摘要
为了保证遗传算法能够尽快收敛到全局最优解,避免早熟现象发生,提出了适应度标定公式,保证适应度函数值总为正值。新的适应度函数能够正确引导群体的发展方向,提高选择压力;提出了相似度概念,保留相似性差的个体,剔除相似性个体。在不增加群体规模的前提下,增加了群体的多样性。为了有效地对图像进行分割,提出基于改进遗传算法的图像分割方法,采用Otsu公式,找出分割图像最优阈值。给出不同改进遗传算法计算实例比较和不同图像分割方法效果图。
In order to get optimal global solution and avoid prematurity a fitness normalization formula is introduced and it always gets a positive value. The new formula can guide the population to a proper direction and increase the press for selection of individuals. The similarity is defined to increase the varieties of individuals without increasing the size of population, thus solving the problem of local optimized solution. In order to solve the problem how to segment an image, an image segmentation method based on improved genetic algorithms is proposed. The method can find out the optimal threshold of the segmentation object by the Otsu formula. Different calculated results are obtained with different improved methods. Different segmented images are given by different segmentation methods.
出处
《数据采集与处理》
CSCD
北大核心
2005年第2期130-134,共5页
Journal of Data Acquisition and Processing
关键词
遗传算法
图像
图像分割
最优阈值
genetic algorithms
image
image segmentation
optimal threshold