Improving Productivity and Quality in Manufacturing by Applying Computer Vision Systems (Image Processing Technique)

摘要

Automated inspection systems is the target for all automated organizations. The objective zero defect considers as a challenge for many industries since there are many factors effect on production line. Increase scrape items effect on productivity and environment; rework produced items as bottle neck for production lines and decrease products rates. The objectives for the research project focused on four main concerning, Evaluate automated inspection and control system in manufactures. Redesign online inspection system for some industrial case studies for the purpose of enhancing Quality control, tracking quality control in manufacturing systems and embedded improving computer vision systems in the production lines levels, and reduce defect items by correct parameters during the production lines. Research project focused on some case studies like (plastic, hot stamping, assembly, and textile) industries in Malaysia and Iraq. Computer vision systems was the common methods since it considers as non-destruction testing system. One of the machine vision systems techniques is image processing technique. Image processing algorithm implement by using MATLAB and Simulink. The developed points in this research focused on interpret defects and signal feedback for correcting deviations in the setting parameter for the fabrication machines. This system will help manufacturers to understand faults for their products online during fabrication route. Three main functions were using feature matching, color recognition and orientation and recognize the object functions. The results for this system showed that the ability for the system to know the weak points in the produced items and the production systems and accurate them with keeping on the stability for the automated system. Growth in information technology and cameras will improve system capabilities in different fields and adaptable for heavy environments.