Research on Defect Detection in Automated Fiber Placement Processes Based on a Multi-Scale Detector

• DOI: 10.3390/electronics11223757
• Publication Date: 2022-11-17T08:27:44Z
• Authors: Yongde Zhang; Wei Wang; Qi Liu; Zhonghua Guo; Yangchun Ji
• Keywords: Pyramid (geometry); Feature (linguistics); Pascal (unit); Fuse (electrical); Pruning

Various surface defects in automated fiber placement (AFP) processes affect the forming quality of components. In addition, defect detection usually requires manual observation with naked eye, which leads to low production efficiency. Therefore, automatic solutions for recognition have high economic potential. this paper, we propose a multi-scale AFP algorithm, named spatial pyramid feature fusion YOLOv5 channel attention (SPFFY-CA). The (SPFFY) adopts dilated convolutions (SPDCs) fuse maps extracted different receptive fields, thus integrating information. For obtained from concatenate function, (CA) can improve representation ability network and generate more effective features. sparsity training pruning (STP) method is utilized achieve slimming, ensuring efficiency accuracy detection. experimental results PASCAL VOC our datasets demonstrate effectiveness scheme, achieves superior performance.