Computer Engineering and Applications ›› 2013, Vol. 49 ›› Issue (3): 230-233.

Previous Articles     Next Articles

Segmentation FCM algorithm combined hue histogram added window

LIU Jiawen1, LIANG Guangming2, LIU Renren1, XIE Yating1, DING Jianwen3   

  1. 1.School of Information Engineering, Xiangtan University, Xiangtan, Hunan 411105, China
    2.College of Electronic Science and Engineering, National University of Defense Technology, Changsha 410073, China
    3.Ave Science and Technology, Changsha 410013, China
  • Online:2013-02-01 Published:2013-02-18

融合加窗色调直方图的彩色图像FCM分割算法

柳佳雯1,梁光明2,刘任任1,谢雅婷1,丁建文3   

  1. 1.湘潭大学 信息工程学院,湖南 湘潭 411105
    2.国防科技大学 电子科学与工程学院,长沙 410073
    3.长沙爱威科技,长沙 410013

Abstract: This paper proposes a fast FCM algorithm (MHFCM) combined hue histogram added window, to direct at conventional FCM algorithm which is characterized by the great amount of calculation, inferior resistance to noise. Hue histogram with added window is used for getting background hue from color images. Efficacious information without background information is used for clustering on FCM. Experiment results show that the MHFCM algorithm enhances segmentation speed and overcomes the effects of background noise on segmentation results to segment more precisely.

Key words: color image segmentation, hue histogram added window, Fuzzy C-Means(FCM) algorithm

摘要: 针对传统FCM算法计算量大、对噪声敏感的缺点,提出了一种融合加窗色调直方图的快速FCM算法(MHFCM)。对彩色图像的色调直方图加窗得到背景色调,对去除背景信息后的有效信息进行FCM聚类分割。实验结果表明,MHFCM算法提高了图像的分割速度,并且克服了图像中背景噪声对分割效果的影响,使分割更加准确。

关键词: 彩色图像分割, 加窗色调直方图, 模糊C均值聚类算法(FCM)