A two pass labeling algorithm for automatic schistosome egg detection and counting

A method is proposed to detect schistosome eggs in a microscopic environment connected to a computer. The advantage of this method is the use of a simple two pass algorithm. The first pass detects and analyzes the elementary cases of isolated objects. The second pass introduces an enhancement process to improve the classification ratio. Although the noisy environment of images results from the nature of the support, the segmentation process gives satisfactory results using an adaptive statistical thresholding method. The detection is improved by an a priori knowledge of the geometry of the eggs. The method is simple and its portability to a microcomputer environment is envisaged to obtain a transportable machine for mass screening campaigns.< >