Electronic component surface quality detection method and system based on machine vision

Through image illumination reflection decoupling and fusion feature defect detection model, the problem of distinguishing real defects and pseudo defects in traditional detection methods is solved, and high-precision detection under complex lighting conditions is achieved.

CN120125589BActive Publication Date: 2025-08-12KUNMING UNIV OF SCI & TECH
2 Cites 0 Cited by

Patent Information

Application Number
CN202510617153.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-14
Publication Date
2025-08-12
Estimated Expiration
2045-05-14

AI Technical Summary

Technical Problem

Traditional electronic component surface quality detection methods fail to effectively distinguish between real defects and pseudo defects caused by light, and are difficult to cope with multi-scale and low-contrast defects, resulting in a degradation of detection performance under complex lighting conditions.

Method used

The original image is decoupled into reflectivity maps and light distribution maps by image illumination reflection decoupling method. Combined with the fusion feature defect detection model, through frequency band optimization, structural tensor analysis and dynamic feature fusion, the significance of defect features is enhanced and noise and pseudo-defects are removed.

Benefits of technology

It improves the contrast and detectability of defect areas, significantly improves detection accuracy and anti-interference ability, and can effectively deal with multi-scale defects.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention discloses a method and system for detecting the surface quality of electronic components based on machine vision. The method includes collecting original images, basic image processing, image illumination and reflection decoupling, building a quality detection model, and detecting the surface quality of components. The present invention relates to the field of electronic component image processing technology, and specifically refers to a method and system for detecting the surface quality of electronic components based on machine vision. The present invention obtains original component detection images by collecting original images; adopts basic image processing methods such as image cleaning, image enhancement, and image standardization; adopts an image illumination and reflection decoupling method to eliminate illumination unevenness and reflection interference, and improve the contrast and detectability of defective areas; and adopts a fusion feature defect detection model for component surface quality detection, which can effectively handle multi-scale defects and significantly improve detection accuracy and anti-interference ability.
Need to check novelty before this filing date? Find Prior Art