A method and system for detecting tiny scratches on glass screen based on small sample deep learning
By using a small sample deep learning model in glass screen detection combined with a three-color light multi-band optical imaging method, the problems of inefficient and insufficient accuracy of traditional detection methods are solved, and efficient and accurate detection of micro scratches is achieved.
CN119359720BActive Publication Date: 2025-03-28HUNAN JINGWEI ELECTRONIC TECH CO LTD
0 Cites 0 Cited by
Patent Information
- Application Number
- CN202411919961.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-25
- Publication Date
- 2025-03-28
- Estimated Expiration
- 2044-12-25
AI Technical Summary
Technical Problem
Traditional glass screen micro scratch detection methods are inefficient and susceptible to subjective factors, and cannot meet the needs of high-quality and high-precision detection.
Method used
Using a detection system based on small sample deep learning, the glass screen image is collected under three separate color lights of red, green and blue through the image acquisition device, and combined with the small sample deep learning model for image recognition to generate scratch defect recognition results.
Benefits of technology
It realizes high accuracy of micro scratch detection in small samples, improves detection efficiency and accuracy, and meets the high-quality and high-precision detection needs.
✦ Generated by Eureka AI based on patent content.
Abstract
The present invention proposes a method and system for detecting tiny scratches on glass screens based on small sample deep learning, in order to solve the difficulties and shortcomings of detecting tiny scratch defects in the current glass screen production and processing technology. First, a small sample deep learning model is designed, and efficient training is achieved in a small sample environment through steps such as training the model with tiny scratch sample images and fine-tuning the model. Subsequently, the images captured by the image acquisition device are identified using the trained deep learning model, and a comprehensive judgment is made on whether there are tiny scratch defects on the glass screen under the image recognition of three-color light, and the identification results are displayed through a display device. The present invention solves the problem of detecting tiny scratches on the glass screen by using three-color light source assistance and high-resolution image acquisition technology combined with a small sample deep learning model, and realizes high-accuracy defect detection in small sample conditions through technologies such as data enhancement, pre-training and model fine-tuning.
Need to check novelty before this filing date? Find Prior Art