Detection of image forgery for forensic analytics.

Sri, Chintakrindi Geaya; Bano, Shahana; Trinadh, Vempati Biswas; Valluri, Venkata Viswanath; Thumati, Hampi

Abstract

Due to the technical revolution in digital image processing, different advanced image manipulation software has been used in recent years to produce new unrealistic images without leaving evidence of what is happening in the world, so it would be difficult to detect tampering visually. Digital image forgeries have many techniques, but it is still very difficult to identify copy-move forgery. Therefore, we use a robust algorithm in this paper to detect copy-move forgery based on the descriptor speed-up robust feature (SURF) as a key-point detection, high-pass filtering as a matching feature, nearest neighbor used as a clustering algorithm to divide the entire image. By swapping the matched feature points with the corresponding super pixel blocks, the doubtful regions are identified, and then, the corresponding blocks are combined on the basis of similar local color features (LCF). Finally, to obtain the suspected forged areas, morphological close operation was applied. The results of the study indicate that the proposed method achieves considerable output based on key-point detection compared to other forgery detection methods used in the current method in order to address the research challenges.

Citation

SRI, C.G., BANO, S., TRINADH, V.B., VALLURI, V.V. and THUMATI, H. 2022. Detection of image forgery for forensic analytics. In Aurelia, S., Hiremath, S.S., Subramanian, K. and Biswas, S.K. (eds.) Select proceedings of the 2021 International conference on sustainable advanced computing (ICSAC 2021), 5-6 March 2021, Bangalore, India. Lecture notes in electrical engineering, 840. Singapore: Springer [online], pages 321-338. Available from: https://doi.org/10.1007/978-981-16-9012-9_26

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2021 International conference on sustainable advanced computing (ICSAC 2021)
• Start Date: Mar 5, 2021
• End Date: Mar 6, 2021
• Acceptance Date: Feb 14, 2021
• Online Publication Date: Mar 31, 2022
• Publication Date: Dec 31, 2022
• Deposit Date: Jul 4, 2024
• Publicly Available Date: Jul 4, 2024
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Pages: 321-338
• Series Title: Lecture notes in electrical engineering
• Series Number: 840
• Series ISSN: 1876-1100; 1876-1119
• ISBN: 9789811690112
• DOI: https://doi.org/10.1007/978-981-16-9012-9_26
• Keywords: Forgery detection; Image processing; Digital forensics; Copy-move forgery; SURF descriptor; High pass filters; Local colour features
• Public URL: https://rgu-repository.worktribe.com/output/2063941