Steel plate surface defect detection method based on multistage characteristics of convolutional neural network

Abstract

The invention provides a steel plate surface defect detection method based on multistage characteristics of a convolutional neural network and relates to the technical field of steel plate defect detection. The method comprises the following steps: selecting a baseline network, pre-training the baseline network, and establishing a special defect detection data set for fine-tuning training; building an overall detection network and a multistage characteristic fusion network, and merging the two networks to obtain a defect detection network; finally, setting a loss function of the defect detection network, training hyper-parameters, and training the defect detection network to enable the baseline network, the multistage characteristic fusion network and a RPN (Risk Priority Number) to sharethe convolutional layer and calculated amount, thereby obtaining the completely trained defect detection network model and further detecting the steel plate surface defects. The steel plate surface defect detection method based on multistage characteristics of the convolutional neural network provided by the invention has strong defect classification ability, and specific types and accurate position information of the defects can be completely acquired. Moreover, configuration of hardware needed by detection is reduced.

Description

technical field [0001] The invention relates to the technical field of steel plate defect detection, in particular to a method for detecting steel plate surface defects based on convolutional neural network multi-level features. Background technique [0002] Surface defect detection is one of the important links in industrial manufacturing. A good detection result can guide the subsequent quality evaluation system to achieve accurate evaluation of steel plate quality. A complete defect detection system should be able to obtain the specific category and precise location of each defect. [0003] The ultimate goal of the surface defect detection method is to obtain the category and location of the target defect. The early traditional method is to use different methods to extract manual features, such as HOG and LBP, and then use the sliding window method to obtain the approximate location of the target, and finally use a classifier for classification based on these regions, su...

AI Technical Summary

Problems solved by technology

However, there are still problems with the defect detection method based on deep learning: such as the lack of corresponding data, the inability to obtain the precise location of the defect, and insufficient classification ability.

Images