Research on surface defect detection method of aluminum-plastic blister package based on improved DeepLabv3+
Aiming at solving the problems of low accuracy of small-scale defect detection,poor edge segmentation effect in complex texture background and slow detection speed of traditional semantic segmentation model for aluminum-plastic blister package defect detection,this paper proposes an im-proved surface defect detection method for aluminum-plastic blister package.Firstly,the backbone feature extraction network replaces the original Xception network with the lightweight MobileNetV 2 network,which significantly reduces the number of model parameters.Secondly,the efficient channel attention(EC A)module is cascaded in the feature extraction module and atrous spatial pyramid poo-ling(ASPP)module to accelerate the global feature fusion and reduce the loss of detail information,thus improving the segmentation accuracy of the model for small-scale defects.Finally,a boundary re-finement module is added to the decoder of DeepLabv3+to improve the segmentation accuracy of the model on the edges of the defect region under the complex texture background of the aluminum-plastic surface.Experimental verification is carried out on the self-built capsule board image dataset,and the results show that the proposed method has a maximum improvement of 14.50％in mean intersection over union(MIoU)score,a maximum reduction of 92.71 ms in single-image prediction time,and a maximum reduction of 47.67 MiB in the number of parameters compared with the traditional Deep-Labv3+,PSPNet,HRNet and other models.The method in this paper has a high recognition accura-cy and efficiency,which can realize the fast detection of aluminum-plastic blister package surface de-fects and has high applicability.
万方数据
维普
