Haze is a common degradation in outdoor Internet of Things (IoT) deployments—such as smart cities, environmental monitoring, and intelligent transportation—where front-end camera nodes operate under dynamic weather and lighting conditions. Dense and spatially non-homogeneous haze severely degrades image quality and undermines downstream tasks, while classical model-based dehazing often fails under real-world violations of the assumed degradation model. To address this, we propose a frequency-prior guided dehazing framework tailored for resource-constrained IoT devices. Our approach explicitly decomposes images in the frequency domain and treats phase and amplitude spectra asymmetrically: a lightweight Mamba module with a cyclic strategy is employed to model phase spectra and capture high-frequency structural cues, thereby guiding a Transformer to recover fine details in heavily hazed regions; in parallel, a lightweight channel-attention mechanism is applied to the amplitude spectrum to attenuate haze-related components. To enhance structural fidelity and perceptual quality, we introduce a phase-spectrum contrastive loss alongside multi-stage intermediate supervision. Based on the lightweight design, our model is well-suited for image dehazing task in IoT scenarios. Extensive testing across multiple devices validates the dehazing capability of our network. The excellent performance on multiple non-homogeneous and synthetic datasets further validates the effectiveness of our proposed model.

Efficient compression of massive 3D point clouds remains challenging under limited storage and bandwidth. The Moving Picture Experts Group (MPEG) has developed the geometry-based point cloud compression (G-PCC) standard and is currently at the FDIS stage, referred to as Enhanced G-PCC. This paper proposes a novel inter prediction attribute coding based on the Region Adaptive Hierarchical Transform (RAHT) and entropy coding framework to enhance the coding efficiency. 1)Multi-Reference Frames: A low-delay prediction architecture is introduced to improve temporal correlation utilization, which has not yet been explored in G-PCC inter RAHT attribute compression, with a lightweight motion-based switch to disable multi-reference prediction for large motion. 2)Occupancy Information Difference: The inter eligibility based on occupancy information difference scheme is employed to select valid inter reference blocks, while hybrid prediction scheme based on occupancy information difference is further introduced to enhance prediction accuracy. 3)Correlation Entropy Coding: A new entropy coder is developed to exploit the intrinsic correlation among color components. By jointly leveraging these techniques, the proposed method efficiently exploits spatial and temporal regularities of dynamic point clouds, achieving significant performance gains. Experimental results show that it outperforms the state-of-the-art Enhanced G-PCC reference software (TMC13-v31), with average coding gains of 3.17% for reflectance on the Cat3 dataset and 2.81%, 16.92%, and 12.21% for the Luma, Cb, and Cr components on the Cat2 dataset under MPEG common test conditions. The method,Occupancy Information Difference, has already been adopted into the MPEG Enhanced G-PCC standard.

Gaussian SLAM systems excel in real-time rendering and fine-grained reconstruction compared to NeRF-based systems. However, their reliance on extensive keyframes is impractical for deployment in real-world robotic systems, which typically operate under sparse-view conditions that can result in substantial holes in the map. To address these challenges, we introduce DenseSplat, the first SLAM system that effectively combines the advantages of NeRF and 3DGS. DenseSplat utilizes sparse keyframes and NeRF priors for initializing primitives that densely populate maps and seamlessly fill gaps. It also implements geometry-aware primitive sampling and pruning strategies to manage granularity and enhance rendering efficiency. Moreover, DenseSplat integrates loop closure and bundle adjustment, significantly enhancing frame-to-frame tracking accuracy. Extensive experiments on multiple large-scale datasets demonstrate that DenseSplat achieves superior performance in tracking and mapping compared to current state-of-the-art methods.

Current Simultaneous Localization and Mapping (SLAM) methods based on Neural Radiance Fields (NeRF) or 3D Gaussian Splatting excel in reconstructing static 3D scenes but struggle with tracking and reconstruction in dynamic environments, such as real-world scenes with moving elements. Existing NeRF-based SLAM approaches addressing dynamic challenges typically rely on RGB-D inputs, with few methods accommodating pure RGB input. To overcome these limitations, we propose Dy3DGS-SLAM, the first 3D Gaussian Splatting (3DGS) SLAM method for dynamic scenes using monocular RGB input. To address dynamic interference, we fuse optical flow masks and depth masks through a probabilistic model to obtain a fused dynamic mask. With only a single network iteration, this can constrain tracking scales and refine rendered geometry. Based on the fused dynamic mask, we designed a novel motion loss to constrain the pose estimation network for tracking. In mapping, we use the rendering loss of dynamic pixels, color, and depth to eliminate transient interference and occlusion caused by dynamic objects. Experimental results demonstrate that Dy3DGS-SLAM achieves state-of-the-art tracking and rendering in dynamic environments, outperforming or matching existing RGB-D methods.

Realistic animatable human avatars from monocular videos are crucial for advancing human-robot interaction and enhancing immersive virtual experiences. While recent research on 3DGS-based human avatars has made progress, it still struggles with accurately representing detailed features of non-rigid objects (e.g., clothing deformations) and dynamic regions (e.g., rapidly moving limbs). To address these challenges, we present STG-Avatar, a 3DGS-based framework for high-fidelity animatable human avatar reconstruction. Specifically, our framework introduces a rigid-nonrigid coupled deformation framework that synergistically integrates Spacetime Gaussians (STG) with linear blend skinning (LBS). In this hybrid design, LBS enables real-time skeletal control by driving global pose transformations, while STG complements it through spacetime-adaptive optimization of 3D Gaussians. Furthermore, we employ optical flow to identify high-dynamic regions and guide the adaptive densification of 3D Gaussians in these regions. Experimental results demonstrate that our method consistently outperforms state-of-the-art baselines in both reconstruction quality and operational efficiency, achieving superior quantitative metrics while retaining real-time rendering capabilities. Our code is available at https://github.com/jiangguangan/STG-Avatar

Gaussian Splatting SLAMs have made significant advancements in improving the efficiency and fidelity of real-time reconstructions. However, these systems often encounter incomplete reconstructions in complex indoor environments, characterized by substantial holes due to unobserved geometry caused by obstacles or limited view angles. To address this challenge, we present Manhattan Gaussian SLAM, an RGB-D system that leverages the Manhattan World hypothesis to enhance geometric accuracy and completeness. By seamlessly integrating fused line segments derived from structured scenes, our method ensures robust tracking in textureless indoor areas. Moreover, The extracted lines and planar surface assumption allow strategic interpolation of new Gaussians in regions of missing geometry, enabling efficient scene completion. Extensive experiments conducted on both synthetic and real-world scenes demonstrate that these advancements enable our method to achieve state-of-the-art performance, marking a substantial improvement in the capabilities of Gaussian SLAM systems.

In hyperspectral anomaly detection (HAD), tensor low-rankness is essential for effectively separating background and anomaly. However, most of the current low-rank-based methods do not use the spatial-spectral low-rankness and the nonlocal self-similarity simultaneously. To address this issue, we propose a tensor double nuclear norm-based tensor approximation (TDNN-TA) model with all the priors in a unified convex framework, which can be efficiently handled through a well-organized alternating direction method of multipliers. Especially, to thoroughly model the background by tensor approximation, we propose a tensor double nuclear norm (TDNN), which achieves a more precise and flexible exploration of low-rankness and nonlocal self-similarity by applying different low-rank constraints to the global tensor and the group tensor. Moreover, to explore the intrinsic characteristics of different priors by various tensor ranks, we employ the Fourier transform-based three-directional tensor nuclear norm to approximate the nonlocal group tensor rank, and the framelet-based three-modal tensor nuclear norm to approximate the global tensor rank. Experimental results validated on several real hyperspectral datasets demonstrate that TDNN-TA is effective in detecting different sizes of anomalous targets and achieves competitive results for various scenes.

With the advancement of spectral imaging technology, hyperspectral image (HSI) resources have been rapidly expanding, making cross-dataset HSI classification a critical technique and an inevitable trend for large-scale Earth observation applications. However, most existing approaches transfer from one HSI to another in a supervised way, which limits their ability to leverage multi-source HSI information. Therefore, this paper proposes an unsupervised cross-dataset HSI classification method based on parallel adversarial domain adaptation (PADA), which learns and integrates task-relevant and domain-invariant knowledge from multi-source HSIs to classify a target HSI. Specifically, a source-and-target shared information mining module is designed to mine transferable knowledge from each source HSI. This module employs parallel adversarial learning between a spectral-spatial feature extractor and a task-relevant controller with a domain-invariant discriminator to learn task-relevant and domain-invariant features, thereby mitigating domain shift. Moreover, a source-to-target transferability learning module is proposed to evaluate the cross-dataset transferability of knowledge, which computes domain correlation score to evaluate inter-domain correlation and guide adaptive knowledge transfer, effectively suppressing redundant information and reducing negative transfer caused by low-correlated sources. Finally, a multi-source collaborative classification module is developed to accomplish cross-dataset knowledge transfer, which designs correlation-aware fusion strategy to produce the final classification result by integrating information from each source, ensuring balanced and robust decision-making. Extensive experiments conducted under challenging multi-source cross-dataset settings validate the classification performance and the domain extensibility of the proposed method, demonstrating its potential for large-scale Earth observation applications.

Hyperspectral target detection (HTD) identifies objects of interest from complex backgrounds at the pixel level, playing a vital role in Earth observation. However, the limited target priors constrain the ability to obtain sufficient features or patterns for background-target discrimination, and spectral variation further exacerbates the difficulty of achieving reliable and robust performance. To address these challenges, this article proposes an efficient self-supervised HTD method with a pyramid state space model (SSM), named HTD-Mamba, which employs spectrally contrastive learning to distinguish between target and background based on the similarity measurement of intrinsic features. Specifically, to obtain sufficient training samples and leverage spatial contextual information, we propose a spatial-encoded spectral augmentation (SESA) technique that encodes all surrounding pixels within a patch into a transformed view of the center pixel. In addition, to explore global band correlations, we divide pixels into continuous group-wise spectral embeddings and introduce Mamba to HTD for the first time to model long-range dependencies of the spectral sequence with linear complexity. Furthermore, to alleviate spectral variation and enhance robust representation, we propose a pyramid SSM as a backbone to capture and fuse multiresolution spectral-wise intrinsic features. Extensive experiments conducted on four public datasets demonstrate that the proposed method outperforms state-of-the-art methods in both quantitative and qualitative evaluations. The code is available athttps://github.com/shendb2022/HTD-Mamba.

Handheld ophthalmoscopes often produce lowquality fundus images, which negatively affect the performance of deep learning models in diagnosing fundus diseases. Existing methods for fundus image enhancement mainly focus on visual improvement rather than optimizing the enhancement outcomes for specific diagnostic tasks. We propose a method for processing low-quality fundus images and diagnosing fundus diseases using an enhancement model. To address the lack of paired low/highquality glaucoma images, we artificially degraded images from the BrG public dataset. We then trained the RFormer enhancement model using transfer learning to transfer knowledge from images captured by traditional medical equipment to those captured by portable devices. The proposed method is designed with downstream diagnostic tasks in mind, creating a feedback loop that uses diagnostic model loss to guide realtime updates to the enhancement model. This approach not only boosts diagnostic performance but also enhances the perceived quality of the output images.

With the increasing application of multi-object tracking in practical scenarios, the problem of tracking object loss due to occlusion is receiving more and more attention. However, multi-object tracking techniques still face significant challenges, and the accuracy and robustness of tracking need to be further improved. To address these problems, a multi-object tracking algorithm based on motion estimation and adaptive appearance matching is proposed, using BoT-SORT as a baseline. First, a method called ME-DIOU is proposed to compute the similarity of targets based on motion estimation. This method improves the accuracy of matching tracked targets in successive frames when there is a loss of appearance features. Secondly, an adaptive thresholding method for appearance feature matching is proposed to improve the stability of appearance models in different scenes. Experiments were conducted on public datasets and the results show that the improved algorithm has better HOTA, MOTA and IDF1 values compared to the BoT-SORT algorithm. Compared to similar algorithms, the improved algorithm has more stable performance in dealing with occlusion and other problems, reducing the risk of losing tracking targets after occlusion.

The tracking-by-detection paradigm typically in-volves two primary steps: object detection and data association. In recent years, researchers have begun to introduce appearance features into the association stage to reduce ID switches and improve trajectory reconstruction capability. Although this approach has made significant progress, it still encounters various challenges, such as the susceptibility of appearance features to environmental noise and the inappropriate combination of motion and appearance information. To address above problems, we use the advanced motion-based OC-SORT as the baseline for improvement in this work. Firstly, we integrate appearance features into the association stage and use an adaptive appearance updating strategy to ensure the robustness of these features. Secondly, we adaptively adjust the weighting of motion and appearance features during the association stage, according to the number of consecutive frames in which the trajectory fails to match any detections. This adaptive adjustment aids in recovering objects that were lost due to long-term occlusion. Finally, we implement a camera motion compensation module, which enhances the accuracy of capturing object motion information correcting image displacement in real time, thereby further optimizing tracking performance. We validate the improved algorithm using multiple public datasets, and the experimental results show that our method significantly improves the stability of tracking.

As Whole Slide Images(WSI) are characterized by their extremely high resolution and lack of pixel-level annotation, Multiple Instance Learning(MIL) has become an effective tool for addressing WSI classification challenge. However, current MIL methods typically rely on the independent and identically distributed assumption, neglecting correlation among different instances and only analyze WSI at a single scale, failing to effectively utilize the complementarity between multi-scale features. To enhance classification performance and obtain more accurate diagnostic results, we propose a MIL model based on multi-scale Graph-Transformer framework. It utilizes graph convolutional network to capture spatial, semantic, and scale correlations among patches, and performs graph pooling to select discriminative patches in order to reduce computational costs. Additionally, we introduce a multi-scale feature fusion module aimed at integrating features across different scales. The experimental results on the TCGA-NSCLC and TCGA-RCC datasets show that our method achieves consistent high performance for different cancer types, obtaining accuracy of 0.9281 and 0.9508 on the above two datasets, respectively. Compared with other comparison methods, our method has surperior classification performance.

Time series forecasting (TSF) has extensive applications in domains, such as energy, traffic, and weather prediction. Currently, existing literature has designed many architectures that combine deep learning models in the frequency domain, and effective results have been achieved. However, handling complex-valued arithmetic poses a challenge for most frequency domain-based models. Additionally, features extracted solely in either the time or frequency domain are not comprehensive enough. To solve these problems, we propose a fully complex network (FCNet) in this work, where all network layers are adapted to handle complex-valued computations to simultaneously learn the information in the real and imaginary parts. First, we utilize time-frequency conversion to obtain time-frequency domain signals. And then we design the time-frequency filter-enhanced block to effectively capture global features from time-frequency signals. Finally, we design the complex-valued time-frequency Transformers Block, which separately extracts information from the time and frequency domains. Experimental evaluations on eight data sets from five benchmark domains demonstrate that our model significantly outperforms state-of-the-art methods in TSF. Code is available athttps://github.com/ZHU-0108/FCNet-main.

As one of the most important application for remote sensing monitoring, land cover mapping has witnessed notable advancements in data acquisition, algorithmic diversity, and classification accuracy. Despite the instrumental role data-driven algorithms have played in the development of global land cover products, their inherent limitations as “black box” methods often fall short of meeting end-users’ specific requirements. In this study, built upon the foundation of the earlier land cover monitoring platform [FROM-GLC plus(FGP)], a data and knowledge dual-driven framework (FGP 2.0) was developed as a user-adaptive framework for intelligent remote sensing land cover mapping. By incorporating ontology-based semantic descriptions with advanced data-driven algorithms, FGP 2.0 provides the capacity for both traditional annual mapping and emerging dynamic mapping. Our results illustrate that FGP 2.0 significantly improves the overall accuracy of annual maps by ~5%, and dynamic maps by ~20% compared to FGP. Moreover, an operational dynamic mapping tool has been developed on the Google Earth engine (GEE), enabling the generation of near-real-time land cover maps for any given place. With an extensible and flexible mapping framework, FGP 2.0 demonstrates the potential of customized land cover monitoring results to suit different application scenarios. This innovative approach not only meets the current demand for reliable annual and dynamic land cover maps but also sets a new benchmark for the integration of geoscientific expertise with machine learning techniques in remote sensing monitoring.

Hyperspectral target detection (HTD) is an important issue in earth observation, with applications in both military and civilian domains. However, conventional representation-based detectors are hindered by the reliance on the unknown background dictionary, the limited ability to capture nonlinear representations using the linear mixing model (LMM), and the insufficient background-target recognition based on handcrafted priors. To address these problems, this paper proposes an interpretable representation network that intuitively realizes LMM for HTD, making nonlinear feature expression and physical interpretability compatible. Specifically, a subspace representation network is designed to separate the background and target components, where the background subspace can be adaptively learned. In addition, to further enhance the nonlinear representation and more accurately learn the coefficients, a lightweight multi-scale Transformer is proposed by modeling long-distance feature dependencies between channels. Furthermore, to supplement the depiction for target-background discrimination, a constrained energy minimization (CEM) loss is tailored by minimizing the output background energy and maximizing the target response. The effectiveness of the proposed method is demonstrated on four benchmark datasets, showing its superiority over state-of-the-art methods. The code for this work is available at https://github.com/shendb2022/HTD-IRN for reproducibility purposes.

The problem of target detection in hyperspectral images is an unsupervised binary classification problem with extremely uneven samples. To highlight the target and suppress the background as much as possible, this paper proposes a target detection algorithm based on dual sparse constraints. Specifically, the original image can be decomposed into a background image and a target image. Combined with sparse representation, the target detection problem can be transformed into a problem of optimizing the target and background coefficient matrices. This problem can be solved by the alternating direction method of multipliers. Considering that both the target dictionary and background dictionary are unknown, this paper also proposes a dictionary construction algorithm based on spectral similarity and clustering to obtain relatively complete and pure target and background dictionaries. The experimental results demonstrate that the proposed method outperforms other competing methods in both quantitative performance and visual effect. Code and datasets are available at https://github.com/shendb2022/DSC.

Change detection, which gives a quantitative analysis of the change information for the target area, is an important technology for lots of remote sensing tasks. Supervised change detection methods can achieve satisfactory performance when there are enough labeled samples, which is a harsh requirement for change detection tasks using remote sensing images. To solve this problem, we propose a change detection method based on semi-supervised segmentation and adversarial learning. The proposed method can learn knowledge from both the limited labeled samples and the abundant unlabeled samples to improve the generalization performance of the model. Firstly, the segmentation maps of both labeled samples and unlabeled samples are obtained through a segmentation network. Then, the labeled samples is used to train the discriminator network, which is responsible for distinguishing the segmentation prediction from the ground truth. Finally, the discriminator output is used as a measurement for self-training to minimize the feature difference between the segmentation prediction and the ground truth. Experimental results on the Sun Yat-Sen University (SYSU) dataset show that the proposed method can use unlabeled samples to improve the quality of prediction, which guarantees the proposed method can achieve good performance with few labeled samples.

Hyperspectral image has been widely used in the field of remote sensing due to the rich spectral and spatial information. During land-cover investigation, new types of ground objects appear constantly, which need the classification model to make quick judgments. However, a trained hyperspectral classification model can only make predictions on pre-defined classes, which limits its application. In order to deal with the aforementioned issue, we propose an incremental learning method based on constantly updated classifier, which is able to recognize new classes by few-shot samples and classify old classes without storing any old samples. Specifically, we propose a decoupled structure of feature representation module and classifier module, the feature representation is trained in the initial stage and then frozen, while the classifier module is updated in the next incremental stages, which alleviates knowledge forgetting as well as over-fitting. Besides, we adopt attention mechanism with a kernel of cosine distance to measure the similarity between prototypes and test samples of each class, which is more robust with few-shot samples. Extensive experiments and analyses based on typical hyperspectral images verified the effectiveness of the proposed method.

Ship target detection in synthetic aperture radar (SAR) images is essential for many applications in marine monitoring and port security. Though considerable developments have been achieved, there still exist some issues toward multiscale and dense ship targets in complex inshore scenes. Under such common but challenging situations, it is difficult to extract effective target information, which drives the missing alarm rate rising dramatically. In complex scenes, it is hard to disentangle background noise from ship target information, which causes false alarm frequently. In this article, an anchor-free SAR ship detection method based on key-point estimation and attention mechanism is proposed to address the aforementioned issues. Specially, an anchor-free framework with skip connections and aggregation nodes is designed to fuse multiresolution features and detect multiscale ship targets. Moreover, a key-point estimation module is proposed to eliminate the undetected ship targets caused by dense target distribution. Furthermore, a channel attention module is explored to enhance network attention on ship targets and suppress background noise. Sufficient experimental results on the open SAR ship detection dataset demonstrate that compared with some state-of-the-art methods, the proposed method is able to achieve higher detection accuracy with a lower false alarm rate.

The visual loop-closure detection for autonomous underwater vehicles (AUVs) is a key component to reduce the drift error accumulated in simultaneous localization and mapping tasks. However, due to viewpoint changes, textureless images, and fast-moving objects, the loop closure detection in dramatically changing underwater environments remains a challenging problem to traditional geometric methods. Inspired by strong feature learning ability of deep neural networks, we propose an underwater loop-closure detection method based on a variational autoencoder network in this article. Our proposed method can learn effective image representations to deal with the challenges caused by dynamic underwater environments. Specifically, the proposed network is an unsupervised method, which avoids the difficulty and cost of labeling a great quantity of underwater data. Also included is a semantic object segmentation module, which is utilized to segment the underwater environments and assign weights to objects in order to alleviate the impact of fast-moving objects. Furthermore, an underwater image description scheme is used to enable efficient access to geometric and object-level semantic information, which helps to build a robust and real-time system in dramatically changing underwater scenarios. Finally, we test the proposed system under complex underwater environments and get a recall rate of 92.31% in the tested environments.

Hyperspectral image classification, which is a crucial task for various remote sensing applications, can achieve qualified performance under a conventionalized assumption, i.e., there are sufficient samples in every concerned class. However, in field investigation, collecting enough samples for every class is extremely difficult, which results in defective training sets with insufficient or imbalanced samples and deteriorates the classification performance dramatically. In order to address this issue, we propose a hyperspectral image classification method based on a task-specific learning network. The proposed network works under an episode-based framework, which learns general knowledge from the tasks with sufficient samples by metalearning and relation learning and then inference specific knowledge for the tasks with few samples by parameters adjustment and representation comparison. In particular, a task-specific feature learner is designed to learn unbiased features, and a comparison-based classifier is utilized to adapt the minority classes. As a result, the proposed method can obtain a qualified overall accuracy over all samples with a comparative averaged accuracy over all classes. Extensive experiments on three real hyperspectral images show that our method can achieve state-of-the-art performance under both the few-shot and imbalanced training set settings.

The cross-data set knowledge is vital for hyperspectral image classification, which can reduce the dependence on the sample quantity by transferring knowledge from other data sets and improve the training efficiency by sharing knowledge between different data sets. However, due to the capturing environment change and imaging equipment difference, domain shift troubles the exploitation of the cross-data set knowledge. To address the aforementioned issue, this article proposes an unsupervised cross-data set hyperspectral image classification method based on adversarial domain adaptation. The proposed method, which employs multiple classifiers to build a discriminator and uses variational autoencoders to constitute a generator, works in an adversarial manner to drive the target samples under the support of the source domain. In particular, the classification error and the classification disagreement are considered in the objective function, which helps to align different domains while keeping the boundaries of different classes. Experimental results of the multidomain data set demonstrate that the proposed method can transfer and share cross-data set knowledge and achieve state-of-the-art performance without using the labeled information of the target data set.

Ship detection from SAR images is an important remote sensing application. However, in complex scenes, i.e., the shore or harbor area, traditional ship detection methods cannot disentangle background information from the target ship, and the detection performance drops dramatically. Moreover, the severe coherent speckle noise also challenges ship detection from SAR images. In order to address the aforementioned issues, this paper proposes a SAR image ship detection method based on scene interpretation to improve the performance of ship detection in complex scenes. Firstly, segmentation algorithm based on Mask R-CNN is utilized to interpret the scene into two catalogs, i.e., the sea and the land. Then, ship detection algorithm based on Faster R-CNN is performed on the sea area and the land area respectively. Finally, non-maximum suppression is used to integrate detection results. Experimental results on SAR ship detection dataset illustrate that the proposed method produces high detection accuracy and low false alarm rate in complex scenes.

Recently, Fully Convolutional Network (FCN) seems to be the go-to architecture for image segmentation, including semantic scene parsing. However, it is difficult for a generic FCN to predict semantic labels around the object boundaries, thus FCN-based methods usually produce parsing results with inaccurate boundaries. Meanwhile, many works have demonstrate that level set based active contours are superior to the boundary estimation in sub-pixel accuracy. However, they are quite sensitive to initial settings. To address these limitations, in this paper we propose a novel Deep Multiphase Level Set (DMLS) method for semantic scene parsing, which efficiently incorporates multiphase level sets into deep neural networks. The proposed method consists of three modules, i.e., recurrent FCNs, adaptive multiphase level set, and deeply supervised learning. More specifically, recurrent FCNs learn multi-level representations of input images with different contexts. Adaptive multiphase level set drives the discriminative contour for each semantic class, which makes use of the advantages of both global and local information. In each time-step of the recurrent FCNs, deeply supervised learning is incorporated for model training. Extensive experiments on three public benchmarks have shown that our proposed method achieves new state-of-the-art performances. The source codes will be released at https://github.com/Pchank/DMLS-for-SSP.

Street Scene Parsing (SSP) is a fundamental and important step for autonomous driving and traffic scene understanding. Recently, Fully Convolutional Network (FCN) based methods have delivered expressive performances with the help of large-scale dense-labeling datasets. However, in urban traffic environments, not all the labels contribute equally for making the control decision. Certain labels such as pedestrian, car, bicyclist, road lane or sidewalk would be more important in comparison with labels for vegetation, sky or building. Based on this fact, in this paper we propose a novel deep learning framework, named Residual Atrous Pyramid Network (RAPNet), for importance-aware SSP. More specifically, to incorporate the importance of various object classes, we propose an Importance-Aware Feature Selection (IAFS) mechanism which automatically selects the important features for label predictions. The IAFS can operate in each convolutional block, and the semantic features with different importance are captured in different channels so that they are automatically assigned with corresponding weights. To enhance the labeling coherence, we also propose a Residual Atrous Spatial Pyramid (RASP) module to sequentially aggregate global-to-local context information in a residual refinement manner. Extensive experiments on two public benchmarks have shown that our approach achieves new state-of-the-art performances, and can consistently obtain more accurate results on the semantic classes with high importance levels.

Cross-scene classification of hyperspectral image (HSI) has been increasingly researched due to its crucial utilization in practical applications. However, cross-scene data generally perform distribution discrepancy, which hampers the transfer learning performance. To address this issue, deep conditional distribution adaptation networks (DCDAN) are proposed for HSI cross-scene classification, which aim to reduce the distribution shift between a source domain and a target domain. The proposed deep network adopts a conditional constraint to match the class conditional distributions across domains, where a great number of training samples from the source domain and a small number of training samples from the target domain are utilized to train the deep model. Cross-scene classification results on two HSIs demonstrate that the proposed network is able to yield superior performance compared with some related methods.

Hyperspectral image, which contains high-resolution spectral information as well as large-scale spatial information, has been widely used in various classification applications of remote sensing area. However, due to the insufficient of the labeled samples in the training set and the unbalance of sample quantity between different classes, traditional supervised classification methods are difficult to achieve satisfying performance. In order to address above issues, this paper studies on how to predict classification parameters more effectively, and finds out that the parameters of the fully-connected layer in the classifier are closely related to the output of the feature mapping layer. Based on above fact, this paper proposes a hyperspectral image classification method base on parameter prediction network, which adapts a pre-trained neural network to novel categories by directly predicting the parameters of classifier from the feature data of the hyperspectral image. Experimental results and analysis demonstrate the competitive performance of the proposed method over other state-of-the-art classification methods based on neural network when the number of labeled samples is very small.

Due to the rapid development of learning-based methods, hyperspectral image classification has achieved remarkable progress. However, since the semantic discrepancy between different land-cover types, the feature distributions of different classes are so nonuniform that the classifier can not measure them with a single rule. In order to consider semantic knowlage in classification, this paper propose a knowledge guided classification method based on hierarchical class tree and deep learning. The proposed method fuses similar classes into several super classes by the knowledge of the confusion matrix, and classify multi-level super classes with different deep networks. Extensive experiments on two hyperspectral images demonstrate that the proposed method can utilize semantic information of different land-cover types, and give better performance than traditional methods.

In existing deep network-based image super-resolution (SR) methods, each network is only trained for a fixed upscaling factor and can hardly generalize to unseen factors at test time, which is non-scalable in real applications. To mitigate this issue, this paper proposes a resolution-aware network (RAN) for simultaneous SR of multiple factors. The key insight is that SR of multiple factors is essentially different but also shares common operations. To attain stronger generalization across factors, we design an upsampling network (U-Net) consisting of several sub-modules, in which each sub-module implements an intermediate step of the overall image SR and can be shared by SR of different factors. A decision network (D-Net) is further adopted to identify the quality of the input low-resolution image and adaptively select suitable sub-modules to perform SR. U-Net and D-Net together constitute the proposed RAN model, and are jointly trained using a new hierarchical loss function on SR tasks of multiple factors. Experimental evaluations demonstrate that the proposed RAN compares favorably against the state-of-the-art methods and its performance can well generalize across different upscaling factors.

Change detection is an important task to identify land-cover changes between the acquisitions at different times. For synthetic aperture radar (SAR) images, inherent speckle noise of the images can lead to false changed points, which affects the change detection performance. Besides, the supervised classifier in change detection framework requires numerous training samples, which are generally obtained by manual labeling. In this paper, a novel unsupervised method named saliency-guided deep neural networks (SGDNNs) is proposed for SAR image change detection. In the proposed method, to weaken the influence of speckle noise, a salient region that probably belongs to the changed object is extracted from the difference image. To obtain pseudotraining samples automatically, hierarchical fuzzy C-means (HFCM) clustering is developed to select samples with higher probabilities to be changed and unchanged. Moreover, to enhance the discrimination of sample features, DNNs based on the nonnegative- and Fisher-constrained autoencoder are applied for final detection. Experimental results on five real SAR data sets demonstrate the effectiveness of the proposed approach.

Deep learning-based classification methods are competent to achieve an excellent performance under one necessary condition, i.e., there are sufficient labeled samples in each class, which is extremely impractical in most of the remote sensing tasks. To improve the performance with small training sets, we resort to other hyperspectral images and design a two-phase relation learning network that can be transferred between different images for general information sharing and fine-trained on a specific hyperspectral image for individual information learning. Specifically, we use a relation learning method to compare samples and deal with the task inconsistency between different data sets, and we adopt an episode-based training strategy to mimic the testing setup and learn the transferable comparison ability. Benefited from these two strategies, the proposed network takes the advantage of extra knowledge for information supplement and learns to compare rather than to classify for information exploration, which guarantees a reasonable performance even with small training sets. Extensive experiments and analysis on three benchmarks demonstrate that the proposed method can provide an effective solution for hyperspectral image classification with small training sets, which makes it possible to work on large-scale applications of earth observation with less effort on field investigation.

For hyperspectral image classification, there is a large gap between the theoretical method and the practical application. Hyperspectral image classification in theoretical research trains a new classifier for each data set, which is ineffective and even infeasible in large-scale applications. In this paper, we make a preliminary attempt to recycle the classification model to new data sets in an unsupervised way. Specially, we propose a cross-data set hyperspectral image classification method based on deep domain adaptation. The proposed method contains three modules: domain alignment module that learns to minimize the domain discrepancy with the guide of an irrelevant task, task allocation module that learns to classify on the source domain with the regulation of domain alignment, and domain adaptation module that transfers both the alignment ability and classification ability to the target domain by an adaptation strategy. As a result, with the information of an irrelevant task on dual-domain data sets, we can minimize the domain discrepancy and transfer the task-relevant knowledge from the source domain to the target domain in an unsupervised way. Extensive experiments on three hyperspectral images demonstrate the effectiveness of our method compared with other related methods when dealing with new data sets.

Device-free activity recognition (DFAR) is a promising technique that detects the activity of a target by analyzing the influence of its existence on surrounding wireless links. It realizes target sensing without the participation or even awareness of the target. The key question of DFAR is how to characterize the influence of the target on wireless links. Existing works mostly utilize statistical features, such as mean and variance in time-domain, and energy as well as entropy in frequency-domain, to characterize the influenced signals. However, statistical features provide only partial information. This paper explores the method on how to characterize the distribution of the signal as a whole. Specifically, we present a novel coherence histogram, which leverages the spatial structural characteristics to better characterize the distribution of the wireless signal. The coherence histogram captures not only the occurrence probability of received signal strength (RSS) measurements, but also the spatial relationship between adjacent RSS measurements as well. Experimental results show that our coherence histogram-based DFAR system could achieve an accuracy of more than 96%, which significantly outperforms other state-of-the-art DFAR systems remarkably.

In this paper, we propose a novel feature learning framework for video person re-identification (re-ID). The proposed framework largely aims to exploit the adequate temporal information of video sequences and tackle the poor spatial alignment of moving pedestrians. More specifically, for exploiting the temporal information, we design a temporal residual learning (TRL) module to simultaneously extract the generic and specific features of consecutive frames. The TRL module is equipped with two bi-directional LSTM (BiLSTM), which are respectively responsible to describe a moving person in different aspects, providing complementary information for better feature representations. To deal with the poor spatial alignment in video re- ID datasets, we propose a spatial-temporal transformer network (ST2N) module. Transformation parameters in the ST2N module are learned by leveraging the high-level semantic information of the current frame as well as the temporal context knowledge from other frames. The proposed ST2N module with less learnable parameters allows effective person alignments under significant appearance changes. Extensive experimental results on the largescale MARS, PRID2011, ILIDS-VID and SDU-VID datasets demonstrate that the proposed method achieves consistently superior performance and outperforms most of the very recent state-of-the-art methods.

Convolutional neural networks (CNN), which are able to extract spatial semantic features, have achieved outstanding performance in many computer vision tasks. In this paper, hybrid neural networks (HNN) are proposed to extract both spatial and spectral features in the same deep networks. The proposed networks consist of different types of hidden layers, including spatial structure layer, spatial contextual layer, and spectral layer. All those layers work as organic networks to explore as much valuable information as possible from hyperspectral data for classification. Experimental results demonstrate competitive performance of the proposed approach over other state-of-the-art neural networks methods. Moreover, the proposed method is a new way to deal with multidimensional data with deep networks.

The classification of polarimetric synthetic aperture radar (PolSAR) image is of crucial significance for SAR applications. In this letter, a superpixel restrained deep neural network with multiple decisions (SRDNN-MDs) is proposed for PolSAR image classification, which not only extracts effective superpixel spatial features and degrades the influence of speckle noises but also deals with the limited training samples. First, the polarimetric features of coherency matrix and Yamaguchi decomposition are extracted as initial features, and superpixel segmentation is conducted on the Pauli color-coded image to acquire the superpixel averaged features. Then, an SRDNN based on sparse autoencoders is proposed to capture superpixel correlative features and reduce speckle noises. After that, MDs, including nonlocal decision and local decision, are developed to select credible testing samples. Finally, our deep network is updated by the extended training set to yield the final classification map. Experimental results demonstrate that the proposed SRDNN-MD yields higher accuracies compared with other related approaches, which indicate that the proposed method is able to capture superpixel correlative information and adds the information of unlabeled samples to improve the classification performance.

A large receptive field of deep networks can better incorporate image context and benefits image super-resolution (SR) in many ways. However, common techniques, like strided pooling and convolutional operations, are not directly applicable to SR due to severe image detail losses. In this letter, we circumvent this issue by proposing a new network architecture, namely the information-compensated (IC) downsampling block. It first uses pooling layers to downsample input feature maps and then immediately upsamples the feature maps back to the original size. To further compensate for information loss, skip connections are added to propagate lost features caused by downsampling to the upsampled output. In addition, pixelwise recurrent units are also applied to the downsampled feature maps to model context coherence. Compared with traditional pooling layers, the IC downsampling blocks cannot only enlarge receptive field and better capture image context, but also preserve image details, which are essential to SR. The final network consists of a stack of IC downsampling blocks and can be trained in an end-to-end manner. Experimental results verify that the proposed method performs favorably against the state-of-the-art approaches.

Synthetic aperture radar (SAR) image classification is a fundamental process for SAR image understanding and interpretation. With the advancement of imaging techniques, it permits to produce higher resolution SAR data and extend data amount. Therefore, intelligent algorithms for high-resolution SAR image classification are demanded. Inspired by deep learning technology, an end-to-end classification model from the original SAR image to final classification map is developed to automatically extract features and conduct classification, which is named deep recurrent encoding neural networks (DRENNs). In our proposed framework, a spatial feature learning network based on long-short-term memory (LSTM) is developed to extract contextual dependencies of SAR images, where 2-D image patches are transformed into 1-D sequences and imported into LSTM to learn the latent spatial correlations. After LSTM, nonnegative and Fisher constrained autoencoders (NFCAEs) are proposed to improve the discrimination of features and conduct final classification, where nonnegative constraint and Fisher constraint are developed in each autoencoder to restrict the training of the network. The whole DRENN not only combines the spatial feature learning power of LSTM but also utilizes the discriminative representation ability of our NFCAE to improve the classification performance. The experimental results tested on three SAR images demonstrate that the proposed DRENN is able to learn effective feature representations from SAR images and produce competitive classification accuracies to other related approaches.

Convolution neural network (CNN) utilizes alternating convolutional and pooling layers to learn representative spatial information when the training samples are sufficient. However, for pixelwise classification of hyperspectral image, some important information is neglected by CNN, such as the erased information by the pooling operation and the appearance information from lower layers. Moreover, the lack of training samples is a common situation in remote sensing area, which afflicts CNN with overfitting problem. To address the aforementioned issues, this paper designs an end-to-end deconvolution network with skip architecture to learn the spectral-spatial features. The proposed network starts with two branches, i.e., the spatial branch and spectral branch. In the spatial branch, a band selection layer is designed to reduce parameters and remit the overfitting problem, unpooling and deconvolution operations are utilized to recover the erased information of the pooling layers and learn pixelwise spatial representation hierarchically, and the skip architecture is constructed for merging the deep semantic information with the shallow appearance information. In the spectral branch, a contextual deep network is employed for learning deep spectral features. Experimental results on three benchmark data sets reveal the competitive performance of the proposed approach over several related methods.

Recent advances in device-free wireless sensing (DFS) have shown that it may eventually evolve traditional wireless networks into smart networks which could sense surrounding target location and activity information without equipping the target with any devices. Despite its promising application prospects, one challenging problem to be solved is that the performance of the DFS system degrades significantly in complex scenarios, such as through-wall and non-line-of-sight (NLOS) scenarios. To alleviate this problem, this paper seeks to explore and exploit more informative features from not only the time domain and frequency domain, but also the spatial structural domain. We partition the time domain and frequency domain measurement matrices into basic structure blocks, adopt self-organizing map networks to cluster the blocks into a number of categories, so as to make it feasible to characterize the block distributions. We further adopt coherence histograms to characterize the distribution of the blocks by considering the spatial relationship between adjacent blocks. Thanks to the additional information provided by the spatial structural domain, extensive experimental results achieved in through-wall and NLOS scenarios confirm the outstanding performance of the proposed multi-domain features based DFS system.

Fully convolutional neural networks (FCNs) have shown outstanding performance in many dense labeling problems. One key pillar of these successes is mining relevant information from features in convolutional layers. However, how to better aggregate multi-level convolutional feature maps for salient object detection is underexplored. In this work, we present Amulet, a generic aggregating multi-level convolutional feature framework for salient object detection. Our framework first integrates multi-level feature maps into multiple resolutions, which simultaneously incorporate coarse semantics and fine details. Then it adaptively learns to combine these feature maps at each resolution and predict saliency maps with the combined features. Finally, the predicted results are efficiently fused to generate the final saliency map. In addition, to achieve accurate boundary inference and semantic enhancement, edge-aware feature maps in low-level layers and the predicted results of low resolution features are recursively embedded into the learning framework. By aggregating multi-level convolutional features in this efficient and flexible manner, the proposed saliency model provides accurate salient object labeling. Comprehensive experiments demonstrate that our method performs favorably against state-of-the-art approaches in terms of near all compared evaluation metrics.

Deep convolutional neural networks (CNNs) have delivered superior performance in many computer vision tasks. In this paper, we propose a novel deep fully convolutional network model for accurate salient object detection. The key contribution of this work is to learn deep uncertain convolutional features (UCF), which encourage the robustness and accuracy of saliency detection. We achieve this via introducing a reformulated dropout (R-dropout) after specific convolutional layers to construct an uncertain ensemble of internal feature units. In addition, we propose an effective hybrid upsampling method to reduce the checkerboard artifacts of deconvolution operators in our decoder network. The proposed methods can also be applied to other deep convolutional networks. Compared with existing saliency detection methods, the proposed UCF model is able to incorporate uncertainties for more accurate object boundary inference. Extensive experiments demonstrate that our proposed saliency model performs favorably against state-of-the-art approaches. The uncertain feature learning mechanism as well as the upsampling method can significantly improve performance on other pixel-wise vision tasks.

Marine reclamation plays an increasingly important role in expanding living space, which should be monitored to ensure legitimate development. In this paper, a patch-based recurrent neural network is developed for change detection of marine reclamation. To capture spatial difference of image patches in two images, a patch-based recurrent neural network is proposed to extract features, where patches from two multispectral images are stacked as a sequence for inputting. After training the deep network, Softmax classifier is applied to detect the changed region. It is illustrated that our network can obtain the difference of two images to improve detection accuracies. Experiments on the study area of the Jinzhou Bay demonstrate that the proposed method outperforms other approaches.

Classification of multisensor data provides potential advantages over a single sensor in accuracy. In this paper, deep bimodal autoencoders are proposed for classification of fusing synthetic aperture radar (SAR) and multispectral images. The proposed deep network based on autoencoders is trained to discover both independencies of each modality and correlations across the modalities. Specifically, the sparse encoding layers in the front are applied to learn features of each modality, then shared representation layers in the middle are developed to learn fused features of two modalities, finally softmax classifier in the top is adopted for classification. Experimental results demonstrate that the proposed network is able to yield superior classification performance compared with some related networks.

The mobile communication system has become increasingly complicated with the dramatic growth of user's requirement in quality of service (QoS). The high fluctuation of traffic data makes conventional cell association schemes difficult to guarantee satisfactory service in accord with traffic demand. In this study, the authors propose a novel QoS‐aware cell association scheme in a heterogeneous network. Utilising traffic prediction achieved by support vector regression, the user can decide the best cell according to the future traffic demand. The authors aim at maximising the total throughput with consideration of channel gains and blocking probability in different cells and formulate the cell association as a binary combinatorial optimisation problem. Since users are selfish for their own benefits without the global information, the authors turn this problem into a game theoretical framework. To obtain the Nash equilibrium with low computation complexity, a heuristic dynamic selection algorithm is proposed by updating selection probability which enables each user to associate with the best cell independently. Numerical simulation results show that the proposed algorithm achieves a remarkable throughput gain. The number of satisfied users increases substantially under the different density of users compared with conventional cell association schemes.

In recent years, the vehicular ad‐hoc networks (VANETs) have attracted plenty of attention. However, VANETs can hardly meet surging demands of mobile data. Device‐to‐device (D2D) communication is regarded as a promising method for providing the reliable connectivity between vehicles. The heterogeneous networks including conventional cellular network and VANETs, where the cellular links and the D2D‐based vehicle‐to‐vehicle communication links coexist by reusing the same spectrum resources, result in a more complicated interference scenario. Thus the access mode switch and resource allocation between cellular and VANETs becomes a challenging issue. In this study, the authors formulate an optimal access mode selection and resource allocation scheme as a user aggregate utility maximisation problem with the joint consideration of various network topologies, transmission delay and power reduction. To reduce the high computation complexity, they provide a distributed algorithm that converges to a near‐optimal solution via augmented Lagrangian technique. Numerical results show that remarkable throughput gains are achieved relative to the existing work, especially for large‐scale networks. The network throughput can be considerably improved as well due to power reduction.

Detection of a marine floating raft is significant for ocean utilization, which provides a basis for marine ecosystem protection. In this case study, supervised classifiers of weighted fusion-based representation are proposed to detect marine floating raft using synthetic aperture radar images. To remove the speckle noise and obtain more discriminative features, a weighted low-rank matrix factorization (WLRMF) model is developed to optimize features before detection, where the matrix of patch features is decomposed to acquire the denoised features. Weighted fusion-based representation classifiers (WFRCs) with weighted multiplication are proposed to combine the sparse representation classifier (SRC) and the collaborative representation classifier (CRC) for floating raft detection, which can capture the competition between the floating raft and water surface as well as the collaboration within-class samples. Experiments on the study area of the Bohai Sea confirm that the proposed approach produces better results than some related methods. It is demonstrated that the WLRMF model extracts effective features and overcomes the influence of speckle noise at the same time, and the WFRC model is able to take advantages of the SRC in competition and CRC in collaboration for improving detection accuracies.

The classification of a synthetic aperture radar (SAR) image is a significant yet challenging task, due to the presence of speckle noises and the absence of effective feature representation. Inspired by deep learning technology, a novel deep supervised and contractive neural network (DSCNN) for SAR image classification is proposed to overcome these problems. In order to extract spatial features, a multiscale patch-based feature extraction model that consists of gray level-gradient co-occurrence matrix, Gabor, and histogram of oriented gradient descriptors is developed to obtain primitive features from the SAR image. Then, to get discriminative representation of initial features, the DSCNN network that comprises four layers of supervised and contractive autoencoders is proposed to optimize features for classification. The supervised penalty of the DSCNN can capture the relevant information between features and labels, and the contractive restriction aims to enhance the locally invariant and robustness of the encoding representation. Consequently, the DSCNN is able to produce effective representation of sample features and provide superb predictions of the class labels. Moreover, to restrain the influence of speckle noises, a graph-cut-based spatial regularization is adopted after classification to suppress misclassified pixels and smooth the results. Experiments on three SAR data sets demonstrate that the proposed method is able to yield superior classification performance compared with some related approaches.

Speckle noises are inherent issues in synthetic aperture radar (SAR) images, which hampers the analysis and interpretation of SAR images. In this paper, we propose an iterative low-rank representation algorithm for SAR image despeckling. The original SAR image is first transformed to the logarithmic image, which is then filtered iteratively by the proposed low-rank representation model. Specifically, in each iteration, similar patches measured by the Mahalanobis distance are collected into a group, and then filtered by the nuclear regularized low-rank representation. Finally, all of the filtered patches are aggregated to form the denoised image. Experimental results demonstrate that the proposed algorithm is able to yield state-of-the-art SAR image despeckling performance.

Polarimetric synthetic aperture radar (PolSAR) images are widely applied in terrain and ground cover classification. Feature extraction and classifier design are both important in Pol- SAR image classification. In this paper, various target decompositions are applied to obtain different polarimetric features. Since that neighboring pixels usually belong to the same species, they can be simultaneously represented through linear combinations of training samples. Therefore, a collaborative representation-based classifier with spatially joint regularization is adopted for classification. Experimental results demonstrate that the joint collaborative representation model performs better than other state-of-the-art methods, such as support vector machine and simultaneous sparse representation.

This paper presents a hyperspectral image classification method based on deep network, which has shown great potential in various machine learning tasks. Since the quantity of training samples is the primary restriction of the performance of classification methods, we impose a new prior on the deep network to deal with the instability of parameter estimation under this circumstances. On the one hand, the proposed method adjusts parameters of the whole network to minimize the classification error as all supervised deep learning algorithm, on the other hand, unlike others, it also minimize the discrepancy within each class and maximize the difference between different classes. The experimental results showed that the proposed method is able to achieve great performance under small training set.

Appearance-based human re-identification is challenging due to different camera characteristics, varying lighting conditions, pose variations across camera views, etc. Recent studies have revealed that color information plays a critical role on performance. However, two problems remain unclear: (1) how do different color descriptors perform under the same scene in re-identification problem? and (2) how can we combine these descriptors without losing their invariance property and distinctiveness power? In this paper, we propose a novel ensemble model that combines different color descriptors in the decision level through metric learning. Experiments show that the proposed system significantly outperforms state-of-the-art algorithms on two challenging datasets (VIPeR and PRID 450S). We have improved the Rank 1 recognition rate on VIPeR dataset by 8.7%.

The robustness of inter‐session network coding is still an open issue in lossy wireless networks. The traditional XOR based network coding cannot work well if the overhearing is unperfect. Especially, the coding node cannot know the overheard information in time. In this paper, we consider a robust network coding method, namely R 2 NC which uses random linear network coding to encode packets together in the inter‐session level, to resist the unperfect overhearing problem. With this method, coding node can always know the solvability of coded packets without the knowledge of overheard information. We analyse the performance of R 2 NC method with both lossy links of output and overhearing in the classic X ‐topology model, and give a necessary condition for the existence of coding gain. Finally, we design an optimal coding algorithm and a relay selection algorithm for R 2 NC to achieve its maximal transmission efficiency. Through ns‐2 simulations, we demonstrate that R 2 NC plays a good performance in terms of throughput, delay and overhead, and is robust against losses on output and overhearing links.

Synthetic aperture radar (SAR) image classification is a hot topic in the interpretation of SAR images. However, the absence of effective feature representation and the presence of speckle noise in SAR images make classification difficult to handle. In order to overcome these problems, a deep convolutional autoencoder (DCAE) is proposed to extract features and conduct classification automatically. The deep network is composed of eight layers: a convolutional layer to extract texture features, a scale transformation layer to aggregate neighbor information, four layers based on sparse autoencoders to optimize features and classify, and last two layers for postprocessing. Compared with hand-crafted features, the DCAE network provides an automatic method to learn discriminative features from the image. A series of filters is designed as convolutional units to comprise the gray-level cooccurrence matrix and Gabor features together. Scale transformation is conducted to reduce the influence of the noise, which integrates the correlated neighbor pixels. Sparse autoencoders seek better representation of features to match the classifier, since training labels are added to fine-tune the parameters of the networks. Morphological smoothing removes the isolated points of the classification map. The whole network is designed ingeniously, and each part has a contribution to the classification accuracy. The experiments of TerraSAR-X image demonstrate that the DCAE network can extract efficient features and perform better classification result compared with some related algorithms.

Outage probability is one of the most important performance measures for cooperative communication systems. And the closed-form solution on exact outage probability is essential for many further studies, such as optimal power allocation and rate control. However, as far as we know, the issue remains open for multiuser linear network coded cooperation (LNCC), which is a technology integrating linear network coding into cooperative communication. Consequently, in this paper, we investigate the multiuser LNCC system with multiple cooperation time slots and many-to-one communication pattern. All the possible outage scenarios are fully considered and the closed-form solution of the system's exact outage probability is theoretically derived. In order to obtain the diversity order of the LNCC system, the asymptotic outage probability is also analysed. The theoretical analyses are verified by plenty of Monte Carlo simulations. In order to demonstrate the benefits introduced by LNC to the cooperative system, the outage performance comparison between the LNCC and the traditional Decode-and-Forward (DF) cooperation system is carried out. The results show that given the same number of users and cooperation time slots, the LNCC system's outage probability is greatly lower than that of the traditional DF cooperation system.

The receiver with message propagation (MP) decoder algorithm realises the decoding of traditional Luby transform (LT) encoded packages when there is only one unknown packet adjacent to the encoded package. Error packets can be recovered by decoding with jointing product code composed of shifted Luby transform (SLT)‐encoded package and error packets. In this study, error correction scheme based on single feedback of SLT‐encoded package was adopted for the recovery of received partial error packets over the erasure channel. The scheme of an MP decoder with an error correction code was proposed, and expand shifted robust soliton distribution (ESRSD) appropriate for the scheme feature was designed. Experimental results indicate that the scheme effectively improves the efficiency of the decoder. Compared with the traditional shifted robust soliton distribution, the ESRSD has better decoding performance for the LT code.

In this study, the authors investigate how to find correspondences (pixel‐level matches) between two video sequences with overlapping views, recorded by different stationary uncalibrated video cameras, without imposing any constrains on, or requiring any a priori knowledge of, the viewing or the illumination conditions under which these videos were obtained. They propose a matching method using a low‐level feature, spatiotemporal activity (STA), which effectively combines both the photometric features and the statistic features and makes the proposed method robust to the pose, illumination and geometric effects. They formulate the author's matching problem as a conditional random field model and optimise it by using graph cuts, instead of applying exhaustive pixel‐wise matching. A qualitative analysis of the performance improvement compared with the temporal activity‐based matching method (TAM) is given by inference from a mathematical model and the quantitative results compared with the TAM and the scale‐invariant feature transform (SIFT) are presented with real life examples. The experiments show that their method significantly outperforms the state‐of‐the‐art methods with a higher accuracy and is orders of magnitudes faster than the comparable ones.

For realizing robust target tracking with wireless sensor networks in the circumstance where the propagation parameters of the characteristic signal emitted by the target are unknown, a novel tracking algorithm under the particle filter framework is proposed. We propose a scheme to realize particle weight calculation without the prior knowledge about the propagation parameters of the target's characteristic signal. With the use of the monotonic relationship of the distance and the received signal strength, we define the signal characteristic sequence and particle distance sequence and utilize the modified sequence distance between the signal characteristic sequence and the particle distance sequence as the criterion to calculate the particle weight blindly with simple lightweight operations. Simulation results demonstrate the effectiveness of the proposed algorithm. Copyright © 2011 John Wiley & Sons, Ltd.

Convergecast is probably the most common communication style in wireless sensor networks (WSNs). And network coding (NC) is a promising concept to improve throughput or reliability of convergecast. Most of the existing works have mainly focused on exploiting these benefits without considering its potential adverse effect. In this paper, we argue that network coding may not always benefit convergecast. This viewpoint is discussed within four scenarios: The network-coding-aided (NC-aided) and the none-network-coding (none-NC) convergecast schemes with or without automatic repeat request (ARQ) mechanisms. The most concerned performance metrics, including packet collection rate, energy consumption and end-to-end delay, are investigated. Theoretical analyses and simulation results show that the way network coding operates, i.e., conscious overhearing and the prerequisite of successfully decoding, could naturally diminish its advantages in convergecast. And NC-aided convergecast schemes may even be inferior to none-NC ones when the wireless link delivery ratio is high enough. The conclusion drawn in this paper casts a new light on how to effectively apply network coding to practical WSNs.

Due to its ability of realizing device free localization with wireless networks, the radio tomography becomes a promising technique that draws considerable attention. Traditional radio tomography makes use of the received signal strength (RSS) of wireless links to realize location estimation. However, the RSS measurement is particularly sensitive to noise. Inspired by the fact that similar to the RSS, the time-of-flight (TOF) measurement also changes significantly when some objects shadow the wireless link, and the fact that compared with the RSS, the TOF measurement is robust to noise, a novel TOF-based radio tomography is proposed in this paper. With the TOF measurements of the shadowed links as observation information, a modified particle filter algorithm which utilizes the compressive sensing technique to produce the importance distribution of the particle set is proposed, so as to realize localization and tracking with under-sampled measurements by making full use of the space-domain sparse and time-domain gradually changed feature of the location information. The experiments with the 802.15.4a chirp spread spectrum ranging hardware are presented to confirm the proposed scheme.

Network coding is one of the most promising techniques to increase the reliability and reduce the energy consumption for wireless sensor networks (WSNs). However, most of the previous works mainly focus on the network coding for multi-cast or uni- cast in WSNs, in spite of the fact that the converge-cast is the most common communication style in WSNs. In this paper, we investigate, for the first time as far as we know, the feasibility of acquiring network coding benefits in converge-cast, and we present that with the ubiquitous convergent structures self-organized during converge-casting in the network, the reliability benefits can be obtained by applying linear network coding. We theoretically derive the network coding benefits obtained in a general convergent structure, and simulations are conducted to validate our theoretical analysis. The results reveal that the network coding can improve the network reliability considerably.

In this paper, an energy-efficient optimal relay selection strategy which is jointly optimized with the energy-efficient optimal power allocation solution for AF cooperative transmission is proposed. The relay selection criterion is the distance to the optimal relay locations where the minimum transmission power of the source, the relay or their total can be achieved. To determine those most energy-efficient relay locations, a universal algorithm with low computational complexity and easy implementation is also presented in this paper. The simulations are conducted to validate our theoretical analysis. The results show that with the relay selected by the proposed strategy, the cooperative transmission can achieve considerably high energy-efficiency.

A compressive sensing-based approach to solve the problem of tracking targets in the deployment area of the wireless networks without the need of equipping the target with a wireless device has been proposed. We present a dynamic statistical model for relating the change of the received signal strength between the node pairs to the spatial location of the target. On the basis of the model, the problem is formulated as a sparse signal reconstruction problem, and we propose a novel Bayesian greedy matching pursuit (BGMP) algorithm to tackle the signal reconstruction problem even from a small set of measurements. The BGMP iteratively seeks the contribution of each pixel for multi-times to compensate for the inaccuracy of the measurement matrix, and builds the enumeration region based on the past estimations to speed up the algorithm and improve its reconstruction performance simultaneously. Experimental results demonstrate the effectiveness of our approach and confirm that the BGMP algorithm could achieve satisfactory localisation and tracking results.

Abstract Benefitting from its ability to estimate the target state's posterior probability density function (PDF) in complex nonlinear and non‐Gaussian circumstance, particle filter (PF) is widely used to solve the target tracking problem in wireless sensor networks. However, the traditional PF algorithm based on sequential importance sampling with re‐sampling will degenerate if the latest observation appear in the tail of the prior PDF or if the observation likelihood is too peaked in comparison with the prior. In this paper, we propose an improved particle filter which makes full use of the latest observation in constructing the proposal distribution. Thequality prediction functionis proposed to measure the quality of the particles, and only the high quality particles are selected and used to generate the coarse proposal distribution. Then, acentroid shift vectoris calculated based on the coarse proposal distribution, which leads the particles move towards the optimal proposal distribution. Simulation results demonstrate the robustness of the proposed algorithm under the challenging background conditions. Copyright © 2010 John Wiley & Sons, Ltd.