Method and system for detecting cable body label of fiber distribution box

Through the combination of multi-scale feature extraction and fusion, global feature calculation and weight allocation, and rotation detection head module, the complex problem of the label detection environment of the fiber-separated box line body is solved, and efficient and accurate label recognition is achieved.

CN119693329APending Publication Date: 2025-03-25CHINA TELECOM DIGITAL INTELLIGENCE TECH CO LTD HEBEI BRANCH
0 Cites 0 Cited by

Patent Information

Application Number
CN202411763457.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-03
Publication Date
2025-03-25

AI Technical Summary

Technical Problem

The wire body label detection environment of fiber-separated box is complex, with problems such as dense wire body, label occlusion and large changes in label angle, making it difficult for existing detection methods to achieve efficient and accurate identification.

Method used

Multi-scale feature extraction and fusion, global feature calculation and weight allocation, and rotation detection head module are adopted to achieve efficient and accurate detection of fiber-separated box line body labels through the combination of feature extraction network, cross-channel self-attention module, visual state holding module and rotation detection head module.

Benefits of technology

Effectively respond to the complex environmental problems in the label detection of fiber-separated box lines, improve the accuracy and robustness of label inspection, and can effectively identify challenges such as dense line bodies, label occlusion and large changes in label angles.

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

Abstract

The invention provides a method and system for detecting a cable body label of a fiber distribution box, and relates to the technical field of image detection. The method comprises the following steps: acquiring a target image containing a label on each line body in a target fiber distribution box; inputting the target image into a feature extraction network to extract multi-scale feature information of different dimensions, performing feature map size alignment through a convolution and / or interpolation method, and performing feature information fusion of different scales by using a cross channel self-attention module to obtain first feature information; inputting the first feature information into a visual state keeping module to calculate global feature information and distribute different weight information according to the distance of the features so as to obtain second feature information; and inputting the second feature information into a rotary detection head module to generate size, category and angle information of the target so as to obtain label detection result information. According to the scheme, the line body label in the fiber distribution box with a complex line body environment can be efficiently and accurately identified.
Need to check novelty before this filing date? Find Prior Art