| Title | Image Forensics using Optimal Normalization in Challenging Environment |
| Publication Type | Conference Paper |
| Year of Publication | 2021 |
| Authors | Agarwal, Saurabh, Jung, Ki-Hyun |
| Conference Name | 2021 International Conference on Electronics, Information, and Communication (ICEIC) |
| Date Published | jan |
| Keywords | Co-occurrence, digital images, feature extraction, Human Behavior, Image coding, image filtering, Image forensics, Image operations, Metrics, pubcrawl, resilience, Resiliency, Scalability, Support vector machines, Tools, Transform coding |
| Abstract | Digital images are becoming the backbone of the social platform. To day of life of the people, the high impact of the images has raised the concern of its authenticity. Image forensics need to be done to assure the authenticity. In this paper, a novel technique is proposed for digital image forensics. The proposed technique is applied for detection of median, averaging and Gaussian filtering in the images. In the proposed method, a first image is normalized using optimal range to obtain a better statistical information. Further, difference arrays are calculated on the normalized array and a proposed thresholding is applied on the normalized arrays. In the last, co-occurrence features are extracted from the thresholding difference arrays. In experimental analysis, significant performance gain is achieved. The detection capability of the proposed method remains upstanding on small size images even with low quality JPEG compression. |
| DOI | 10.1109/ICEIC51217.2021.9369794 |
| Citation Key | agarwal_image_2021 |
Image Forensics using Optimal Normalization in Challenging Environment