Histogram-based Threshold Segmentation of Video Frames using Otsu’s Method

Authors

B. Sathiyaprasad
Department of Computer Science and Engineering, Annamalai University, Tamilnadu
K. Seetharaman
Department of Computer Science and Engineering, Annamalai University, Tamilnadu
B. Satheesh Kumar
Department of Computer Science and Engineering, Annamalai University, Tamilnadu

Synopsis

Video segmentation plays an important role in image analysis and computer vision system. Automatic threshold is mostly used for segmentation because the method is easy to implement, and the time complexity is less when compared to other segmentation methods. In many domains the Otsu’s method is used for image segmentation. In our case, the videos are segregated to number of frames and each frame is converted to grays scale. Foreground and Background are segmented on a gray scale images and compute a threshold value using Histogram-based threshold. This paper proves the Otsu’s threshold gives better performance when compared to the existing methods.

ICICCAS20 Cover Page
Published
July 29, 2020