Golden finger defect detection method based on mixed multi-fine-grained frequency

By mixing multiple fine-grained frequencies of gold finger defect detection method, residual convolution and frequency attention modules are used to improve detection accuracy in complex scenarios, solving the problems of insufficient detection accuracy and stability in existing technologies.

CN120655645AActive Publication Date: 2025-09-16SHENZHEN ZHIJIAN HUITONG TECH CO LTD +1
2 Cites 3 Cited by

Patent Information

Application Number
CN202511157385.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-08-19
Publication Date
2025-09-16
Estimated Expiration
2045-08-19

AI Technical Summary

Technical Problem

Existing technologies have difficulty ensuring the accuracy and stability of gold finger defect detection in complex scenarios such as complex backgrounds, high reflective interference, multi-scale targets, and blurred edges.

Method used

A gold finger defect detection method based on hybrid multi-fine-grained frequencies is adopted. Encoding is performed through the residual convolution module, and the hybrid cross-frequency attention module extracts spatial features under different receptive fields and performs frequency domain decomposition. The pooling operation is combined to enhance the key frequency band response, and finally defect detection is performed through the decoding structure.

Benefits of technology

It significantly improves the detection accuracy and stability in complex scenes such as complex backgrounds, high reflective interference, multi-scale targets and edge blur, and enhances the detection capability of fine-grained defects.

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

Abstract

The invention discloses a gold finger defect detection method based on mixed multi-fine-grained frequency, relates to the technical field of gold finger defect detection, and solves the technical problem that gold finger defect detection precision and stability are difficult to ensure in a complex scene. The method comprises the following steps: coding a training data set to obtain a first feature map; inputting the first feature map into a hybrid crossover frequency attention module to obtain a second feature map, and inputting the second feature map into a multi-fine-grained frequency attention module to obtain a frequency response feature map; performing pooling operation on the frequency response feature map, adding output results in a channel dimension, and outputting a double-path attention map; performing element multiplication operation on the double-path attention map and the frequency response feature map to generate a frequency enhancement feature map; decoding based on the first feature map and the frequency enhancement feature map, and outputting a golden finger prediction image; and inputting the golden finger image to be detected to obtain a detection result. The method improves the recognition capability and robustness of the key abnormal region, and improves the detection precision and stability.
Need to check novelty before this filing date? Find Prior Art