An improved image processing method for particle measurement

Abstract

For the particle measurement under the condition of uneven illumination, defocus blur as well as consisting of touching particles, an effective analysis method was proposed in this paper. Firstly, through analyzing and determining the optimal imaging plane of particles, the level of defocus blur can be reduced, and then the influence of uneven illumination can also be eliminated by using an adaptive thresholding method, finally, with the aid of image reconstruction method to modify the Euclidean distance transform result of binary image and combined with the watershed transform, touching particles can be efficiently segmented, besides, which can also help to reduce the error rate during the segments of peanut-shaped particles. According to the experimental analysis, this proposed method can effectively reduce the analytical error of particle characteristics when compared with the conventional method.