提出了一种阈值分割方法。本算法将模糊集理论和遗传算法有机地结合起来 ,它是一种有监督的分割方法 ,首先确定图像阈值的数目 ,然后将灰度图像模糊化 ,确定准则函数 ,采用遗传算法求得最优解 ,最后经过后处理得到最终分割结果。实验表明该方法运算速度快 ,且对噪声不敏感 ,具有较高的鲁棒性。 A threshold segmentation method is proposed. This algorithm combines fuzzy set theory and genetic algorithm organically. It is a supervised segmentation method. Firstly, the number of image thresholds is determined. Then the gray image is blurred to determine the criterion function, and the genetic algorithm is used to obtain the optimal Solution, and finally through the post-processing to get the final segmentation results. Experimental results show that this method is fast, sensitive to noise and has high robustness.