Open-set Supervised Anomaly Detection (OSAD) strategy seeks to detect novel anomalies that are unseen during training. However, existing OSAD works fail to learn a comprehensive margin that separates normal and anomalous samples in the feature space owing to the lack of restriction on abnormal distribution. To this end, we propose a new OSAD scheme based on Margin Metric Learning, which fully exploits the differences between normal and abnormal events in the latent feature dimension. The developed framework embodies three major components. First, video clips including normal events and restricted anomalies are input into an Attention-embedded Spatial Convolutional Network to extract the spatial feature sequences. Then, the spatial feature sequences are input into a Temporal Convolutional Siamese Network to further obtain the temporal features. Second, a novel Quadruplet Contrastive Loss is designed based on the spatial-temporal features of normal and abnormal video clips to conduct the margin learning, which enlarges the inter-class distances between anomalous and normal instances as well as reduces the intra-class distances inside anomalous instances and normal instances. This contributes to acquire compact normal and abnormal distributions. Finally, during the testing stage, a simplified metric distance-based anomaly detection strategy is proposed to calculate the anomaly score of testing clip. Extensive experiments and ablation studies on the Avenue and ShanghaiTech datasets demonstrate the effectiveness and efficiency of our proposed method for discovering unseen anomalous events via limited types of aware anomalies.

Coal mine power load fluctuation serves as a critical precursor to safety accidents, with its clustering analysis holding significant value for safety production monitoring. To address the challenges of high-dimensional load data feature extraction, the separation between feature learning and clustering tasks in existing deep clustering methods, and the embedding space distortion caused by KL divergence loss, this paper proposes a Stacked Autoencoder with Combined Improved Loss for Coal Mine Daily Load Deep Embedded Clustering (SAE-CS-K-means). The methodology comprises three main components: Firstly, a layer-wise pretraining mechanism of stacked autoencoders extracts deep temporal features from load data, achieving effective mapping from high-dimensional data to low-dimensional embedding space. Secondly, a novel CS clustering loss function is designed to optimize intra-cluster compactness and inter-cluster differentiation. Finally, a joint optimization strategy synchronously updates network parameters and cluster centers, maintaining dynamic equilibrium between reconstruction loss and clustering loss, thereby ensuring embedding space stability and enhancing clustering performance. Experimental validation using actual power consumption data from a Yunnan coal mine demonstrates significant improvement in clustering quality through Silhouette Coefficient (SC), Davies-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI) evaluations. The clustering results provide precise quantitative references for abnormal power consumption diagnosis and reliable technical support for coal mine safety monitoring.

Autonomous Aerial Vehicles (AAVs) based Vehicle Re-Identification (ReID) brings high flexibility to the ReID system, and also brings challenges of complicated shooting views and special occlusions. In this study, we propose a novel framework for UAV-based vehicle ReID, calledCounterfactual Attention and Joint Promotion based ReID(CAJP-ReID), to deal with fine-grained feature extraction and occlusions. Based on the mechanism of using randomly generated counterfactual attention intervention to train, theCounterfactual Attention Learning Network(CAL-Net) is proposed to learn the fine-grained features of vehicle images, for distinguishing similar vehicles in different ID categories. In order to enhance the diversity of the dataset and the robustness of the network to occluded images, theCounterfactual Attention Enhancement Module(CAE-Module) is proposed based on counterfactual attention mechanism, by the data-enriching mechanism of cropping and erasing. TheHard-sensitive Binomial Joint Promotion Loss(HBJP-Loss) is proposed to comprehensively consider the relative distance and absolute distance of positive and negative samples of vehicle images, and further improves the accuracy of vehicle re-identification. Experiments on two public datasets show that the proposed method achieves State-of-the-Art performance.

Simultaneous polarimetric radars (SPRs) are powerful tools for measuring the polarization scattering matrix of targets in a single pulse. However, a major challenge associated with SPRs is the design of orthogonal waveforms that can effectively minimize the interference arising from simultaneous transmission and reception. Currently, frequency modulation (FM) and phase-coded modulation are the most commonly used signal structures. However, these signal structures suffer from certain limitations that necessitate the development of new orthogonal waveforms that are both well-suited to high-power transmitters and maximally free in design. This article proposes an approach that leverages polyphase-coded FM waveforms to design orthogonal structures that combine the benefits of FM and phase-coded waveforms. A mathematical model for designing such waveforms is established considering a matched filter at the receiver. The model unifies the parameters of orthogonality and sidelobe level into a fraction expressed in terms of the sidelobe integral level, main-lobe integral level of autocorrelation functions, and entire-lobe integral level of cross correlation functions. This fraction accurately characterizes the orthogonality and sidelobe level of a pair of waveforms. The optimal waveform group can be obtained using a binary gradient descent algorithm, with peak sidelobe levels and isolation levels simultaneously approaching −30 dB. Finally, this article describes the implementation of the proposed orthogonal waveforms on hardware and verifies their performance by presenting experimental results that confirm the effectiveness of the proposed method.

Compared to the steady-state part of the extended target echo, few scholars have paid enough attention to the echo establishment and disappearance part. They contain abundant target information, but have not been fully explored and utilized. Since the pulse duration of both parts is typically short, they can be considered as instantaneous processes. Polarization is an inherent property of electromagnetic waves, which can bring information gain to the analysis of the echo instantaneous part. The objective of this study is to utilize the instantaneous polarization response (IPR) of the echo to achieve super-resolution for two targets with different polarization scattering characteristics. It is analyzed that when two target echoes are “equal-amplitude and opposite-phase", echo decoupling can be achieved, which is the precondition for achieving range super-resolution. While making full use of the polarization information is instrumental in achieving the echo with “equal-amplitude and opposite-phase". A trihedral and a dihedral are selected to verify the effectiveness of the proposed method. Multiple sets of measured data show that the two target echoes do have the instantaneous part. When the signal-to-noise ratio (SNR) of the echo data is about 6-7dB, the range super-resolution and range estimation of the two targets are achieved by using the edge threshold detector.

Action recognition in video understanding is a challenging task, largely because of the complexity and difficulty in temporal modeling, making it suffer from motion information loss and misalignment of temporal attention in spatial dimensions. To overcome these difficulties, we propose a novel temporal modeling method calledAdjoint Enhancement Network(AE-Net), which can fully explore clues of motion and time in the long-range structure. The AE-Net mainly consists of two new modules: theInitial Adjoint Enhancement Module(IAE-Module), which deals with shallow features; and theGlobal Adjoint Enhancement Module(GAE-Module), which deals with global features. With a novel mechanism of parallel spatio-temporal convolution and difference fusion, the IAE-Module is to enhance the degree of motion transformation in shallow network features, exciting the potential of motion flow and avoiding motion information loss. The GAE-Module is proposed to improve the local temporal representation in long-range structures by feeding the enhanced feature differences into a spatial cascade module with residuals to resolve the misalignment of temporal attention in the spatial dimension.The experimental results show that our AE-Net can achieve state-of-the-art results in Something-Something V1, UCF-101 and HMDB-51 datasets.

The deep neural network model based on self-attention (SA) for obtaining rich contextual information has been widely adopted in semantic segmentation. However, the computational complexity of the standard self-attentive module is high, which partly limits the use of this module. In this work, we propose the lightweight self-attention network (LSANet) for semantic segmentation. Specifically, the Lightweight Self-Attentive Module (LSAM) captures information using a hand-designed compact feature representation, and weighted fusion of position information. In the decoder structure, an improved up-sampling module is proposed. Compared with the bilinear upsampling, this method achieves better results in restoring image details. The experimental results on PASCAL VOC 2012, and Cityscapes datasets show the effectiveness of our method, which simplifies operations and improves performance.

The realistic false targets caused by interrupted sampling repeater jamming (ISRJ) can mask the real target, leading to the failure of radar target detection. In this letter, a method based on jointly designing complementary sequences and receiving filters under the signal-to-noise ratio loss constraint is proposed to suppress ISRJ. A gradient-based nonlinear programming solver and the Lagrange multiplier method are used to optimize the sequences and filters alternately. Numerical results show that the proposed method can generate sequences and receiving filters with good correlation properties and anti-ISRJ performance.

Abstract Enhancing network feature representation capabilities and reducing the loss of image details have become the focus of semantic segmentation task. This work proposes the bilateral attention network for semantic segmentation. The authors embed two attention modules in the encoder and decoder structures . Specifically, high‐level features of the encoder structure integrate all channel maps through dense channel relationships learned by the channel correlation coefficient attention module. The positively correlated channels promote each other, and the negatively correlated channels suppress each other. In the decoder structure, low‐level features selectively emphasize the edge detail information in the feature map through the position attention module. The feature expression of semantic segmentation is improved by feature fusion of the two attention modules to obtain more accurate segmentation results . Finally, to verify the effectiveness of the model, the authors conduct experiments on the PASCAL VOC 2012 and Cityscapes scene analysis benchmark data sets and achieve a mean intersection‐over‐union of 74.92% and 66.63%, respectively.

Time-efficient anomaly detection and localization in video surveillance still remains challenging due to the complexity of “anomaly”. In this paper, we propose a cuboid-patch-based method characterized by a cascade of classifiers called a spatial-temporal cascade autoencoder (ST-CaAE), which makes full use of both spatial and temporal cues from video data. The ST-CaAE has two main stages, defined by two proposed neural networks: a spatial-temporal adversarial autoencoder (ST-AAE) and a spatial-temporal convolutional autoencoder (ST-CAE). First, the ST-AAE is used to preliminarily identify anomalous video cuboids and exclude normal cuboids. The key idea underlying ST-AAE is to obtain a Gaussian model to fit the distribution of the regular data. Then in the second stage, the ST-CAE classifies the specific abnormal patches in each anomalous cuboid with reconstruction error based strategy that takes advantage of the CAE and skip connection. A two-stream framework is utilized to fuse the appearance and motion cues to achieve more complete detection results, taking the gradient and optical flow cuboids as inputs for each stream. The proposed ST-CaAE is evaluated using three public datasets. The experimental results verify that our framework outperforms other state-of-the-art works.

In this paper, we present a duo-graph Simultaneous Localization and Mapping (SLAM) method that enables visual-guided robot to efficiently and robustly generate consistent 3D map in a large-scale environment. Recently the conditional independent graph (CI-GRAPH) has been proved to be an efficient and robust method for solving the large-scale SLAM problem. However, it stems from extended kalman filter (EKF) SLAM that suffers from high computational load when number of the landmarks becomes large. Moreover, the big noise in the measurement can lead EKF SLAM far from convergence. These disadvantages are the cases in the visual-guided robot. To solve these problems, we propose a duo-graph structure that sustains the feasibility of a hierarchical graph-based SLAM framework. It implements two level graph-based SLAM: local and global SLAM. Utilizing the characteristic of the duo-graph structure, the local SLAM can efficiently localize itself while the global SLAM can accurately close loops in the large scale environment. In addition, our method can filter massive noisy visual features and eliminate mismatches in the global SLAM process. To demonstrate the superiority of our method, both simulation and real experiments are carried out.