| Genetic algorithm is a probability search algorithm based on a kind of binary number string that uses some kind of encoding technology to be called chromosome. It has strong global search capability and robustness. So it is widely used in function optimization, combination optimization, machine learning and so on. However, genetic algorithm has the defects of premature convergence and local search ability. The paper focuses on improving the local search ability of the genetic algorithm, and proposes an improved niche genetic algorithm (IVNGA). Through genetic control parameters can be adaptively adjust the crossover probability and mutation probability, the Euclidean distance is used to replace the Hamming distance, the better measure the distance from the individual. The local search ability and the convergence speed of the algorithm are improved by introducing simulated annealing algorithm. Finally, the effectiveness and superiority of the proposed algorithm are verified by the results of six function optimization experiments.Image segmentation is an important part of image processing, and it is an important basic work in the field of image processing. Among them, the Otsu is a kind of maximum inter class variance method, which can be called Otsu image segmentation method, which makes it easier to determine the threshold in the selection process. However, the Otsu image segmentation method is more suitable for the case of a single pole. Any single image segmentation method is unable to obtain satisfactory results for all images. So the most recent image segmentation research is the combination of the results. Among them, the image segmentation based on genetic algorithm has great advantages in the efficiency and effectiveness of segmentation. However, using the classical Otsu algorithm and the basic genetic algorithm for image segmentation, there are some problems which are low efficiency, easy to form pseudo solution in advance. In this paper, we propose a new image segmentation method based on Improved Niche Genetic Algorithm (IVNGAMS), which is based on the improved niche genetic algorithm (Otsu), and can be used to control the genetic parameters of the two-dimensional Otsu image segmentation function. Experimental results show that the improved image segmentation method can better improve the global search ability of the algorithm, and improve the ability of genetic algorithm to jump out of local optimal solution, but also more stable and fast convergence to the optimal segmentation threshold, and get better image segmentation results. |
