Application of Microscopic Image Segmentation Technology in Locust-Control Pesticide Research

The microscopic slice image segmentation of the interacting tissues between locust and bio-pesticide is very important in aspects of illuminating the interactive processes between the locust organs and the bio-pesticide, revealing the infective mechanism of the bio-pesticide to locust, and optimizing the biological agriculture chemical preparation. The classic image segmentation algorithms, such as threshold segmentation, region-growing and edge detection, always result in over-segmentation and edge discontinuity for the microscopic slice images. In this paper, we analyzed the locust soft tissue image's characteristics of complex topology and minimal gray scale difference, exploited the C-V model formulated by level set method to extract the features of image, adjusted the parameters of C-V model and examined their influences in the whole process. The algorithm can assure the obtained contours are not sensitive to the initial contour position, can converge to the sunken part of the contours and realize the adaptive segmentation of biological tissue slice images. The experimental results demonstrate the efficiency of the algorithm, which can satisfy the accuracy of microscopic slice image segmentation.