Detecting product surface defects is an important issue in industrial scenarios. In the actual scene, the shooting angle and the distance between the industrial camera and the shooting object often vary, which results in a large variation in the scale and angle. In addition, high-speed cameras are prone to motion blur, which further deteriorates the defect detection results. In order to solve the above problems, this study proposes a surface defect detection model for industrial products based on attention enhancement. The network takes advantage of the lower-level and higher-resolution feature map from the backbone to improve Path Aggregation Network (PANet) in object detection. This study makes full use of multihead self-attention (MHSA), an independent attention block for enhancing the backbone network, which has made considerable progress for practical application in industry and further improvement of the surface defect detection. Moreover, some tricks have been adopted that can improve the detection performance, such as data augmentation, grayscale filling, and channel conversion of input images. Experiments in this study on internal datasets and four public datasets demonstrate that our model has achieved good performance in industrial scenarios. On the internal dataset, the mAP@.5 result of our model is 98.52%. In the RSDDs dataset, the model in this study achieves 86.74%. In the BSData dataset, the model reaches 82.00%. Meanwhile, it achieves 81.09% and 74.67% on the NRSD-MN and NEU-DET datasets, respectively. This study has demonstrated the effectiveness and certain generalization ability of the model from internal datasets and public datasets.

Aiming at the problems of low efficiency, high false detection rate, and poor real-time performance of current industrial defect detection methods, this paper proposes an industrial defect detection method based on an expanded perceptual field and feature fusion for practical industrial applications. First, to improve the real-time performance of the network, the original network structure is enhanced by using depth-separable convolution to reduce the computation while ensuring the detection accuracy, and the critical information extraction from the feature map is enhanced by using MECA (More Efficient Channel Attention) attention to the detection network. To reduce the loss of small target detail information caused by the pooling operation, the ASPF (Atrous Spatial Pyramid Fast) module is constructed using dilate convolution with different void rates to extract more contextual information. Secondly, a new feature fusion method is proposed to fuse more detailed information by introducing a shallower feature map and using a dense multiscale weighting method to improve detection accuracy. Finally, in the model optimization process, the K-means++ algorithm is used to reconstruct the prediction frame to speed up the model’s convergence and verify the effectiveness of the combination of the Mish activation function and the SIoU loss function. The NEU-DET steel dataset and PCB dataset is used to test the effectiveness of the proposed model, and compared to the original YOLOv5s, our method in terms of mAP metrics by 6.5% and 1.4%, and in F1 by 5.74% and 1.33%, enabling fast detection of industrial surface defects to meet the needs of real industry.

The pothole is a common road defect that seriously affects traffic efficiency and personal safety. Road evaluation and maintenance and automatic driving take pothole detection as their main research part. In the above scenarios, accuracy and real-time pothole detection are the most important. However, the current pothole detection methods can not meet the accuracy and real-time requirements of pothole detection due to their multiple parameters and volume. To solve these problems, we first propose a lightweight one-stage object detection network, the AAL-Net. In the network, we design an LF (lightweight feature extraction) module and use the NAM (Normalization-based Attention Module) attention module to ensure the accuracy and real time of the pothole detection process. Secondly, we make our own pothole dataset for pothole detection. Finally, in order to simulate the real road scene, we design a data augmentation method to further improve the detection accuracy and robustness of the AAL-Net. The metrics F1 and GFLOPs show that our method is better than other deep learning models in the self-made dataset and the pothole600 dataset and can well meet the accuracy and real-time requirements of pothole detection.