不同噪声下超像素FCM图像分割算法优化

FCM Image Segmentation Algorithm Optimization for Super-Pixel under Different Noises

图像分割经典模糊C均值(FCM)算法需要较长的运行时间。为了提高图像的分割处理精度,利用超像素方法建立了FCM聚类快速分割图像算法。通过分水岭分割算法对其进行优化处理获得快速FCM(FFCM)算法,并进行了详细的试验分析。研究结果表明:相比较多尺度分水岭超像素(MGR-WT)、稳健边缘超像素(RE-SH)方法,FFCM表现出了更优的性能。设置噪声后,FFCM算法发生了分割效果下降的现象。对于高斯与椒盐噪声,采用FFCM算法表现出了优异的概率指数(PRI)、信息误差(VOI)和全局一致性误差(GCE)性能指标,具有更稳定的分割性能。该研究对提高图像分割效果具有很好的实际应用价值。

The classical fuzzy C-means(FCM)algorithm for image segmentation needs a long running time.In order to improve the accuracy of image segmentation,a fast image segmentation algorithm based on fuzzy C-means clustering is established by using super-pixel method.The fast FCM(FFCM)algorithm is optimized by watershed segmentation algorithm and analyzed in detail.The results show that FFCM has better performance than morphological gradient reconstruction-watershed algorithm(MGR-WT)and robust edge super-pixel hierarchy(RE-SH)methods.After setting the noise,the segmentation effect of FFCM algorithm decreases.For Gaussian noise and salt-and-pepper noise,FFCM algorithm shows excellent probabilistic rand index(PRI),variation of information(VOI)and global consistency error(GCE)performance index,and has more stable segmentation performance.The research has a good practical value for improving the image segmentation effect.