Evaluating the susceptibility of regional landslides is one of the core steps in spatial landslide prediction. Starting from multiresolution image segmentation and object-oriented classification theory, this paper uses the four parameters of entropy, energy, correlation, and contrast from remote-sensing images in the Zigui–Badong section of Three Gorges Reservoir as image texture factors; the original image data for the study area were divided into 2279 objects after segmentation. According to the various indicators of the existing historical landslide database in the Three Gorges Reservoir area, combined with the classification processing steps for different types of multistructured data, the relevant geological evaluation factors, including the slope gradient, slope structure, and engineering rock group, were rated based on expert experience. From the perspective of the object-oriented segmentation of multiresolution images and geological factor rating classification, the C5.0 decision tree susceptibility classification model was constructed for the prediction of four types of landslide susceptibility units in the Zigui–Badong section. The mapping results show that the engineering rock group of a high-susceptibility unit usually develops in soft rock or soft–hard interphase rock groups, and the slope is between 15°–30°. The model results show that the average accuracy is 91.64%, and the kappa coefficients are 0.84 and 0.51, indicating that the C5.0 decision tree algorithm provides good accuracy and can clearly divide landslide susceptibility levels for a specific area, respectively. This landslide susceptibility classification, based on multiresolution image segmentation and geological factor classification, has potential applicability.

This study demonstrates a light field display system using a nematic liquid crystal (LC) microlens array (MLA) and a polymer dispersed liquid crystal (PDLC) film. LC-MLA without polarization effects presented high-resolution intermediate 3D images by adopting a depolarization algorithm. The adopted PDLC film modulated the reconstructed 3D images to deliver full-parallax images efficiently with a wide FOV. The experimental result shows that the peak signal to noise ratio (PSNR) value of photograph accurate display results improves compared to the pure LC-MLA method. The proposed method is an essential step toward high-quality light field display.

Semantic segmentation is a fundamental and important task in the field of remote-sensing images. However, most convolutional neural network (CNN)-based methods often produce undesirable results of blurry object boundaries as convolution operations may inherently fail to retain the edge information while expanding the receptive field. Besides, CNNs struggle to capture subtle relationships between distant parts of an image. This limitation can be particularly detrimental to semantic segmentation as it requires a comprehensive understanding of the global image context. Transformers, nevertheless, excel at processing long-range dependencies. To address the above issues, we propose a novel three-branch network with three new modules, called GDBNet. The Swin-Transformer-based global (G) branch is for capturing global context information. The ResNet-based detail (D) branch and boundary (B) branch for processing local features and edge information. The three modules are feature preservation module (FPM), spatial interaction module (SIM), and balance feature module (BFM). FPM can maximumly retain the context information during the downsampling stage of Swin Transformer in the global branch, whereas SIM and BFM can fuse the information from different branches in a boundary-oriented manner. Experiments show that GDBNet outperforms state-of-the-art methods quantitatively and qualitatively on Vaihingen, Potsdam, and WHDLD datasets, reaching 71.87%, 75.64%, and 62.70% accuracy in MIOU metric, respectively, and generating visually better clearance and coherence of boundaries.

Semantic segmentation of remote sensing images is a key technology for land cover interpretation and target identification. Although convolutional neural networks (CNNs) have achieved remarkable success in this field, their inherent limitation of local receptive fields restricts their ability to model long-range dependencies and global contextual information. As a result, CNN-based methods often struggle to capture the comprehensive spatial context necessary for accurate segmentation in complex remote sensing scenes, leading to issues such as the misclassification of small objects and blurred or imprecise object boundaries. To address these problems, this paper proposes a new hybrid architecture called KPV-UNet, which integrates the Kolmogorov–Arnold Network (KAN) and the Pyramid Pooling Visual State Space Attention (PP-VSSA) block. KPV-UNet introduces a deep feature refinement module based on KAN and incorporates PP-VSSA to enable scalable long-range modeling. This design effectively captures global dependencies and abundant localized semantic content extracted from complex feature spaces, overcoming CNNs’ limitations in modeling long-range dependencies and inter-national context in large-scale complex scenes. In addition, we designed an Auxiliary Local Monitoring (ALM) block that significantly enhances KPV-UNet’s perception of local content. Experimental results demonstrate that KPV-UNet outperforms state-of-the-art methods on the Vaihingen, LoveDA Urban, and WHDLD datasets, achieving mIoU scores of 84.03%, 51.27%, and 62.87%, respectively. The proposed method not only improves segmentation accuracy but also produces clearer and more connected object boundaries in visual results.