Dongsheng Liu

Remote sensing images often degrade during acquisition due to various environmental factors, leading to noise contamination and loss of texture details. Existing methods based on convolutional neural networks (CNNs) are limited by their local receptive fields, making it difficult to effectively model long-range dependencies. Although Transformers possess global modeling capabilities, they face high computational costs and poor scalability in high-resolution remote sensing images. To address these challenges, this paper proposes an efficient remote sensing image denoising network—OSSMDNet—based on the Mamba network and incorporating an omni-directional selective scanning mechanism (OSSM). Its advantages include (1) introducing a multi-directional state-space modeling mechanism to enhance spatial structure perception capabilities and mitigate the limitations of traditional unidirectional modeling; (2) OSSMDNet is designed based on the Mamba architecture, achieving efficient fusion of global context and local details while maintaining linear computational complexity. On multiple remote sensing and natural image denoising datasets such as CBSD68 and DOTA, OSSMDNet significantly outperforms existing CNN-, Transformer-, and Mamba-based methods in terms of Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) metrics, with PSNR and SSIM values 0.14 dB and 0.0033 higher than the most iconic Mamba baseline method, respectively. This demonstrates that the proposed OSSMDNet achieves an excellent balance between accuracy and efficiency.

Hyperspectral unmixing can handle the mixed pixels in hyperspectral images (HSIs). Shadows of objects in observed areas are recorded by sensors, resulting in an HSI contaminated by shadows. Therefore, shadow pollution is a grievous obstacle for unmixing applications. Although shadow pollution occurs frequently in HSIs, previous unmixing studies have never considered the interference caused by shadows. Hence, mitigating shadow interference for unmixing will be significant for further acquiring subpixel information. In this letter, we employ a generative adversarial autoencoder (GAA) to develop a supervised unmixing method that can substantially reduce the impacts of shadow for unmixing. Specifically, we adopt the GAA to establish an anti-shadow unmixing network (GAA-AS), where the encoder block is used to feature reinforcement, and the decoder serves for abundance estimation. Moreover, we adopt a spectral-aware loss (SAL) as the loss function of adversarial training, which makes the discriminator better capture the difference between pixels. Finally, a softmax layer is adopted for the abundance sum-to-one constraint (ASC). Several experiments verify the effectiveness and advantages of our GAA-AS. In the experiment with shadow-polluted data, the proposed GAA-AS improves accuracies by approximately 70% compared to SOTA approaches in the quantitative experiment with synthetic data, and the impacts of shadow pollution are also significantly alleviated in the experiment with real shadow-polluted HSIs. Additionally, note that the proposed GAA-AS is competitive even when no shadow exists in HSIs, verified by the experiment with shadowless data.

Lakes are important natural resources closely related to human survival and development. Based on PIE cloud computing platform, the study uses Landsat images and the empirical normalized water body index (ENDWI) to extract water body information of the large lakes in Sichuan province from 2000 to 2020 in the drought and rainy seasons, respectively, and uses the Mann–Kendall test to obtain the long-term trends of their area and climate. On this basis, the evolution of the lakes and their correlation with climate and human activities are analyzed. The results show that (1) In the past 20 years, the area of Lugu Lake, Qionghai Lake, and Luban Reservoir represent a decreasing trend, with Lugu Lake being the most affected. The area of Ma Lake, Three Forks Lake, and Shengzhong Reservoir increased, with the area of Shengzhong Reservoir increasing significantly; (2) During the drought season, all six lakes showed a decreasing trend in precipitation, with the most apparent decreasing trend for Lugu Lake (Slope = −0.8). Only Lugu Lake showed a decreasing trend in precipitation (Slope = −0.15) during the rainy season. The precipitation of Ma Lake, Three Forks Lake, Luban Reservoir and Shengzhong Reservoir showed a significant increasing trend (Slope value was greater than 1.96); (3) The temperatures of the remaining lakes all decreased in the drought season and increased in the rainy season, except that the temperature of Shengzhong Reservoir decreases throughout the year; (4) The area change of plain lakes is greatly affected by human activities, but the area of plateau lakes is are more impacted by climate. Our study improved the accuracy of long-term water body change monitoring with PIE-Engine Studio. Besides, the findings would provide reference for the implementation of sustainable water resources management in Sichuan Province.

Remote sensing-based change detection (CD) is a critical technique of detecting land surface changes in earth observation. Inspired by recent success of a lightweight CD network– 3M-CDNet, we implemented 9 meaningful modifications to it, for example, the incorporation of MHSA (Multi-Head Self-Attention). This elaborately designed model is termed as 3M-CDNet-V2. Its effectiveness and advantages were demonstrated on three engineering CD datasets, and experimental results indicated that: (1) relative to other state-of-the-art algorithms, the proposed model obtained very competitive or slight better performance on both visual comparison and quantitative metrics evaluation. (2) By applying a novel transfer learning strategy, 3M-CDNet-V2 could perform well on the small dataset. (3) The incorporation of MHSA brings a substantial accuracy improvement while moderately increasing the computational complexity. (4) The late fusion framework and deep supervision contribute most to the performance gain of 3M-CDNet-V2, while the introduction of low-level features to the classifier by skip connection could guide the model to focus on detailed spatial information such as small changes, narrow shaped objects, or accurate boundaries. We hope that our 3M-CDNet-V2 model helps in improving the understanding of network architecture design for CD.

This paper proposes a new ship detection algorithm via saliency analysis on multiscale superpixels. First, Synthetic Aperture Radar (SAR) imagery is over-segmented into many homogenous regions, called superpixels. In contrast to prior work, these superpixels are generated with various scales to handle size variation of ships in the SAR imagery. At each scale, local contrast and ship geometrical characteristics are considered to enhance the saliency of ship pixels. Then, the final saliency map is computed by integrating saliency values of all scales. Experimental results on two real SAR images demonstrate the efficiency of the proposed algorithm.

Extraction of agricultural parcels from high-resolution satellite imagery is an important task in precision agriculture. Here, we present a semi-automatic approach for agricultural parcel detection that achieves high accuracy and efficiency. Unlike the techniques presented in previous literatures, this method is pixel based, and it exploits the properties of a spectral angle mapper (SAM) to develop customized operators to accurately derive the parcels. The main steps of the method are sample selection, textural analysis, spectral homogenization, SAM, thresholding, and region growth. We have systematically evaluated the algorithm proposed on a variety of images from Gaofen-1 wide field of view (GF-1 WFV), Resource 1-02C (ZY1-02C), and Gaofen-2 (GF-2) to aerial image; the accuracies are 99.09% of GF-1 WFV, 84.42% of ZY1-02C, 96.51% and 92.18% of GF-2, and close to 100% of aerial image; these results demonstrated its accuracy and robustness.

Coastline detection using a Gaussian Mixture Model (GMM) applied to synthetic aperture radar (SAR) imagery is usually inaccurate due to the inherent noise of SAR data. In addition, the traditional active counter model is sensitive to the initial position of the contour line and requires a large number of iterations to converge to a solution. In this study, we first used the GMM algorithm to segment the SAR images and obtain a coarse land and sea segmentation map. This map is then used as the initial position for a subsequent active contour model. The K distribution was introduced into the local statistical active contour model to better model the SAR image. The Gaussian distribution-based local active contour model and the algorithm detailed in this paper were used to perform coastline extraction experiments on four SAR images. Four GF-3 SAR images with different modes were collected to validate the efficiency of the proposed method. The experimental results show that the coastline extraction methods from SAR images based on the GMM algorithm and the K distribution-based local statistical active contour model (LKDACM) overcame the shortcomings of the traditional active contour model to accurately and quickly detect coastlines, thus enabling the detection of coastline changes.