Lightweight High-Resolution Keypoint Regression Network for Noncooperative Spacecraft Using Enhanced Attention Mechanisms

Extracting key features on noncooperative space targets is a prerequisite for completing primary space missions such as target localization and attitude measurement. However, the space environment is characterized by extreme variations in light intensity, occlusions, and significant differences in image scales, making traditional feature extraction methods challenging to apply to extracting feature points from noncooperative space vehicles. This article proposes a deep learning-based keypoint regression algorithm to improve accuracy in noncooperative space targets. Unlike traditional methods, which struggle with occlusions and lighting changes, our method leverages a hybrid attention mechanism and high-resolution networks (HRNets) to address these challenges effectively. First, a keypoint regression network model for noncooperative spacecraft is proposed, using an HRNet as the backbone. The model incorporates a hybrid attention mechanism during feature extraction and expands the receptive field during feature fusion to improve the accuracy of keypoint regression in occluded scenarios. Second, to address high memory usage issues and poor real-time performance caused by complex network models, a lightweight design is applied to the HRNet, ensuring accuracy and model efficiency. Finally, multiple experimental results validate the performance of the proposed deep learning-based keypoint regression algorithm for noncooperative space targets.