Process Capability Analysis for Image Data based on Coefficient of Variation

As the application of machine vision system in industry becomes more and more widespread, product information is more and more presented in the form of image data. Process capability analysis of image data has become an important issue in statistical process control. This analysis is important to measure that products meet the quality standards. This paper proposes a method for analyzing process capability based on the coefficient of variation of image data. Firstly, the average value of the gray value of each region after segmentation is extracted by super-pixel segmentation of the image data. Secondly, the upper and lower limits of specifications are determined by the maximum and minimum values of the average values of each region. Finally, the process capability index and its nonconformity rate are calculated. Experiments are conducted on printed circuit board (PCB) as an example, and the nonconforming rate in actual production are compared. The results show that the method can effectively analyze the image data for process capability and improve the efficiency of quality control.