空间控制技术与应用2023，Vol.49Issue(6)：94-103,10.DOI:10.3969/j.issn.1674-1579.2023.06.010
融合多尺度及注意力机制的表面缺陷检测算法
The Surface Defect Detection Algorithm Based on Multi-Scale Feature Fusion and Attention Mechanism
摘要
Abstract
The impeller blades of the engine are a key component of the propulsion system of a space spacecraft and play an important role in the success and efficiency of space missions.In order to solve the above problems,this paper proposes a defect detection algorithm(EF-CenterNet)that integrates multi-scale features and attention mech-anism,and uses the lightweight EPSANet network as the backbone of the CenterNet algorithm to effectively in-tegrate the PSA segmentation attention mechanism,pay attention to more important defect features,and enhance the feature extraction ability of the network.At the same time,the FPN structure is added after the feature layer output by the backbone feature extraction network to further integrate multi-scale information,that is,low-resolu-tion high-level semantic information and high-resolution low-level feature information,so as to improve the defect detection accuracy of the algorithm.Experimental results show that the proposed EF-CenterNet algorithm achieves an average accuracy of 96.74％in the self-made dataset,which is 1.81％higher than that of the baseline Center-Net algorithm,and an average accuracy of 77.37％in the public dataset.关键词
叶轮叶片/缺陷检测/注意力机制/多尺度/深度学习Key words
impeller blade/defect detection/attention mechanism/multi-scale/deep learning分类
信息技术与安全科学引用本文复制引用
步斌,张梦怡,王超,王村松,薄翠梅,彭浩..融合多尺度及注意力机制的表面缺陷检测算法[J].空间控制技术与应用,2023,49(6):94-103,10.基金项目
国家重点研发计划项目(2021YFB3301300)、江苏省高等学校自然科学研究面上项目(21KJB520007)、国家自然科学基金(62203213)和江苏省自然科学基金(BK20220332)National Key Research and Development Program of China(2021YFB3301300),Natural Science Research of Jiang-su Higher Education Institutions of China(21KJB520007),National Natural Science Foundation of China(62203213)and Natural Science Foundation of Jiangsu Province(BK20220332) （2021YFB3301300）
