T-Shirt Defect Detection Using Yolo11
Dr. Hemavathy R Asst.Professor, Dept. of AIML
Sri Shakthi Institute of Engineering and Technology
Coimbatore
Nishanth S Dept. of AIML
Sri Shakthi Institute of Engineering and Technology
Coimbatore
Milind Krishna Dept. of AIML
Sri Shakthi Institute of Engineering and Technology
Coimbatore
Dhanasekar D Dept. of AIML
Sri Shakthi Institute of Engineering and Technology
Coimbatore
Ashin K Minil Dept. of AIML
Sri Shakthi Institute of Engineering and Technology
Coimbatore
Abstract: In the face of detection problems posed by complex T-shirt texture backgrounds, different sizes, and different types of defects, commonly used object detection networks have limitations in handling target sizes. Therefore, when the target types are more diverse, false detections or missed detections are likely to occur. In order to meet the stringent requirements of T-shirt defect detection, we propose a novel ACYOLOv11-based T-shirt defect detection method. This method fully considers the optical properties, texture distribution, imaging properties, and detection requirements specific to T-shirts. First, the Atreus Spatial Pyramid Pooling (ASPP) module is introduced into the YOLOv11 backbone network, and the feature map is pooled using convolution cores with different expansion rates. Multiscale feature information is obtained from feature maps of different receptive fields, which improves the detection of defects of different sizes without changing the resolution of the input image. Secondly, A convolution squeeze-and-excitation (CSE) channel attention module is proposed, and the CSE module is introduced into the YOLOv11 backbone network. The weights of each feature channel are obtained through self-learning to further improve the defect detection and anti-jamming capability.
Keywords: T-shirt defect; surface defect detection; deep learning; attention mechanism
