A curve matching image processing technique for analysis of coolant flow patterns

A curve matching technique is proposed to process experimentally obtained images to permit analysis of coolant flow patterns, which will be used to determine suitable parameters for an in-process optical measurement method to be used in the machining processes. In the proposed method, an array of ideal images is created based on the theoretical results and the disparities with the actual images are examined through a recursive process. To quantify the assessment of the disparities, an error index is used to determine the most suitable match. The working principle of the proposed technique is presented and the recursive algorithm to examine the image disparities described. To improve the computational efficiency, techniques to offer the necessary enhancement are described. The results of the experimental test demonstrated the effectiveness of the proposed method.