Building a Statistical Model to Detect Foreground Objects and using it in Video Steganography

Autor: Mithal Hadi Jebur, Fanar Ali Joda, Mohammed Abdullah Naser
Jazyk: Arabic<br />English
Rok vydání: 2023
Předmět:
Zdroj: Baghdad Science Journal, Vol 20, Iss 6 (2023)
Druh dokumentu: article
ISSN: 2078-8665
2411-7986
DOI: 10.21123/bsj.2023.7926
Popis: Video steganography has become a popular option for protecting secret data from hacking attempts and common attacks on the internet. However, when the whole video frame(s) are used to embed secret data, this may lead to visual distortion. This work is an attempt to hide sensitive secret image inside the moving objects in a video based on separating the object from the background of the frame, selecting and arranging them according to object's size for embedding secret image. The XOR technique is used with reverse bits between the secret image bits and the detected moving object bits for embedding. The proposed method provides more security and imperceptibility as the moving objects are used for embedding, so it is difficult to notice the changes in the moving objects instead of using background area for embedding in the video. The experimental results showed the better visual quality of the stego video with PSNR values exceeding 58 dB, this indicates that the proposed method works without causing much distortion in the original video and transmitted secret message.
Databáze: Directory of Open Access Journals