An Improved 2-D Maximum Entropy Threshold Segmentation Method Based on PSO

An improved two-dimensional maximum entropy threshold segmentation method optimized by PSO is proposed. Two-dimensional maximum entropy segmentation method has a better segmentation effect because it not only reflects the gray distribution information of image pixels, but also reflects the related information of neighborhood space. However, it has a poor anti-noise ability, and it may result in over-segmentation. Aiming at this situation, an improved segmentation threshold decision function is proposed, and the particle swarm optimization is used to optimize the choice of threshold in order to further improve the accuracy of threshold selection. Experimental results show that the method is effective to image segmentation, and the speed of segmentation is improved and it also conquers over-segmentation brought by the method of literature.