Construction method of automobile paint film defect recognition system based on deep learning

Abstract

The invention discloses a construction method of an automobile paint film defect recognition system based on deep learning. The method comprises the following steps: collecting and labeling a plurality of to-be-detected vehicle body paint film original images, labeling comprising labeling defect areas and categories in the vehicle body paint film original images, after labeling, generating an xmlformat file for each labeled picture, and forming a labeled sample set; performing multiple times of multi-angle clipping sampling on each defect area marked in each picture of the marked sample set by utilizing a sampling block to obtain a defect area sampling set; the sampling block being square; training and testing the defect area sampling set by using a deep learning algorithm, and constructing an automobile paint film defect recognition model based on deep learning; and constructing an automobile paint film defect identification system based on deep learning. The method can effectively assist the recognition of the defects of the automobile paint film, makes up the defects of manual visual detection, and improves the quality of the automobile body paint film.

Description

technical field [0001] The invention relates to the technical field of deep learning, in particular to a deep learning method for constructing a vehicle paint film defect identification system. Background technique [0002] Automobile painting is a crucial link in the automobile manufacturing process. The painted body needs to be inspected and modified for surface paint film defects. The traditional industrial line defect detection system adopts human eye initial inspection and manual re-inspection. Due to the influence of human eye resolution, resolution speed and subjective consciousness of inspection workers, and long-term intensive work and reflection of white light will lead to workers' vision Fatigue, the efficiency of manual inspection is not high, and missed inspections often occur. SUMMARY OF THE INVENTION [0003] In view of the deficiencies of the prior art, the present invention aims to provide a deep learning method for constructing a vehicle paint film defec...

AI Technical Summary

Problems solved by technology

[0003] Aiming at the deficiencies of the prior art, the present invention aims to provide a method for constructing a car paint film defect recognition system based on deep learning, which can effectively assist the recognition of car paint film defects, make up for the shortcomings of manual visual inspection, and improve the quality of car body paint film

Images