WCSE

Semi-Automatic Moving Objects Segmentation and Tracking Base on Background Subtraction Using Fuzzy C-Means

Abstract— In this paper we present a semi-automatic technique for moving objects segmentation and tracking. Clustering as a segmentation method has been used for many applications. The problem of clustering in many segmentation methods is a need of high performance and low computational cost. We proposed the Fuzzy C-Means method for clustering moving objects. To evaluate the performance, we compare FCM against K-Means and SOM algorithm. Semi-automatic used by human for create image ground truth for measure performance by MSE and PSNR. Based on experiment the MSE of Fuzzy C-Means is lower than K-Means and SOM. Also PSNR of FCM is higher than K-Means and SOM. The result proved that Fuzzy C-Means is promising to cluster pixels in moving objects segmentation.

Index Terms— semi-automatic, fuzzy c-means, moving object segmentation

Moch Arief Soeleman, Mauridhy Hery Purnomo, Mochamad Hariadi
Department of Electrical Engineering Institut Sepuluh Nopember Surabaya, INDONEDIA
Moch Arief Soeleman
Department of Computer Science Dian Nuswantoro University Semarang, INDONESIA
Mochamad Hariadi, Kondo Kunio, Masanori Kakimoto, Mikami Koji
School of Media Science Tokyo University of Technology Tokyo, JAPAN

[Download]

Cite: Moch Arief Soeleman, Mauridhy Hery Purnomo, Mochamad Hariadi, Kondo Kunio, Masanori Kakimoto, Mikami Koji, "Semi-Automatic Moving Objects Segmentation and Tracking Base on Background Subtraction Using Fuzzy C-Means," 2015 The 5th International Workshop on Computer Science and Engineering-Information Processing and Control Engineering (WCSE 2015-IPCE), pp. 487-493, Moscow, Russia, April 15-17, 2015.