Textile defect on-line detection and closed-loop control system and method

Abstract

This invention provides an online detection and closed-loop control system and method for textile defects, comprising an image acquisition module, an image preprocessing module, a feature extraction module, a defect identification module, a data fusion module, and a control module connected in sequence, and also includes a confidence perception unit and an online incremental learning engine. This invention achieves quantitative fusion of identification confidence and feedback confidence through the confidence perception unit, and, in conjunction with the online incremental learning engine, completes effective sample screening and dynamic adjustment of the learning rate based on the fused confidence, realizing online incremental optimization of the detection model. This effectively adapts to dynamic changes in fabric types and working environments, significantly improving the generalization ability to identify minor and irregular defects in textiles, avoiding accuracy decay after long-term model operation, and balancing the requirements of high-speed production lines for real-time detection, identification accuracy, and long-term operational stability, thus comprehensively improving the intelligent management and control level of textile production.

Images