Improved YOLOV4-CSP Algorithm for Detection of Bamboo Surface Sliver Defects With Extreme Aspect Ratio

Abstract

Bamboo surface defect detection provides quality assurance for bamboo product manufacture in industrial scenarios, an integral part of the overall manufacturing process. Currently, bamboo defect inspection predominantly relies on manual operation, but manual inspection is very time-consuming as well as labor-intensive, and the quality of inspection is not guaranteed. In recent years, a few visual inspection systems based on traditional image processing have been deployed in some factories. However, traditional machine vision algorithms extract features in tedious steps and have poor performance along with poor adaptability in the face of complex defects. Accordingly, many scholars are committed to seeking deep learning methods to accomplish surface defect detection. However, existing deep learning object detectors struggle with specific industrial defects when directly applied to industrial defect detection, such as sliver defects, especially for ones with extreme aspect ratios. To this end, this paper proposes an improved model based on the advanced object detector YOLOV4-CSP, which introduces asymmetric convolution and attention mechanism. The introduction of asymmetric convolution enhances the feature extraction in the horizontal direction of the bamboo strip surface, improving the performance in detecting sliver defects. In addition, convolutional block attention module(CBAM), a hybrid attention module, which combines channel attention with spatial attention, is utilized to promote the representation ability of the model by increasing the weights of crucial channels and regions. The proposed model achieves outstanding performance in the general categories and excels in the hard-to-detect categories. Some enterprise's bamboo strip dataset experiments verify that the model can reach 96.74% mAP for the typical six surface defects. Meanwhile, we also observe significant improvements when extending our model to aluminum datasets with similar characteristics.

Investigation of You Only Look Once Networks for Vision-based Small Object Detection

“…Fig. 1 illustrates the backbone, neck, and head of the YOLOv5 architecture, which may be understood by analyzing its structural code [18,21,23]: Cross Stage Partial Networks (CSP) and the focal structure make up the backbone. The focus structure down samples the input data dimension while the original data is kept.…”

CPAM: Cross Patch Attention Module for Complex Texture Tile Block Defect Detection

“…To help models better focus on defective regions, many scholars have enhanced the visual perception capabilities of models by adding attention mechanisms. For instance, Li et al [15] create a Multiscale Residual Attention Unit (MRAU) and apply it to micromotor armature surface defect detection, Cheng et al [16] present DE-block for metal surface defect detection, and Guo et al [4] utilize CBAM for bamboo surface sliver defects detection. In the field of tile block defect detection, Wan et al [9] use CBAM to help the model to further improve the focus on the defect region.…”

“…In the field of industrial defect detection, according to [2][3][4]9,11,15,16], this study uses Precision (P), Recall (R), F1-score (F1), Average Precision of A category (AP), Average Precision of Multiple categories (mAP) are used to evaluate the proposed method. The calculation methods of Precision, Recall, F1-score, AP, and mAP are shown in Equations ( 7)- (11).…”

“…Vision-based inspection methods have developed rapidly in recent years, and many scholars have applied methods based on deep learning models to the field of industrial defect detection, such as Fabric Defect Detection [2], Electrical Insulator Defects Detection [3], and Bamboo Surface Sliver Detection [4]. In the field of tile defect detection, tile defect detection is still a difficult task due to the special characteristics of tile datasets, the great resolution of tile images, and the many types and tiny sizes of defects.…”

Pneumonia detection based on RSNA dataset and anchor-free deep learning detector

“…For example, we use a large preset bounding box with an approximate aspect ratio of 1:2 for lung detection with a large rectangular object, but if we use this object detection framework for lung nodule detection, we need to use a smaller square preset bounding box with an aspect ratio of 1:1. In the case of more complex when dealing with more complex subjects or setups, more complex parameter settings are required, which also places higher demands on the user 11 , which is clearly impractical and not user-friendly for clinicians without code and computer engineering experience.…”