The primary challenges in image-level weakly supervised semantic segmentation (WSSS) lie in addressing the under-activation issue of target pixels and mitigating the co-occurrence phenomenon in class activation maps. In recent years, Vision-Language Models (VLM) have demonstrated exceptional performance across various vision tasks, ‌primarily attributed to their cross-modal semantic alignment capabilities achieved through contrastive learning mechanisms‌. Leveraging VLM’s capability to capture fine-grained visual-textual correspondences, this paper proposes a novel Vision-Language Driven Prompt Learning (VLD-PL) framework that addresses two fundamental challenges in WSSS by establishing explicit semantic correspondences between textual descriptors and visual components, ultimately enabling efficient semantic segmentation. The VLD-PL framework consists of two core components Auxiliary Class Matching (ACM) and Background Class Filtering (BCF). The ACM module dynamically identifies semantically relevant auxiliary classes through feature alignment between image and textual embeddings, effectively enlarging target activation while mitigating co-occurrence interference by expanding semantic coverage. Simultaneously, the BCF constructs image-specific background prompts and adaptively refines background feature representations, achieving precise suppression of irrelevant background regions. These dual mechanisms synergistically address both target localization accuracy and background noise suppression, achieving state-of-the-art performance on both the PASCAL VOC 2012 and MS COCO 2014 benchmarks.

Point cloud completion referring to completing 3D shapes from partial 3D point clouds is a fundamental problem for 3D point cloud analysis tasks. Benefiting from the development of deep neural networks, researches on point cloud completion have made great progress in recent years. However, the explicit local region partition like kNNs involved in existing methods makes them sensitive to the density distribution of point clouds. Moreover, it serves limited receptive fields that prevent capturing features from long-range context information. To solve the problems, we leverage the cross-attention and self-attention mechanisms to design novel neural network for point cloud completion with implicit local region partition. Two basic units Geometric Details Perception (GDP) and Self-Feature Augment (SFA) are proposed to establish the structural relationships directly among points in a simple yet effective way via attention mechanism. Then based on GDP and SFA, we construct a new framework with popular encoder-decoder architecture for point cloud completion. The proposed framework, namely PointAttN, is simple, neat and effective, which can precisely capture the structural information of 3D shapes and predict complete point clouds with detailed geometry. Experimental results demonstrate that our PointAttN outperforms state-of-the-art methods on multiple challenging benchmarks. Code is available at: https://github.com/ohhhyeahhh/PointAttN

Class activation maps generated by image classifiers are widely used as priors for image-level weakly supervised semantic segmentation. However, these activation maps mainly focus on the sparse discriminative regions, which has been a bottleneck for the segmentation task. Based on our observations, the activation maps actually capture almost the entire target regions, and some regions with lower activation values are easily to be neglected. Thus, to solve the issue, we propose an adaptive activation network with two branches to recalibrate the low-confidence regions in the activation maps. Specifically, an activation enhancement branch is designed to redistribute the activation values by leveraging attention mechanism. Since multi-scale images can provide complementary information, a scale adaptation branch is paralleled to supervise the activation enhancement branch. The mutual supervision and fusion of the two branches can promote the less-discriminative parts, and deactivate the background regions. Based on them, a simple yet effective denoising module is proposed to further improve the quality of pseudo masks, which makes use of the large scale predictions of the trained segmentation network. Extensive experiments on the PASCAL VOC 2012 and MS COCO 2014 benchmarks show that our method achieves state-of-the-art performance, demonstrating the effectiveness of our algorithm. Code will be made publicly available.

Abstract With the progress of communication technology, the increasing wireless devices are hooked up the networks, and the consequent security problems cannot be ignored. At present, the solutions mainly include secret key system and physical layer security (PLS). However, secret key system that relies on highly complex algorithms for encryption and decryption still has the risk of being cracked by powerful computers. In this paper, the PLS of Beaulieu‐Xie fading channels is studied. To improve the transmission quality and support higher speed data transmission, the single‐input multiple‐out (SIMO) system considered is the Wyner eavesdropper model with decoding and forwarding (DF) relay over Beaulieu‐Xie fading. The closed‐form theoretical formulas for secure outage probability (SOP) and strictly positive secrecy capacity (SPSC) are obtained by complex unified functional representation, which includes two cases: and . Moreover, the theoretical formulas are verified by Monte Carlo results. By comparing the simulation results with the theoretical analysis, the authors discussed the factors to improve the secure performance. Experiments show that the larger ratio of , the smaller ratio of , smaller signal‐to‐noise ratio (SNR) of eavesdropping channels, smaller threshold and more legitimate antennas are beneficial to the improvement of anti‐interference capability for the relay networks with Beaulieu‐Xie distributions when the legal channel link R→D is in a state of high SNR.

The combination of nonorthogonal multiplex access and unmanned aerial vehicles (UAVs) can improve the energy efficiency (EE) for Internet of Things (IoT). On the condition of interference constraint and minimum achievable rate of the secondary users, we propose an iterative optimization algorithm on EE. First, with a given UAV trajectory, the Dinkelbach method-based fractional programming is adopted to obtain the optimal transmission power factors. By using the previous power allocation scheme, the successive convex optimization algorithm is adopted in the second stage to update the system parameters. Finally, reinforcement-learning-based optimization is introduced to obtain the best UAV trajectory.

Though deep learning-based saliency detection methods have achieved gratifying performance recently, the predicted saliency maps still suffer from the boundary challenge. From the perspective of foreground-background separation, this article attempts to extract the edge information of objects by exploiting the difference between different color channels in the RGB color space and establishes a novel multicolor contrast extraction (MCE) mechanism to improve the learning ability of exquisite boundary information of the network. To make full use of the MCE outputs and RGB colors, and well depict and capture the complementary information between them, we devise a novel Siamese densely cooperative fusion (DCF) network (SDFNet) for saliency detection, which consists of two effective components: boundary-directed feature learning (BDFL) and DCF. The BDFL provides joint learning for both MCE and RGB modalities through a Siamese network, while the DCF module is devised for complementary feature discovery, in order to effectively combine the features learned from two modalities. Experiments on five well-known benchmark datasets demonstrate that the proposed method outperforms the state-of-the-art approaches in terms of different evaluation metrics. We provide a detailed analysis of these results and indicate that our joint modeling of MCE and RGB colors helps to better capture the object details, especially in the object boundaries.

Video snapshot compressive imaging (SCI) system enables high-frame-rate imaging by projecting multiple frames into a 2D snapshot measurement during a single exposure, and the original video frames can be reconstructed by solving an optimization problem. However, existing methods usually cannot achieve a good balance between reconstruction time and reconstruction quality, which has become a major obstacle for practical application of video SCI. In order to cope with this issue, we propose a residual ensemble network to learn the explicit inverse mapping from the 2D snapshot measurement to the original video. Specifically, the proposed network aims to exploit the spatiotemporal correlations between video frames for improving reconstruction quality. The spatiotemporal correlations of video frames demonstrate multiple types, including intra-frame spatial correlation, inter-frame forward and backward temporal correlation. With the purpose of fully capturing these differentiated correlations, we design four sub-networks, namely, a pseudo-3D U-shape sub-network, two residual sub-networks, and a serial forward and backward recurrent sub-network, and further assemble these four sub-networks into an ensemble network through alternate residual links. This ensemble network can effectively fuse the predictions of each sub-network and maintain spatiotemporal consistency between video frames. We further design a compound loss function to guide the network learning, and the new video can be fast reconstructed by simply feeding its 2D snapshot measurement into the learned network. The experimental results demonstrate that our network can significantly improve the reconstruction quality while maintaining low computational cost.

In this article, we tackle the saliency detection task from an interesting perspective: we focus both on salient regions (or foreground) detection and nonsalient regions (or background) detection instead of only the foreground and propose a novel complementarity-aware attention network. It is a unified framework with two branches, namely, positive attention module (PAM) and negative attention module (NAM), for the foreground and background detection, respectively. More specifically, the PAM exploits a position self-attention mechanism to enhance the discriminant ability of feature representation, which can detect most of the salient object regions. Meanwhile, the NAM is designed to detect the background regions, aiming to pop out the missing object parts and details in the prediction map produced by the PAM. By fusing these two attention modules together, NAM can provide complementary cues to assist PAM for precise object detection. Furthermore, in order to capture more multiscale contextual information, we introduce a bidirectional structure with multisupervision to the proposed complementarity-aware attention module for performance improvement. Experiments on five benchmark datasets show that the proposed framework achieves comparable results compared with the state-of-the-art saliency detection methods.

Internet of Medical Things (IoMT) is an application-specific extension of the generalized Internet of Things (IoT) to ensure reliable communication among devices$C_{i}$, designed for the medical industry. However, a challenging issue associated with these networks, i.e., IoMT and IoT, is to ensure the authenticity of both source and destination modules and further guarantee the integrity of the multimodal data in the emergencies such as the COVID-19 pandemic. Various mechanisms for device authentication have been presented in the literature to resolve both devices and data’s authenticity, integrity, and privacy. Still, authentication of mobile device-to-server (in both homogeneous and heterogeneous IoMT) is not explicitly addressed for the black-hole attack. In this article, a device-to-server andvice versamutual authentication scheme are presented to ensure secure communication sessions among numerous mobile devices$C_{i}$and server$S_{j}$in the operational IoMT. The proposed scheme is a hybrid of medium access control (MAC) and enhanced on-demand vector (EAODV)-enabled routing schemes. In the proposed scheme, an offline phase is introduced to complete the registration process of member devices with the concerned server module. It blocks every possible entry of the potential intruder devices$A_{k}$in the operational IoMT. A mobile device$C_{i}$interested in initiating a communication session with a particular server$S_{j}$is needed to pass the mutual authentication process. As a result, only registered devices$C_{i}$are allowed to communicate. Additionally, a reliable encryption and decryption scheme is used to ensure data reliability during these communication sessions. Simulation results verify the exceptional performance of the proposed mutual authentication scheme in terms of authenticity, security, and integrity of both devices and data in the operational IoMT.

This paper addresses the core issue of how to learn powerful features for saliency. We have two major observations. First, feature maps of different layers in convolutional neural networks play different roles in saliency detection. Second, different feature channels in the same layer are not of equal importance to saliency, and they often have different response to foreground or background. To address these problems, a stacked U-shape network with channel-wise attention is presented to effectively utilize these features, which mainly consists of a parallel dilated convolution (PDC) module and a multi-level attention cascaded feedback (MACF) module. More specifically, PDC aims to enlarge the receptive field without increasing the computation and effectively avoid the gridding problem. MACF is innovatively designed to adaptively select the cross-layer complementary information, and the inter-dependencies between different channel maps in the same layer can be depicted well. Finally, we adopt a multi-layer loss function to improve the commonly used binary cross entropy loss which treats all pixels equally. The extensive experiments on five saliency detection datasets demonstrate that the proposed method outperforms the state-of-the-art approaches.

With the development of Internet of Things (IoTs), devices are now connecting and communicating together on a heretofore unheard‐of scale, forming huge heterogeneous networks of mobile IoT‐enabled devices. For beyond 5G‐ (B5G‐) enabled networks, this raises concerns in terms of spectral resource allocation and associated security. Cognitive radio is one effective solution to such a spectrum sharing issue which can be adopted to these B5G networks, which works on the principle of sharing spectrum between primary and secondary users. In this paper, we develop the confidentiality of cognitive radio network (CRNs) for IoT over k‐μ fading channels, with the information transmitted between secondary networks with multiple cooperative eavesdroppers, under the constraint of the maximum interference that the primary users can tolerate. All considered facilities use a single‐antenna receiver. Of particular interest, the minimum limit values of secure outage probability (SOP) and the probability of strictly positive secrecy capacity (SPSC) are developed for this model in a concise form. Finally, the Monte Carlo simulations for the system are provided to support the theoretical analysis presented.

This paper investigates a secure wireless-powered sensor network (WPSN) with the aid of a cooperative jammer (CJ). A power station (PS) wirelessly charges for a user equipment (UE) and the CJ to securely transmit information to an access point (AP) in the presence of multiple eavesdroppers. Also, the CJ are deployed, which can introduce more interference to degrade the performance of the malicious eavesdroppers. In order to improve the secure performance, we formulate an optimization problem for maximizing the secrecy rate at the AP to jointly design the secure beamformer and the energy time allocation. Since the formulated problem is not convex, we first propose a global optimal solution which employs the semidefinite programming (SDP) relaxation. Also, the tightness of the SDP relaxed solution is evaluated. In addition, we investigate a worst-case scenario, where the energy time allocation is achieved in a closed form. Finally, numerical results are presented to confirm effectiveness of the proposed scheme in comparison to the benchmark scheme.

In this study, the authors propose a novel framework for top‐down (TD) saliency detection, which is well suited to locate category‐specific objects in natural images. Saliency value is defined as the probability of a target based on its visual feature. They introduce an effective coding strategy called locality constrained contextual coding (LCCC) that enforces locality and contextual constraints. Furthermore, a contextual pooling operation is presented to take advantages of feature contextual information. Benefiting from LCCC and contextual pooling, the obtained feature representation has high discriminative power, which makes the authors' saliency detection method achieving competitive results with existing saliency detection algorithms. They also include bottom‐up cues into their framework to supplement the proposed TD saliency algorithm. Experimental results on three datasets (Graz‐02, Weizmann Horse and PASCAL VOC 2007) show that the proposed framework outperforms state‐of‐the‐art methods in terms of visual quality and accuracy.

This paper presents the design, implementation and test of a watch with the capability to measure multiple physiological signals, such as photoplethysmography (PPG), electrocardiography (ECG), and some physiological parameters extracted from PPG and ECG. Based on the PPG and ECG signals, the pulse rate and heart rate could be extracted respectively. Moreover, the pulse transmit time (PTT) could be extracted combining PPG and ECG, which was the key to calculate blood pressure (BP). The circumstance conditions (temperature, humidity and barometric pressure) were detected and displayed on the screen. A motion state was used to toggle the watch screen. Meanwhile all data measured could be sent to an android smart phone, using a Texas Instruments' (TI's) Bluetooth low energy chip CC2540. The PPG wave, ECG wave, heart rate, as well as the systolic blood pressure (SBP) and diastolic blood pressure (DBP) can be displayed on the android client in real time. Compared with the sport band and other smart watches, this multi-parameter health monitoring watch had the advantage of monitoring more parameters. We had made a test to compare this watch with standard equipment, Benavidez T5 and SUNTECH OSAR2, and the result showed that the relative error of heart rate, pulse rate, SBP and DBP against reference were 1.76 ±1.64bpm, 1.76 ±1.64bpm, 4.01 ± 2.75 mmHg, 4.64 ± 3.31 mmHg. These results demonstrated that this watch has a broad prospect in the field of mobile health care.

The continuous monitoring of blood pressure (BP) has been found to significantly predict the risk of severe cardiovascular disease. Pulse arrival time (PAT), generally extracted from synchronized photoplethysmogram (PPG) and electrocardiogram (ECG) signals, is widely adopted in noninvasive blood pressure studies. However, motion artifact and physical activities introduce different levels of noise to the ECG and PPG signals in wearable devices, resulting in large fluctuations in PAT-based BP estimation, which may confuse and mislead users. We explored the potential of Kalman filter to enhance the stability of continuous BP estimation. We developed an Android application collecting data from the wearable device via Bluetooth technology, two Kalman filters were designed and implemented to evaluate the systolic blood pressure (SBP) and pulse pressure (PP) separately, whose gains were adjusted automatically by signal quality indicators. Validation experiments were performed on 6 volunteers to ensure the effectiveness of Kalman filters, and the preliminary results compared with a standard commercial sphygmomanometer showed that our approach can achieve higher stability than the method without Kalman filtering.

The death of the patients is an important event in the intensive care unit (ICU), mortality risk prediction thus offers much information for clinical decision making. However, Patient ICU mortality prediction faces challenges in many aspects, such as high dimensionality, imbalance distribution. In this paper, we modified the cost-sensitive principal component analysis (CSPCA), which is denoted by MCSPCA, to solve these problems. This modified method not only reduced the feature dimensionality but also better handled the imbalanced problem of the benchmark data. A support vector machine (SVM) model was used as the classifier to identify ICU mortality risk. As for SVM parameters optimization, a chaos particle swarm optimization (CPSO) was used to optimize the penalty parameter and the kernel parameter. The proposed model was compared with several contrast models (such as model without PCA). The test results indicate that the proposed model showed highest AUC of 0.7718 and minimum consumption time of 814s.

This paper has presented a wearable sleep apnea monitoring wrist watch, its micro controller controls sensors to detect user's respiratory airflow, blood oxygen saturation(SpO2), electrocardiogram(ECG) and breathing movement in real-time. Then our algorithm uses these physiological parameters to estimate whether the apnea occurs or not. In addition, this wrist watch provides motion state of users and some environmental parameters like temperature, humidity and atmospheric pressure. A liquid crystal display(LCD) module is used to show the parameters above. This wrist watch can send all the physiological parameters to the application on smartphone or computer by Bluetooth, where doctors can use these data for further analysis. Based on experimental test, the highly-integrated wrist watch can provide reliable measurement results. It avoids hospital's high cost and complicated procedures, and expands the screening scope of sleep apnea syndrome.

In this paper, we address the problem of fusing various saliency detection methods such that the fusion result outperforms each of the individual methods. We observe that the saliency regions shown in different saliency maps are with high probability covering parts of the salient object. With image regions being represented by the saliency values of multiple saliency maps, the object regions have strong correlation and thus lie in a low-dimensional subspace. Meanwhile, most of background regions tend to have lower saliency values in various saliency maps. They are also strongly correlated and lie in a lowdimensional subspace that is independent of the object subspace. Therefore, an image can be represented as the combination of two low rank matrices. To obtain a unified low rank matrix that represents the salient object, this paper presents a double low rank matrix recovery model for saliency fusion. The inference process is formulated as a constrained nuclear norm minimization problem, which is convex and can be solved efficiently with the alternating direction method of multipliers (ADMM). Furthermore, to reduce the computational complexity of the proposed saliency fusion method, a saliency model selection strategy based on the sparse representation is proposed. Experiments on five datasets show that our method consistently outperforms each individual saliency detection approach and other state-of-the-art saliency fusion methods.

Local feature embedding considers two constraints: intra-image spatial and inter-image feature affinity in the embedding process. However, it does not work well for the image classification task when the images are with intra-class variation, background clutter, etc. In this paper, we enhance the manifold structure by adding the class label of images into the embedding process. Since class labels are used in the training, our method can be considered as supervised. Four constituents are included in our model: feature consistency, spatial consistency, intra-class compactness and inter-class separability. With the defined Hausdorff distance between two images, different classifiers are exploited for classification. Extensive experiments on seven datasets demonstrate the effectiveness of our proposed image classification model.

We propose an adaptive filtering approach for interferograms, which is a modification to the Lee adaptive complex filter. Based on local frequency estimates, we compute the normal orientation of local phase fringes. A directionally dependent filtering window is aligned perpendicular to the normal orientation of local phase fringes (i.e., along local phase fringes) by interpolation, making the pixels included in the filtering window have approximately more homogeneous values. Moreover, the computation of the filter parameter does not require local phase unwrapping in the real plane. This filter minimizes the loss of signal and reduces the level of noise. By using two sets of simulated data, its effectiveness can be seen in terms of the fidelity to noise-free phases, fringe preservation, and residue reduction.