An improved emperor penguin optimization based multilevel thresholding for color image segmentation

This paper proposes a multi-threshold image segmentation method based on improved emperor penguin optimization (EPO). The calculation complexity of multi-thresholds increases with the increase of the number of thresholds. To overcome this problem, the EPO is used to find the optimal multilevel threshold values for color images. Then, the Gaussian mutation, the Levy flight and the opposition-based learning are employed to increase the search ability of EPO algorithm and balance the exploitation and exploration. The IEPO algorithm optimizes the Kapur’s multi-threshold method to conduct experiments on Berkeley images, Satellite images and plant canopy images. As the experimental results show, the IEPO is the effective method for color image segmentation and have higher segmentation accuracy and less CPU time.