Intelligent detection method for inductance winding defects based on machine vision

By employing a dual-modal fusion detection method combining optical images and Barkhausen noise signals, along with a superconducting quantum interference device (QFID) probe array and shear wave flow field tensor technology, the problem of the inability to comprehensively detect surface and internal defects in traditional inductor winding detection methods has been solved, achieving high-precision defect identification and localization.

CN122193372APending Publication Date: 2026-06-12SHENZHEN SOREDE ELECTRONIC CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHENZHEN SOREDE ELECTRONIC CO LTD
Filing Date
2026-02-27
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Traditional inductor winding inspection methods cannot effectively detect surface and internal defects in the windings, and the consistency and repeatability of the inspection results are insufficient, making it difficult to meet high precision requirements.

⚗Method used

A dual-modal fusion detection method combining optical images and Barkhausen noise signals, along with a superconducting quantum interference device probe array and shear wave flow field tensor technology, was adopted to achieve comprehensive detection of defects in inductor windings.

🎯Benefits of technology

It improves the comprehensiveness and accuracy of defect detection, enabling simultaneous identification of surface and internal defects, providing detailed diagnostic information, and offering testing solutions for military or aerospace-grade products with high precision and high reliability requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122193372A_ABST
    Figure CN122193372A_ABST
Patent Text Reader

Abstract

The present application relates to a machine vision-based intelligent detection method for inductance winding defects, and relates to the technical field of machine vision, comprising the following steps: step 1: applying an alternating magnetizing field to the inductance to be detected, and performing background elimination and notch filtering on the matrix sequence of the Barkhausen noise magnetic field; step 2: respectively performing discrete shear wave transformation on adjacent front and rear image frames in the image frame sequence of the region of interest to form a shear wave flow field tensor sequence; step 3: performing sliding window segmentation on the envelope signal and extracting statistical features to form a Barkhausen noise feature tensor; step 4: fusing the spatial feature map sequence and the time feature map sequence to obtain a time sequence aggregation feature map; and step 5: up-sampling the time sequence aggregation feature map to generate a defect positioning heat map and determine the defect position coordinates; the present application realizes comprehensive detection of surface defects and internal defects through the dual-mode fusion of optical images and Barkhausen noise signals.
Need to check novelty before this filing date? Find Prior Art