The sperm video segmentation based on dynamic threshold

Abstract

In this paper, the method of threshold segmentation is introduced, which focus on the image characteristics of the sperm video, base on the dynamic threshold and combine with region growing arithmetic. The method is based on the movement characteristics and the brightness characteristics of the objective sperm to distinct, and then uses the region growing algorithm to calculate the sperm region, finally according to this gray area to determine the threshold. The results show that this method has better performance to divide the sperm goal.