A method for Extracting the Actual contour size of Object images under Specific conditions

Edge is one of the most basic features of image. How to extract image edge information quickly and accurately is a research hotspot in the field of image recognition at home and abroad. Image edge is a set of pixels whose gray value changes step by step or similar to roof changes. It contains the most critical information in the image. How to quickly and effectively extract the edge information of the image is the key point in the whole image processing process. In order to improve the acquisition accuracy of edge image information processing, eliminate the influence of edge burr and shadow, and establish an efficient edge contour segmentation fitting model, this paper proposes a method to extract the actual contour size of the target image under specific conditions. Firstly, the image of the target object to be measured is corrected by affine transformation to ensure high measurement accuracy. Then, the gray image of the target object is obtained by gray processing of the transformed image, and the sub-pixel edge contour image of the image can be obtained. Then the image edge contour segmentation and fitting to get the fitting image. Finally, in the experimental part, use the correlation operator of Halcon to obtain the outer contour of the edge of the target object image and the distance of the center of the circle on the calibration plate image. Then the actual physical size of the edge of the target object to be measured can be finally obtained by the ratio method. Experiments show that the method is of great significance to improve the recognition accuracy of target object images.