Automatic building extraction from high-resolution remote sensing imagery has various applications, such as urban planning and land use management. However, the existing methods for automatic building extraction focus on regional accuracy but not on boundary quality. As a result, these building extraction methods yield discontinuous results due to tree and shadow occlusions and complex building roof materials. In this article, BOMSC-Net, which can consider boundary optimization and multi-scale context awareness, is proposed to address these two problems. BOMSC-Net is an encoder-to-decoder network composed of a basic network, a multi-scale context-aware module (MSCAM), and a direction feature optimization module (DFOM). The basic network learns to predict building segmentation maps through attention gates (AGs) that connect low-level feature jumps to advanced semantics. MSCAM is used as a bridge, which uses graph reasoning blocks (GRBs) to solve intra-class discontinuities. Finally, the DFOM constructs a direction field (DF) and then uses direction information for boundary correction to achieve fine building segmentations. Experiments were conducted on the open Massachusetts building dataset, WHU aerial building dataset, and Inria aerial image labeling dataset. The quantitative and qualitative analyses demonstrate that the proposed BOMSC-Net can help building extraction overcome interruptions caused by shadows and occlusions, extract buildings with different scales and materials, and realize boundary optimization and complete segmentation of buildings.

The rapid development of Earth observation technology has promoted the continuous accumulation of images in the field of remote sensing. However, a large number of remote sensing images still lack manual annotations of objects, which makes the strongly supervised deep learning object detection method not widely used, as it lacks generalization ability for unseen object categories. Considering the above problems, this study proposes a few-shot remote sensing image object detection method that integrates context dependencies and global features. The method can be used to fine-tune the model with a small number of sample annotations based on the model trained in the base class, as a way to enhance the detection capability of new object classes. The method proposed in this study consists of three main modules, namely, the meta-feature extractor (ME), reweighting module (RM), and feature fusion module (FFM). These three modules are respectively used to enhance the context dependencies of the query set features, improve the global features of the support set that contains annotations, and finally fuse the query set features and support set features. The baseline of the meta-feature extractor of the entire framework is based on the optimized YOLOv5 framework. The reweighting module of the support set feature extraction is based on a simple convolutional neural network (CNN) framework, and the foreground feature enhancement of the support sets was made in the preprocessing stage. This study achieved beneficial results in the two benchmark datasets NWPU VHR-10 and DIOR. Compared with the comparison methods, the proposed method achieved the best performance in the object detection of the base class and the novel class.

Military conflicts have a significant impact on national security and the ecological environment. Effective detection methods for ground-moving objects in complex scenarios can be further utilized to assess damage and provide recommendations for security and environmental restoration. Current ground-moving object detection in optical remote sensing imagery struggles with balancing detection accuracy and real-time performance, hindering timely threat assessment. To address this, the study proposes GMODet, a real-time detection method incorporating region awareness and semantic-spatial interaction to enhance detection of partially occluded and fine-grained objects in complex environments. The framework includes three modules: the Region Awareness Module (RAW), Cross-scale Context-aware Feature Aggregator (CCFA), and Semantic-Spatial Progressive Interaction Module (SPIM), focusing on extracting discriminative features for contextual, multi-scale, and semantic-spatial information. A new dataset, GMOD Dataset, is constructed with four object types and high scene complexity, alongside experiments on the publicly available MVRSD Dataset. GMODet achieves state-of-the-art performance, with mAP ₅₀ ,mAP ₇₅ , and mAP scores of 65.5%, 48.5%, and 42.3% on the GMOD dataset, outperforming the second-best results by 1.9%, 5.1%, and 1.5%, respectively. On the MVRSD Dataset, it achieves mAP ₅₀ ,mAP ₇₅ , and mAP scores of 88.2%, 75.2%, and 61.7%, respectively. Notably, with an inference time of just 25 seconds on large-scale images (9,152×9,152 pixels), GMODet showcases outstanding accuracy, speed, robustness, and generalization in ground-moving object detection.

In order to meet the requirements of high accuracy, high real-time and low number of parameters in lane detection algorithm in industrial practice, a lightweight network detection algorithm based on MoblieNet+UNet is proposed. This method takes UNet as the main structure of the network and applies MoblieNet to the process of network encoding. MoblieNet can decompose standard convolution into deep convolution and pointwise convolution when encoding. It can significantly reduce the amount of network parameters and improve the real-time performance of detection algorithm. At the same time, the UNet network framework can also segment the pixels with high precision, which ensures the accurate extraction of lane information. The experimental results show that the average cross merge ratio of the training set is 87% and 82.1 % respectively on the MHDD and bctd data sets, and the average cross merge ratio of the verification set is 84.5% and 81.6% respectively. The average time of a single frame in the PC terminal is 40ms. Experimental results show that the algorithm can extract lane line features in complex environment efficiently and accurately.

1. Wildfires are among the most destructive natural disasters, causing significant loss of life and property while profoundly impacting global climate change (Balch et al. 2022; Cunningham, William...