An Approach of Data Hiding in Video Steganography using Object Detection
Mukesh Dalal1, Manisha Singh2, Arun Kumar3, Charu4, Mamta Juneja5
1Mukesh Dalal, UIET, Panjab University, Chandigarh, India.
2Manisha Singh, UIET, Panjab University, Chandigarh, India.
3Arun Kumar, UIET, Panjab University, Chandigarh, India.
4Charu, UIET, Panjab University, Chandigarh, India.
5Mamta Juneja, UIET, Panjab University, Chandigarh, India
Manuscript received on 18 June 2019 | Revised Manuscript received on 25 June 2019 | Manuscript published on 30 June 2019 | PP: 2460-2466 | Volume-8 Issue-5, June 2019 | Retrieval Number: E7715068519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In today’s digital world, the growth of internet and communication technology has significantly increased the transmission of data in large amount. As a result, the secure transmission of data has become a pre-requisite. Generally, for video steganography whole frame/frames are utilized to conceal secret data which usually causes visual distortion. To overcome this problem, the paper presents a new video steganography approach based on moving object detection and tracking where the secret data is embedded in LSB (a least significant bit) of moving objects. Background subtraction using GMM (Gaussian mixture model) is used for detection in objects that possess motion. Albeit the embedding in moving objects is cumbersome, however, pixels selection of moving objects in videos provides more security than embedding in each frame considering as an individual image. Also, moving objects causes less distortion after embedding making it hard for the intruder to detect, ensuring the imperceptibility of the stego-video. Eventually, the proposed technique is compared with LSB embedding exclusively in frames of the video, and the results illustrated that the proposed scheme performed better than the simple LSB embedding technique.
Keywords: Background Subtraction, LSB, Object Detection, Video Steganography.
Scope of the Article: Image Processing