Gypsum board defect detection method and system based on bidirectional structured light

By combining bidirectional structured light and twin neural networks in gypsum board production, the randomness problem caused by unidirectional laser detection is solved, defect detection with higher accuracy and robustness is achieved, costs are reduced and detection efficiency is improved.

CN120685647APending Publication Date: 2025-09-23BEIXIN BUILDING MATERIALS (TIANJIN) CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202510789450.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-13
Publication Date
2025-09-23

AI Technical Summary

Technical Problem

The existing gypsum board defect detection system uses unidirectional laser detection, which results in unconstrained detection image features, strong detection randomness, and reduced detection accuracy.

Method used

A method combining bidirectional structured light and twin neural networks is adopted. Structured light is emitted from both sides of the gypsum board at a preset angle using a structured light emitter. Bidirectional detection images are collected using a high-speed industrial camera. Image recognition training is performed using the twin neural network. A defect detection network is constructed and defect recognition is performed in combination with a classifier.

Benefits of technology

The accuracy and robustness of defect detection are improved, detection errors are reduced, costs are reduced, detection efficiency is improved, and external environmental interference is avoided.

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

Abstract

The invention relates to the technical field of plasterboard production, in particular to a plasterboard defect detection method and system based on bidirectional structured light, and the method comprises the following steps: emitting bidirectional structured light for detecting plasterboard defects from two sides of a plasterboard conveying assembly to the surface of a plasterboard at a preset angle by using a structured light emitter; a high-speed industrial camera is used for collecting a bidirectional detection image formed by irradiation of the bidirectional structured light on the surface of the gypsum board and used for detecting the defects of the gypsum board; and performing image recognition training on the bidirectional detection image by using a twin neural network to obtain a defect detection network for obtaining a defect detection result of the gypsum board. According to the invention, the twin neural network is utilized to carry out image recognition training on the bidirectional detection image to obtain the defect detection network for obtaining the defect detection result of the gypsum board, the defect detection features are obtained from different directions, mutual constraint is realized, and the detection error is reduced.
Need to check novelty before this filing date? Find Prior Art