PVC floor surface coating defect detection method and system

By combining frequency domain analysis with adaptive filtering, the problem of stable detection of subtle scratches on the surface of highly reflective PVC flooring was solved. This method enables accurate segmentation and detection of subtle scratches, overcomes the problems of missed detection and computational redundancy in traditional methods, and improves the stability and efficiency of detection.

CN122243867APending Publication Date: 2026-06-19ZHEJIANG GIMIG TECH CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202610160861.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-02-04
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing technologies for detecting minor scratches on highly reflective PVC flooring suffer from poor stability, high false negative rates, and high computational costs. They also struggle to effectively suppress strong specular reflection interference and adaptively enhance and segment minor scratch signals from different directions.

Method used

A method combining frequency domain analysis and adaptive filtering is adopted. The original image of PVC flooring is acquired, preprocessed, and subjected to spectral transformation and analysis. The set of main direction angles of scratches in the frequency domain is extracted, and frequency domain filtering and inverse transformation are performed based on adaptive filter mask to finally segment the scratches.

Benefits of technology

It achieves accurate segmentation and detection of weak, randomly oriented scratches under strong interference, effectively overcoming the problems of missed detection and computational redundancy in traditional methods, and improving the stability and efficiency of detection.

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

Abstract

The invention discloses a PVC floor surface coating defect detection method and system, and relates to the field of coating defect detection.The method comprises the steps that firstly, an original image containing strong specular reflection and low-signal-to-noise-ratio scratches is converted into a frequency domain, and the scratches serve as a linear structure to present directional characteristics in a frequency spectrum; the image is separated from large-area low-frequency background noise and glare signals, scratches in any direction existing in the current image are accurately recognized, and the main direction angle of the scratches is extracted. Then, based on the direction information detected in real time, an adaptive anisotropic filter completely matched with the direction information is dynamically constructed. And finally, through filtering and inverse transformation, feature signals of the scratches are remarkably amplified, and the contrast ratio is greatly improved, so that accurate segmentation and detection of weak and directional random scratches under a strong interference background are realized, and the problems of missing detection and calculation redundancy of a traditional method are effectively solved.
Need to check novelty before this filing date? Find Prior Art