Systém pro kontrolu kvality dílů pomocí strojového vidění

Abstract

This thesis focuses on the design and implementation of a machine vision system intended for automated surface defect inspection of manufactured parts. The goal is to develop a functional solution capable of detecting defects such as scratches, pigment anomalies, or burn marks by comparing a test image to a reference image. The system employs an industrial monochrome camera, diffused LED lighting, and image preprocessing algorithms including contrast normalization, edge enhancement, and morphological operations. In the practical part, the system is tested on real samples and achieves detection accuracy exceeding 95%. The evaluation confirms the crucial importance of consistent lighting and mechanical conditions during image acquisition. The results indicate that the proposed system is suitable for industrial deployment and provides a solid foundation for further development.