A machine vision-based quality detection method for basic cobalt carbonate powder material

CN122265280APending Publication Date: 2026-06-23JIANGXI NUCLEAR IND XINGZHONG NEW MATERIALS
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
JIANGXI NUCLEAR IND XINGZHONG NEW MATERIALS
Filing Date
2026-05-26
Publication Date
2026-06-23

AI Technical Summary

Technical Problem

Existing online detection technologies for basic cobalt carbonate powder suffer from low accuracy, high false alarm rate, inability to achieve multi-dimensional integrated detection, and delayed detection results, making it impossible to promptly report quality anomalies in the production process and easily leading to the generation of batches of unqualified products.

Method used

A machine vision-based detection method is adopted. By acquiring cross-polarized images and multi-exposure image sequences, and combining them with a reference color chart and qualified batch samples, true color restoration is performed. Powder layer morphology correction and thickness interference compensation are carried out, abnormal features are extracted and time-series tracking is performed, and a minimum re-inspection sampling window is generated to achieve batch quality inspection.

Benefits of technology

It improves detection accuracy, reduces false alarm rate, realizes multi-dimensional integrated detection, avoids detection lag, can promptly report quality abnormalities in the production process, and reduces production costs and quality control risks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122265280A_ABST
    Figure CN122265280A_ABST
Patent Text Reader

Abstract

The application discloses a kind of based on machine vision's basic formula cobalt carbonate powder material quality detection method, belong to powder quality on-line detection technical field.The method is by collecting the cross-polarization image and multiple exposure image sequence of basic formula cobalt carbonate powder, combine reference color plate and qualified batch sample to carry out true color recovery, and carry out powder layer morphology correction and thickness interference compensation to true color image, effectively inhibit residual wet reflection, uneven thickness and other pseudo abnormal interference;Again by abnormal feature extraction, time series tracking and clustering obtain stable defect cluster, generate minimum recheck sampling window and recheck guide after batch consistency determination.The application realizes the on-line, real-time, multi-dimensional quality detection of basic formula cobalt carbonate powder, significantly improves defect detection accuracy, reduces false positive rate and missed detection rate, while greatly reducing the recheck range.
Need to check novelty before this filing date? Find Prior Art