سينا سيفي

Nowadays, the use of image processing approach has been proposed in various industries as one of the new methods for quality control of products. Considering the importance of using new approaches based on image processing in the direction of product quality monitoring, in this research, the fuzzy transformation approach based on Fourier transform equations was used to compress the image in order to reduce the computational complexity while maintaining the image quality. Therefore, the aim of this research is to provide a new approach to control the statistical process of image data using information extracted from black-white images based on the fuzzy transformation approach. In this research, first, a fuzzy transformation method based on generalized fuzzy function has been used to reduce the image dimensions while maintaining acceptable quality during image compression, and then the information extracted from the image using the generalized likelihood ratio control chart for detecting changes in process performance is monitored. Also, in this research, different generator functions have been used to check the performance of the model in process monitoring. To eva‎luate the performance of the proposed method, a real-world case study in the ceramic tile industry is considered along with various numerical examples under various scenarios. The results of the comparison between the proposed method and other developed methods show that the developed model in most cases has a better performance than the compared models in detecting the defect after applying changes in lighting intensity. Also, the proposed approach has better performance in fault detection for smaller fault sizes.

پردازش تصوير , تبديل‌هاي فازي , كنترل آماري فرايند , متوسط طول دنباله

image processing , fuzzy transformations , statistical process control , average run length

Sina Seifi

Dr. Rassoul Noorossana