The Technical Research of Ferrography Division of Morphology

Abstract

During the ferrographic image analysis and its technical research, the accuracy of the ferrography segmentation directly affect the analysis accuracy of the ferrograph’s particle. In order to realize the automatic and quick analysis of ferrographic image, First of all, we do morphological close operation to the binary image to eliminate the small holes in the target area, and then use the morphological reconstruction to filter the bright and dark area of the image. In the end, the reconstructed ferrographic image segmentation based on morphological watershed technique. Through this method, we effectively avoided the over-segmentation and realized automatic detection of the objective area and divided the attached and overlapped particles, which largely improved the accuracy of image analysis. Moreover, because we use morphological method to process image, the speed was increased, which satisfied the real time requirements in the practical application. In this paper, the algorithm is realized C++ language, and the effectiveness and accuracy of this application were proved by experiment results.