Change detection (CD) network for process-oriented model design improves detection efficiency through more complete time modeling. However, the networks accumulated by convolutional operations are limited by the localization of convolutional kernels, resulting in limited perception of spatiotemporal relationships. Therefore, in this letter, a process-oriented CD network based on discrete wavelet transform is proposed by combining the frequency-domain information in the convolutional network. Specifically, the network constructs a dual-time image into a multiframe video stream through video modeling and extracts the change features of different scales, frequencies, and directions in video and image features from the frequency-domain perspective with the help of discrete wavelet transform, which enhances the perception of spatiotemporal relationships. Experimental results on the LEVIR-CD, GVLM-CD, and EGY-BCD datasets validate the effectiveness of the network.

Feature selection and representation in infrared small target detection (ISTD) and spatial localization are crucial for detection accuracy. However, existing methods are not highly accurate in small target detection against complex backgrounds. In this article, we propose a patch spatial attention network, termed the semantic token transformer network (STPSA-Net), to detect small targets from a novel perspective. This framework represents images as compact semantic tokens by a semantic token transformer (STT) module and models spatiotemporal context to refine the original features and enhance the representation capability of small target features. The PSAM divides extracted features into patches and integrates spatial and semantic information to restore spatial information and achieve precise localization. Extensive experiments on the SIRST, MFIRST, and NUDT-SIRST datasets show the proposed method’s accurate detection of infrared small targets and its superior performance compared with state-of-the-art approaches.

In recent years, many video anomaly detection methods have mainly used reconstruction and prediction based methods. However, due to the powerful encoding and decoding capabilities of autoencoders in reconstruction methods, the misjudgment rate of abnormal samples is high, and prediction methods are easily affected by environmental changes and data noise. Real time and accurate detection of pedestrian abnormal events still faces huge challenges. This article proposes a fast method VAD-KEMM for detecting abnormal behavior, which uses keyframe extraction and a multi task hybrid model. Firstly, key frames are extracted through segmented clustering and inter frame differences to improve detection efficiency; Then, a dual branch hybrid model is constructed using human skeletal information to improve reconstruction accuracy, and multi task learning is used to enhance prediction ability. The experimental results show that the AUC values of this method on the HR Shanghai Tech and HR Avenue datasets are 76.8% and 87.1%, respectively, indicating high detection efficiency and accuracy.

Convolutional neural networks (CNN) have been widely used in the field of remote sensing (RS) scene classification, which have achieved remarkable results. In RS scene classification, local key objects are particularly crucial for classification results. However, most existing CNN methods directly utilize the deep-level global features of CNN, ignoring object-level information in shallow features or leading to redundant and erroneous information when using shallow features. To fully utilize the important information in shallow features, we proposed an end-to-end contextual spatial-channel attention network (CSCANet) to learn multi-layer feature representations and further improve classification performance by employing shallow object-level semantic information. Firstly, ResNet34 is pre-trained to extract different levels of features. Secondly, a contextual spatial-channel attention module is constructed to generate contextual spatial-channel attention features by exploiting features at different levels. Finally, the triple loss function is combined with the central loss function to guide the model training. Experiments on three public RS scene classification datasets (UC-Merced, AID, and NWPU-RESISC45) demonstrate that the proposed method achieves highly competitive results.

This article reports the findings from simulating the spatial diffusion processes of memes over social media networks by using the approach of agent-based modeling. Different from other studies, this article examines how space and distance affect the diffusion of memes. Simulations were carried out to emulate and to allow assessment of the different levels of efficiency that memes spread spatially and temporally. Analyzed network structures include random networks and preferential attachment networks. Simulated spatial processes for meme diffusion include independent cascade models and linear threshold models. Both simulated and real-world social networks were used in the analysis. Findings indicate that the numbers of information sources and opinion leaders affect the processes of meme diffusion. In addition, geography is still important in the processes of spatial diffusion of memes over social media networks.