An improved floc image segmentation algorithm based on Otsu and particle swarm optimisation

Abstract

Image segmentation algorithm research is of great significance in the process of flocs detection. The paper proposes an improved floc image segmentation algorithm based on particle swarm optimisation (PSO) and Otsu, which takes into account both the motion characteristics of flocs and the real-time requirements of water treatment. Our research process goes as follows. Firstly, grey stretch technique is used to enhance the contrast between the flocs and the background. Then, the segmentation threshold is obtained by using the adaption characteristics of PSO. Finally, our algorithm uses the morphological filtering including the opening and closing operation to handle the segmented flocs image. The purpose is to remove the edge fuzzy zone. Experiments show that the algorithm realises flocs image segmentation accurately and rapidly, which greatly simplifies the calculation of equivalent size and quantity of flocs.