Deep neural networks based error level analysis for lossless image compression based forgery detection.

Sri, Chintakrindi Geaya; Bano, Shahana; Deepika, Tinnavalli; Kola, Nehanth; Pranathi, Yerramreddy Lakshmi

Abstract

The proposed model is implemented in deep learning based on counterfeit feature extraction and Error Level Analysis (ELA) techniques. Error level analysis is used to improve the efficiency of distinguishing copy-move images produced by Deep Fake from the real ones. Error Level Analysis is used on images in-depth for identifying whether the photograph has long passed through changing. This Model uses CNN on the dataset of images for training and to test the dataset for identifying the forged image. Convolution neural network (CNN) can extract the counterfeit attribute and detect if images are false. In the proposed approach after the tests were carried out, it is displayed with the pie chart representation based on percentage the image is detected. It also detects different image compression ratios using the ELA process. The results of the assessments display the effectiveness of the proposed method.

Citation

SRI, C.G., BANO, S., DEEPIKA, T., KOLA, N. and PRANATHI, Y.L. 2021. Deep neural networks based error level analysis for lossless image compression based forgery detection. In Proceedings of the 2021 International conference on intelligent technologies (CONIT 2021), 25-27 June 2021, Hubli, India. Piscataway: IEEE [online]. Available from: https://doi.org/10.1109/CONIT51480.2021.9498357

• Conference Name: 2021 International conference on intelligent technologies (CONIT 2021)
• Conference Location: Hubli, India
• Start Date: Jun 25, 2021
• End Date: Jun 27, 2021
• Acceptance Date: Apr 30, 2021
• Online Publication Date: Aug 4, 2021
• Publication Date: Dec 31, 2021
• Deposit Date: Sep 19, 2023
• Publicly Available Date: Sep 19, 2023
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• ISBN: 9781728185842
• DOI: https://doi.org/10.1109/CONIT51480.2021.9498357
• Keywords: Error-level analysis; Convolution neural networks; Deep learning
• Public URL: https://rgu-repository.worktribe.com/output/2063996