A New Steel Defect Detection Algorithm Based on Deep Learning

Abstract

In recent years, more and more scholars devoted themselves to the research of the target detection algorithm due to the continuous development of deep learning. Among them, the detection and recognition of small and complex targets are still a problem to be solved. The authors of this article have understood the shortcomings of the deep learning detection algorithm in detecting small and complex defect targets and would like to share a new improved target detection algorithm in steel surface defect detection. The steel surface defects will affect the quality of steel seriously. We find that most of the current detection algorithms for NEU-DET dataset detection accuracy are low, so we choose to verify a steel surface defect detection algorithm based on machine vision on this dataset for the problem of defect detection in steel production. A series of improvement measures are carried out in the traditional Faster R-CNN algorithm, such as reconstructing the network structure of Faster R-CNN. Based on the small features of the target, we train the network with multiscale fusion. For the complex features of the target, we replace part of the conventional convolution network with a deformable convolution network. The experimental results show that the deep learning network model trained by the proposed method has good detection performance, and the mean average precision is 0.752, which is 0.128 higher than the original algorithm. Among them, the average precision of crazing, inclusion, patches, pitted surface, rolled in scale and scratches is 0.501, 0.791, 0.792, 0.874, 0.649, and 0.905, respectively. The detection method is able to identify small target defects on the steel surface effectively, which can provide a reference for the automatic detection of steel defects.

Method for Diagnosis of Acute Lymphoblastic Leukemia Based on ViT‐CNN Ensemble Model

“…As one of neural networks used most frequently in deep learning, convolutional neural network (CNN) has strong self-learning ability, adaptive ability, and generalization ability. Traditional image recognition methods require manual feature extraction and classification, while CNN only needs the image data as an input of the network, and the self-learning ability of the network can complete the image classification [ 4 , 5 ]. Nahid et al used a multichannel convolution neural network to identify chest radiographs and diagnose pneumonia.…”

Recognition of Scratches and Abrasions on Metal Surfaces Using a Classifier Based on a Convolutional Neural Network

“…It is promising to create technological approaches that reduce defects in the area of plastic deformation of materials, in particular known solutions [16][17][18]. This article is also one of the steps to such technology, because the exact establishment of the nature of defects, their classification is the first step to diagnose the causes of their occurrence and further optimize the process of rolling strips.…”

Prediction of Inclusion Types From BSE Images: RF vs. CNN

“…Such techniques have been recently applied in the steel industry. Most of the current work has been focused on utilizing computer vision for the detection of surface defects and scratches (Wang et al, 1883;Konovalenko et al, 2021;Zhao et al, 2021). The focus of the work presented here is to utilize machine learning to predict inclusion types from their BSE images.…”