摘要
阈值的选取对图像分割后的效果起着至关重要的作用,本文针对图像分割过程中阈值选取的问题,提出了一种基于PSO优化的改进OTSU图像分割算法。该算法以最大类间方差作为PSO算法适应度函数,以当前分割阈值组合作为粒子的当前位置,阈值更新速度作为粒子的当前速度。通过迭代计算更新粒子位置和速度,最后确定图像分割的最佳阈值。与传统OTSU图像分割算法及基本遗传算法图像分割算法相比,该算法稳定性更好,算法效率更高。
The selection of threshold plays an important role in image segmentation. Based on PSO 'algorithm, an improved method for selecting the optimal threshold in image segmentation is proposed in this paper. In this method, the maximum between -cluster variance is used as the fitness function of PSO algorithm. The combination of threshold are used as the current position of the particles. And the updating veloeity of the threshold is used as the current velocity of the particles. The optimum threshold is made through updating the position and velocity of particles. Compared with the traditional OTSU image segmentation algorithm and the image segmentation method base on basic genetic algorithm, the algorithm introduced in this paper has higher stability and computational efficiency.
出处
《微计算机应用》
2009年第12期13-17,共5页
Microcomputer Applications