• Kesavan K
• Singaravel G
• Deepthi R
• Indhu Priya S V

Digital images are quickly taking over as the main way that content is shared on social media. Adware programs have the ability to mimic similar graphics to briefly convey inaccurate information. Therefore, it's critical to recognize these fakes. The literature has addressed this issue using a variety of digital image fraud detection methods. However, the majority of Some of these methods are limited to identifying specific types of forgeries, including picture splicing or copy-move, which are not used in everyday situations. The research suggests a method to improve the identification of digital picture forgeries by simultaneously detecting two types of image forgeries through transfer learning and deep learning approaches. The suggested method is based on identifying compression quality differences between the surrounding area and the forged region, which is a common sign of tampered digital images. For the purpose of identifying these frauds, a deep learning approach is advised. Studies show that the best identification accuracy (about 97%) may be obtained by using a convolutional neural network model such as MobileNetV2, which also has the advantage of requiring fewer training parameters and shorter training timeframes.

Image Fraud Detection (IFD), Image Compression (IC), Convolutional Neural Network (CNN), Pertained Model.

"DETECTION OF FAKE IMAGES IN SOCIAL MEDIA USING CONVOLUTIONAL NEURAL NETWORKS IN DEEP LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 5, page no.h54-h68, May-2024, Available :http://www.jetir.org/papers/JETIR2405705.pdf

"DETECTION OF FAKE IMAGES IN SOCIAL MEDIA USING CONVOLUTIONAL NEURAL NETWORKS IN DEEP LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 5, page no. pph54-h68, May-2024, Available at : http://www.jetir.org/papers/JETIR2405705.pdf