Image segmentation based on the 2-D maximum entropy value and improved genetic algorithm

Image segmentation, extracting characteristics target from the image for user´s requirements, the optimum threshold selection of image segmentation is the key technique. Traditional 2-d maximum entropy image segmentation algorithms use exhaustive way to find the optimal threshold, which is time-consuming, low efficient, and easy to generate the false division. In order to improve the accuracy and efficiency of image segmentation, this paper puts forward a genetic algorithm of 2- d maximum entropy value for image segmentation and makes some improvements in genetic algorithms coding, crossover operator, and mutation operator. Simulation experiments have proved that the new algorithm can greatly shorten the time for optimization, enhance the anti-noise capability in the segmentation process, and improve the efficiency of image segmentation.