Remote sensing (RS) images change detection (CD) is essential for the surveillance and prevention of geohazards. Nevertheless, the current deep learning (DL)-based CD methods still face challenges such as pseudo changes, missed detections, and edge noise due to the inadequate research of temporal differences and inconsistent viewing angles between the dual-temporal images. In order to improve the perception of spatiotemporal variations and effectively manage complex motion in spatiotemporal data, this paper proposes a spatiotemporal multi-scale fusion and optical flow warping network (SMOW-Net). Initially, the internal fusion property of 3D convolution enables the simultaneous extraction and fusion of feature information in dual-temporal images. The spatiotemporal multi-scale feature encoder (SMFE) module is proposed to mitigate the semantic gap between low-level and high-level features. This module is designed to aggregate complementary feature information between each level through temporal and spatial independent processing and flexible temporal transposed convolutional layers. Furthermore, the optical flow warper (OFW) module is intended to improve the spatiotemporal dynamic modeling capability in order to manage complex motion data effectively, where a two-channel spatial deformation field is autonomously learned by the network to guide feature alignment. The performance advantage of our network over eleven state-of-the-art methods (SOTA) on the GVLM-CD, LEVIR-CD, WHU-CD, S2Looking, and LEVIR-CD+ datasets is validated by experimental results. Finally, we also introduce SMOW-Net-LW, a lightweight variant with significantly reduced model complexity, suitable for resource-constrained settings, while still achieving excellent performance. The code for this work is available at https://github.com/ChundeLiao/SMOW-Net .

Abstract Currently, in the field of processing 3D point cloud data, two primary representation methods have emerged: point‐based methods and voxel‐based methods. However, the former suffer from significant computational costs and lack the ease of handling exhibited by voxel‐based methods. Conversely, the later often encounter challenges related to information loss resulting from downsampling operations, thereby impeding subsequent tasks. To address these limitations, this article introduces a novel density‐based representation method for voxel partitioning. Additionally, a corresponding network structure is devised to extract features from this specific density representation, thereby facilitating the successful completion of classification tasks. The experiments are implemented on ModelNet40 and MNIST demonstrate that the proposed 3D convolution can achieve the‐state‐of‐the‐art performance based on the voxels.

In recent years, neural architecture search (NAS) methods have been proposed for the automatic generation of task-oriented network architecture in image classification. However, the architectures obtained by existing NAS approaches are optimized only for classification performance and do not adapt to devices with limited computational resources. To address this challenge, we propose a neural network architecture search algorithm aiming to simultaneously improve the network performance and reduce the network complexity. The proposed framework automatically builds the network architecture at two stages: block-level search and network-level search. At the stage of block-level search, a gradient-based relaxation method is proposed, using an enhanced gradient to design high-performance and low-complexity blocks. At the stage of network-level search, an evolutionary multiobjective algorithm is utilized to complete the automatic design from blocks to the target network. The experimental results demonstrate that our method outperforms all evaluated hand-crafted networks in image classification, with an error rate of 3.18% on Canadian Institute for Advanced Research (CIFAR10) and an error rate of 19.16% on CIFAR100, both at network parameter size less than 1 M. Obviously, compared with other NAS methods, our method offers a tremendous reduction in designed network architecture parameters.

Moth-flame optimization algorithms are widely employed to solve optimization problems and achieve good performance. However, the algorithms suffer the shortcoming of prematurity because of the early gathering of flames. To solve this problem, the flame fusion mechanism is integrated to improve the exploratory behavior of the moth-flame optimization algorithm. The flame fusion mechanism provides a new way to evaluate the state of flame aggregation based on the distribution of flames and moths. When the concentration of flames is higher than the fusion threshold, the better flame will fuse other flames. And the fused flames will be regenerated to enhance the exploration behavior of the algorithm. At the same time, the fusion rate that determines the probability of flame fusion is introduced. The fusion rate changes during iteration to balance the exploration and exploitation behaviors of the algorithm. The improved moth-flame optimization is validated by ten benchmark functions. The results show that the optimization ability of the improved moth-flame optimization algorithm is improved, and the stability is higher than compared algorithms as well.

Relation extraction is an important task of information extraction. Most existing methods of Chinese language relation extraction are based on word input. They are highly dependent on the quality of word segmentation and suffer from the ambiguity of polysemic words. Therefore, a multi-feature fusion model is presented on the basis of character input, which integrates character-level features, word-level features and entity sense features into deep neural network models. Specifically, to alleviate the ambiguity of polysemy, the entity sense is introduced as external language knowledge to provide supplementary information for understanding the semantics of an entity in a given sentence. The Attention-Based Bidirectional Long Short-Term Memory Networks (Att-BLSTM) are proposed to capture features at the character level. To obtain more structural information, the convolutional layer (C-Att-BLSTM) is built upon the Att-BLSTM to capture features at the word level. Experiments are conducted on a public dataset of SanWen, and show that the proposed model achieves state-of-the-art results.

During the past decade, graph-based semi-supervised learning has become one of the most important research areas in machine learning and artificial intelligence community. In this paper, we propose a sub-graph to construct the graph for semi-supervised learning (SSL). The new graph is scalable so that it can be extended to large-scale data. Based on this graph, we then propose a sub-graph regularization for scalable SSL. It can also project the new-coming data to infer its label for handling out-of-sample problem. Simulation results show that the proposed method can achieve better performance compared with other state-of-the-art graph based SSL methods.

Recently, deep learning has been widely applied in various areas and achieved remarkable research findings. The major reason that makes the deep learning paradigm successful is that it can effectively learn a hierarchical feature structure for the training data. However, most deep learning algorithms rely on massive well-labeled training datasets and hyper-parameter configurations. This paper proposed a novel methodology that uses the geometric characteristics of line-segment representations to optimize the hyper-parameters for the deep networks. The methodology is applied to a line-segment-based stacked auto-encoder to verify its effectiveness. It is found that the line-segment-based visualizations can increase the interpretability of the deep models and facilitate the configurations for the hyper-parameters.

Shape representation is an extremely important and longstanding problem in the field of pattern recognition. Closed contour, which refers to shape contour, plays a crucial role in the comparison of shapes. Because shape contour is the most stable, distinguishable, and invariable feature of an object, it is useful to incorporate it into the recognition process. This paper proposes a method based on genetic algorithms. The proposed method can be used to identify the most common contour fragments, which can be used to represent the contours of a shape category. The common fragments clarify the particular logics included in the contours. This paper shows that the explicit representation of the shape contour contributes significantly to shape representation and object recognition.