Abstract: Maximum between-class variance (Otsu) image segmentation method is a common image threshold segmentation method based on statistical theory, but Otsu image segmentation method has some disadvantages, such as more time-consuming, low segmentation accuracy and false image segmentation. Combining the principles of monkey king genetic algorithms, with Otsu algorithm, image gray, just as optimal threshold, was found. The results show that combined method not only improves the quality of image segmentation but also reduce the computation time. It is very suitable for real-time image processing.