AUTOMATED CONTROL OF MATERIAL DEFECTS IN FOOTWEAR PRODUCTION

Abstract

The article describes the stages of automated material defect control. It is noted that the technical inspection system uses high-resolution cameras to obtain a detailed image of the surface material used for the production of equipment. Pre-processing of the image is necessary to eliminate possible interference. This provides a clean image for further analysis.

Algorithms for determining key image parameters, namely: textural characteristics, color differences and geometric features, help identify your defects. After detecting defects, an initial distribution of defective elements is carried out. The detected defects are distributed by categories depending on their criticality. This allows you to make operational decisions about the further use of the material.

The module automatically decides if the material can be used, whether it needs to be rejected or sent for additional processing. This stage is integrated with production systems for instant response. The analysis results are displayed graphically: defective areas can be marked on the image, and a report is generated that can be used for further quality analysis and process optimization. All information about detected defects is used for continuous learning and improvement of algorithms. This allows the system to adapt to new conditions and increase the accuracy of analysis over time. Analysis of literature sources showed that the use of deep learning algorithms significantly increases the accuracy of defect detection, which can be adapted for material quality control. The use of machine vision, neural network algorithms and intelligent systems allows you to reduce costs, improve product quality and increase the efficiency of production processes. Further research and development in this area will contribute to the further improvement of control technologies and increase the competitiveness of enterprises