Recent advances in deep learning have driven significant success in synthetic aperture radar (SAR) automatic target recognition (ATR), particularly through convolutional neural network (CNN)-based classification algorithms. However, SAR images possess distinctive physical scattering properties owing to their unique imaging mechanism. Many deep learning algorithms rely solely on amplitude information, ignoring phase information, which may result in the loss of information in the original complex-valued SAR image and suboptimal performance. To tackle these problems, this article introduces a SAR target recognition approach based on complex-valued operations, designated as Complex-valued Residual Mish activation and Convolution Block Attention Module (CBAM) net (CRMC-Net). The CRMC-Net effectively utilizes the amplitude and phase information in complex-valued SAR data. Specifically, first, the elements of CNN, including the input and output layers, the convolution layers, the activation functions, and the pooling layers, are extended to the complex-valued domain. Secondly, in order to further enhance the representation ability of the model, multiscale information of the target is extracted through different convolution kernel sizes. Finally, the network constructs many complex-valued operation blocks to enhance the robustness of the designed network, including the complex-valued residual block (CReLBlock), complex-valued Mish (CMish) activation function, and complex-valued convolution block attention module (CCBAM). The experimental results obtained from the Moving and Stationary Target Capture and Recognition dataset (MSTAR) and OpenSARShip2.0 (OPS 2.0) dataset demonstrate that the proposed network model outperforms the traditional real-valued models. It can further reduce the classification error and enhance performance.

Fire poses a serious threat to the global economy, environment, and social stability, highlighting the need for rapid and accurate fire detection. Remote sensing combined with deep learning has outperformed traditional fire assessment methods. However, in early fire stages, small flame areas, class imbalance, and weak feature extraction hinder detection accuracy. This study proposes an end-to-end segmentation model called Flame Detail Enhancement U-Net (FDE U-Net), using Landsat-8 multispectral remote sensing data. The model incorporates the self-Attention and Convolutional mixture (ACmix) module and the Convolutional Block Attention Module (CBAM) into the encoder of the Residual U-Net. ACmix integrates self-attention and convolution to capture global semantic features while maintaining computational efficiency, improving both contextual awareness and local detail. CBAM enhances flame recognition by weighting important channel features and focusing spatially on small flame areas, helping address the class imbalance problem. Additionally, Haar wavelet downsampling is applied to retain image detail and improve the detection of small-scale flame regions. Experimental results show that the FDE U-Net model exhibits robust performance in fire detection, accurately extracting flame regions even when their proportion is low and the background is complex. The F1 score reaches 95.97%, significantly improving the class imbalance problem.

Change detection for synthetic aperture radar (SAR) images effectively identifies and analyzes changes in the ground surface, demonstrating significant value in applications such as urban planning, natural disaster assessment, and environmental protection. Since speckle noise is an inherent characteristic of SAR images, noise suppression has always been a challenging problem. At the same time, the existing unsupervised deep learning-based methods relying on the pseudo labels may lead to a low-performance network. These methods are high data-dependent. To this end, we propose a novel unsupervised change detection method based on curvelet fusion and local patch similarity information clustering (CF-LPSICM). Firstly, a curvelet fusion module is designed to utilize the complementary information of different difference images. Different fusion rules are designed for the low-frequency subband, mid-frequency directional subband, and high-frequency subband of curvelet coefficients. Then the proposed local patch similarity information clustering algorithm is used to classify the image pixels to output the final change map. The pixels with similar structures and the weight of spatial information are incorporated into the traditional clustering algorithm in a fuzzy way, which greatly suppresses the speckle noise and enhances the structural information of the changing area. Experimental results and analysis on five datasets verify the effectiveness and robustness of the proposed method.