Underwater image segmentation based on improved PSO and fuzzy entropy
Yongjie Pang · Ocean Engineering · 2010
Under the influence of lighting condition and water quality,the underwater images have low contrast,uneven gray scales and fuzzy edge of objects.Though tradition threshold methods based on maximum entropy principle could divide the image into objects and background,its complex and time-consuming computation by flange-fuzzy membership function and method of exhaustion is often an obstacle.In this paper,based on the traditional threshold methods,fuzzy entropy is redefined on given images,and the improved PSO is used to search the optimal threshold based on maximum entropy principle.The experiments prove that this novel approach is effective for simple back-ground underwater images.Comparing with the traditional methods,the new method shows better adaptability and noise restraining performance.