Comparative Analysis of Watermark Reconstruction Using Discrete Wavelet Transform and Slantlet Transform for User Identification in Social Media

The use of social media is common in day to day life. People transfer digital data in the form of images, texts or videos to convey the news/information to others. Sometimes, people unintentionally forward the wrong/sensitive information over social media, which can cause negative effects in society. Most of the times this information message is only forwarded, which is being shared by others but the source of that information is the specific user. It is important to find such users, who start spreading this type of digital information over the internet. Such users should be identified and stopped them from doing such activities. A comparative analysis is proposed using two different transforms namely Slantlet transform (SLT) and discrete wavelet transform (DWT) in watermark reconstruction for user identification over the internet. In the proposed robust image watermarking, a 10-digit number is considered and each digit is assigned to a 32 bit randomized Walsh code and 8 additional digits are reserved as padding to obtain the binary watermark. The watermark insertion and extraction process are performed using two different transforms to compare their performance in terms of different watermarking parameters. Although both transforms provide significant results against many signal processing attacks, the experimental evaluation proved that the Slantlet transform gives better results as compared to the DWT.