基于遗传算法的自适应聚类图像阈值分割方法

An Adaptive Method of Clustering Image Threshold Segmentation Based on Genetic Algorithms

文中针对复杂背景条件下的红外图像分割问题,将遗传算法引入最大类间方差法中,同时结合人类视觉感知原理,探讨了一种新的多阈值图像分割方法即基于遗传算法的图像阈值分割方法,该算法引入了一个自动判别且时空可变的目标背景条件和调整最佳分割区域的步骤,提高了分割算法的质量及鲁棒性,克服了传统阈值方法在图像分割中的局限性。通过计算机仿真实验,验证了该方法的有效性。

The paper deals with image segmentation in complex background.Genetic Algorithm is introduced to Otsu method.Together with human visual apperception theory,a new multi-threshold image segmentation,i.e.an method of image threshold segmentation based on Genetic Algorithms,is discussed here.The Method of image segmentation Based on Genetic Algorithms is able to distinguish changeable target backgrounds automatically and regulate the best segmenting region.This brings much more accurate result of segmentation and robustness,which helps overcome drawbacks of traditional threshold methods.Computer emulation experiment shows that the performance of the algorithm provided above are satisfactory.