A Novel Approach for Human Silhouette Extraction from Video Data
In this paper we propose a method for efficient extraction of human silhouette from video sequences. The proposed approach includes background elimination, edge detection, region filling and noise removal using morphological operations to estimate the sil
- PDF / 298,618 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 50 Downloads / 215 Views
Abstract In this paper we propose a method for efficient extraction of human silhouette from video sequences. The proposed approach includes background elimination, edge detection, region filling and noise removal using morphological operations to estimate the silhouette of an image. To the best of our knowledge our proposed approach for silhouette extraction involving background elimination and edge detection is first of its kind. We have applied our proposed technique on Weizmann (standard) dataset and compared the results with the most recent related research work. The comparison results in terms of statistical measures like precision, recall and F-measure clearly show the supremacy of our method and thus justify its novelty.
⋅
⋅
Keywords Human silhouette Foreground detection Edge detection ground subtraction Morphological operation Region filling
⋅
⋅
⋅
Back-
A. Raychaudhuri (✉) Department of Computer Science & Engineering, B. P. Poddar Institute of Management & Technology, 137, VIP Road, Kolkata-52, India e-mail: [email protected] S. Maity ⋅ A. Chakrabarti A. K. Choudhury School of Information Technology, University of Calcutta, JD-2, JD Block, Sector-III, Kolkata-98, India e-mail: [email protected] A. Chakrabarti e-mail: [email protected] D. Bhattacharjee Department of Computer Science and Engineering, Jadavpur University, 188, Raja S. C. Mallick Road, Kolkata-32, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2017 R. Chaki et al. (eds.), Advanced Computing and Systems for Security, Advances in Intelligent Systems and Computing 568, DOI 10.1007/978-981-10-3391-9_8
125
126
A. Raychaudhuri et al.
1 Introduction Human silhouette extraction from video is an important step towards shape based analysis for many human based video applications. Some of the video applications like CCTV camera for indoor surveillance, elderly monitoring, and office floor surveillance etc., where camera is fixed in certain position, extraction of background information is an important step. Our proposed methodology eliminates the background efficiently to accurately extract human silhouettes as foreground objects. Sometimes the color of the foreground is closely matched with some portion of the background, that results distorted foreground detection. Silhouette extraction is an important step for segmenting a human body from a background, which can be used to track the person in video, recognizing human actions or gait based biometric recognition. In the process of human silhouette extraction, background modeling is one of the important tasks. Background subtraction has been an active area of research, and it is a commonly used approach for detecting as well as tracking a person in videos from a fixed camera. An efficient thresholding mechanism is needed after eliminating the background from the current frame to get the noise free foreground. Background subtraction methods have been used extensively to assist in human behavior analysis such as human action recognition and human gait recognition [1–7],
Data Loading...