Intelligent detection method for automobile front face part configuration

Abstract

The invention relates to the field of computer vision, in particular to an intelligent detection method for automobile front face part configuration, and the method comprises the steps: carrying out the image preprocessing of a to-be-detected automobile front face part image and a standard image, and obtaining a to-be-detected region image and a standard detection region image; edge pixels, key pixels and feature descriptors of the edge pixels and the key pixels of the standard detection area image are obtained, the attention of the pixels in the standard detection area image is calculated, a loss function of a K-SVD algorithm is corrected through the attention, and a dictionary matrix and a sparse coding matrix of the standard detection area image are obtained; and decomposing the gray matrix of the to-be-detected area image by using the dictionary matrix of the standard detection area to obtain a sparse coding matrix of the to-be-detected area image, and calculating the difference degree between the to-be-detected automobile front face part image and the standard image according to the sparse coding matrix of the to-be-detected area image and the sparse coding matrix of the standard detection area image. The method is high-efficiency and accurate, and whether qualification is achieved is judged according to the difference degree.

Description

technical field [0001] The application relates to the field of computer vision, in particular to an intelligent detection method for the configuration of front face parts of an automobile. Background technique [0002] The quality of auto parts directly affects the stability and safety of the car. With the increasing competition in the auto industry, the requirements for the quality of the produced cars are becoming more and more strict. If car manufacturers want to stand out in the fierce market competition, they need to The quality of the production vehicles must be continuously improved, and every step of the manufacturing process must be optimized. [0003] The task of the quality supervision station on the tail line of the assembly workshop is to check whether there are any wrong or missing parts in the configuration of the front face of the car, so as to ensure that every car produced is free of defects. At present, most of the inspection of the configuration of the fr...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides an intelligent detection method for the configuration of automobile front face parts, which solves the problem of inefficient and unreliable configuration of automobile front face parts, and adopts the following technical solutions:

Images