CN-122023256-A - Defect positioning method and system of screen bracket based on image processing

CN122023256ACN 122023256 ACN122023256 ACN 122023256ACN-122023256-A

Abstract

The invention relates to the technical field of image data processing, in particular to a defect positioning method and system of a screen bracket based on image processing. The method comprises the steps of obtaining local variation intensity of each pixel point through gradient amplitude difference of the pixel points in the surface image of the screen bracket under gradient operators of different scales, obtaining suspected pixel points in the surface image of the screen bracket through the local variation intensity of each pixel point, calculating target neighborhood radius of the suspected pixel points and variation rate of the suspected pixel points, calculating defect degree of the suspected pixel points, wherein the defect degree is positively correlated with local variation intensity and variation rate of the suspected pixel points and LBP positive value extraction value and LBP negative value extraction value in the target neighborhood radius of the suspected pixel points, and locating defect areas in the surface image of the screen bracket according to comparison results of the suspected pixel points and threshold values in the surface image of the screen bracket, so that accuracy of locating defects of the screen bracket can be effectively improved.

Inventors

MI YONG
LIU BIAO
ZHANG YONGQIN
Wang Yunhuan
TONG XIAOJUN
WU JIANBO

Assignees

东莞汇美模具制造股份有限公司

Dates

Publication Date: 20260512
Application Date: 20251226

Claims (10)

1. The defect positioning method of the screen bracket based on image processing is characterized by comprising the following steps of: obtaining local variation intensity of each pixel point through gradient amplitude difference of the pixel points in the surface image of the screen bracket under gradient operators of different scales; Obtaining a suspected pixel point in the surface image of the screen bracket through the local change intensity of each pixel point, and obtaining a target neighborhood radius of the suspected pixel point according to the gray scale difference of the suspected pixel point in the neighborhood radius ranges of the adjacent neighborhood radii at two sides of the suspected pixel point; Calculating the defect degree of the suspected pixel, wherein the defect degree is positively correlated with the local change intensity and change rate of the suspected pixel, and the LBP positive value extraction value and the LBP negative value extraction value in the target neighborhood radius of the suspected pixel; And responding to the comparison result of the suspected pixel points in the surface image of the screen bracket and the threshold value, and positioning the defect area in the surface image of the screen bracket.
2. The method for positioning defects of a screen bracket based on image processing according to claim 1, wherein the obtaining the local variation intensity of each pixel point by the gradient amplitude difference of the pixel point in the surface image of the screen bracket under different scale gradient operators further comprises: and acquiring a screen bracket diagram of the automobile central control, and preprocessing to obtain gray values of all pixel points in the screen bracket surface image.
3. The method for positioning defects of a screen bracket based on image processing according to claim 1, wherein the obtaining local variation intensity of each pixel point by gradient amplitude difference of the pixel point in the surface image of the screen bracket under different scale gradient operators comprises: the absolute value of the difference value between the gradient amplitude of the pixel under each scale gradient operator and the neighborhood gradient amplitude mean value is recorded as a first index of the pixel, and the difference value between the maximum value and the minimum value in the neighborhood gradient amplitude mean value of the pixel under all scale gradient operators is recorded as a second index of the pixel; and calculating the local variation intensity of the pixel point, wherein the local variation intensity is positively correlated with the first index, the second index and the third index.
4. The method for positioning defects of a screen bracket based on image processing according to claim 1, wherein the obtaining the suspicious pixel point in the surface image of the screen bracket by the local variation intensity of each pixel point comprises: and processing the local variation intensity of each pixel point in the surface image of the screen bracket by using an Ojin method to obtain the suspected pixel point in the surface image of the screen bracket.
5. The method for positioning defects of a screen bracket based on image processing according to claim 4, wherein the step of processing the local variation intensity of each pixel point in the surface image of the screen bracket by the oxford method to obtain a suspected pixel point in the surface image of the screen bracket comprises the steps of: The method comprises the steps of dividing local variation intensity of each pixel point in a screen bracket surface image into a suspected defect class and a normal class by taking the local variation intensity of each pixel point in the screen bracket surface image as a threshold value, calculating local variation intensity inter-class variance between the suspected defect class and the normal class, taking the suspected defect class and the normal class corresponding to the maximum value of the inter-class variance as a final classification result, and taking the pixel points in the suspected defect class in the final classification result as suspected pixel points.
6. The method for positioning defects of a screen bracket based on image processing according to claim 5, wherein the processing of the local variation intensity of each pixel point in the surface image of the screen bracket by the oxford method further comprises: And carrying out morphological operation on the normal pixel points in the surface image of the screen bracket to obtain the suspected pixel points in the normal class.
7. The method for locating defects of a screen bracket based on image processing according to claim 1, wherein the obtaining a target neighborhood radius of a suspicious pixel point comprises: And iteratively obtaining a difference maximum value of the gray average value of the suspicious pixel points in the neighborhood radius range and the gray average value of the suspicious pixel points in the neighborhood radius ranges adjacent to the two sides of the neighborhood radius, and taking the neighborhood radius corresponding to the difference maximum value as the target neighborhood radius of the suspicious pixel points.
8. The method for locating defects in an image processing-based screen bracket according to claim 1, wherein the calculating the defect level of the pseudo pixel point comprises: And calculating the defect degree of the suspected pixel point according to the local change intensity and change rate of the suspected pixel point and the LBP positive value extraction value and the LBP negative value extraction value in the target neighborhood radius, wherein the defect degree is positively correlated with the local change intensity and change rate and the LBP positive value extraction value and the LBP negative value extraction value in the target neighborhood radius.
9. The method for locating defects in an image processing-based screen bracket according to claim 1, wherein the locating the defective area in the surface image of the screen bracket in response to the comparison of the suspicious pixel point in the surface image of the screen bracket with the threshold value comprises: and taking the suspected pixel point with the defect degree larger than the threshold value in the surface image of the screen bracket as a defective pixel point and carrying out connected domain analysis to obtain a defective region in the surface image of the screen bracket.
10. A defect localization system for an image processing based screen support, comprising a processor and a memory, the memory storing computer program instructions which, when executed by the processor, implement the method for defect localization of an image processing based screen support according to any one of claims 1-9.

Description

Defect positioning method and system of screen bracket based on image processing Technical Field The invention relates to the technical field of image data processing, in particular to a defect positioning method and system of a screen bracket based on image processing. Background In modern automotive industry manufacturing, automotive center screen brackets are brackets designed for automotive center consoles for supporting and securing center screens or other related equipment. If the quality of the screen support does not meet the quality standard, the problems of shaking, falling and the like can possibly occur in the running process of the automobile, so that the driving safety problem is generated, and a large potential safety hazard exists. In order to accurately identify quality defects of a bracket, a plurality of bracket image defect detection methods are provided in the prior art, for example, patent application document with publication number of CN119130918A discloses an intelligent detection method and system for frame bracket fracture defects in an electronic product. However, in order to clearly analyze possible defects in the stent, the acquired stent image is usually a high-resolution image, multiple iterative and folding operations increase the complexity of the algorithm, and the prior art uses a template matching method to locate feature contours, so that the prior art depends on the training quality of the model and the completeness of the data set, and if the template is improperly selected or the difference between the template and the picture to be detected is large, matching failure or false detection rate may be increased. Based on this, how to accurately locate the defects in the screen bracket is a problem to be solved by those skilled in the art. Disclosure of Invention In order to solve the technical problem of how to accurately position the defects in the screen bracket, the invention provides a defect positioning method and a defect positioning system for the screen bracket based on image processing. In a first aspect, the present invention provides a method for positioning a defect of a screen bracket based on image processing, which adopts the following technical scheme: the defect positioning method of the screen bracket based on image processing comprises the following steps: Obtaining local variation intensity of each pixel point through gradient amplitude difference of the pixel points in the surface image of the screen bracket under gradient operators of different scales, obtaining a suspected pixel point in the surface image of the screen bracket through the local variation intensity of each pixel point, obtaining a target neighborhood radius of the suspected pixel point according to the gray level difference of the suspected pixel point in the neighborhood radius ranges of the neighborhood point and two sides of the neighborhood point, normalizing the gray level variance value of the suspected pixel point in the target neighborhood radius range to obtain the variation rate of the suspected pixel point, calculating the defect degree of the suspected pixel point, positively correlating the defect degree with the local variation intensity and the variation rate of the suspected pixel point and the LBP positive value extraction value and the LBP negative value extraction value in the target neighborhood radius of the suspected pixel point, and positioning the defect area in the surface image of the screen bracket according to the comparison result of the suspected pixel point and the threshold value in the surface image of the screen bracket. The invention processes the surface image of the screen bracket by the LBP algorithm, and can accurately locate and mark the defect area in the surface image of the screen bracket. In the process, the method and the device take the fact that the geometric structure of the screen bracket can cause high-brightness textures to cover low-brightness textures, so that defect recognition errors are generated by an LBP algorithm into consideration, and based on the method and the device, the suspected pixel points possibly in a defect area are obtained by analyzing gradient change consistency of the pixel points, and defects in the surface image of the screen bracket can be accurately recognized based on the possibility that the suspected pixel points are the defect area. On the basis, the invention also considers the situation that the defect dark edge with lower brightness possibly has identification errors, so the probability that the suspected pixel point is a defect area is corrected through the LBP positive value extraction value and the LBP negative value extraction value, and the defect degree of the suspected pixel point can be accurately obtained by combining the probability that the suspected pixel point is the defect area and the probability of the dark edge, and the accuracy of identifying the defect area in the surface image