Research and Implementation of Digital Image Threshold Segmentation Based on GA-Otsu
In order to improve the global search capability of image segmentation and the accuracy of optimal threshold selection,a digital image threshold segmentation optimization algorithm GA-Otsu is proposed based on the maximum inter class variance method(Otsu)and genetic algorithm(GA).Combining genetic algorithm and Otsu,utilizing the strong global search ability of genetic algorithm,a series of genetic operations can quickly approach the optimal threshold for image segmentation.Using Lena,Cameraman,Peppers,Airplane,Scene and Tree as objects,the image segmentation performance,noise resistance,and timeliness of Otsu algorithm and GA-Otsu algorithm are compared and analyzed.The results show that the GA-Otsu image segmentation method can effectively shorten the segmentation time for digital images while ensuring image segmentation quality,and improve the segmentation limitations of the Otsu algorithm,split time less than 0.07 seconds.
Genetic algorithmThreshold segmentationMaximum inter class variance algorithmImprovement
NETL
NSTL
万方数据
