Image Tamper Detection and Correction Based on Mean Pixel Value and Logistic Map

• Main Authors: Gottimukkala, Anantha Rao, Pradhan, Anita, Kosuru, S.N.V.J. Devi, Swain, Gandharba
• Format: Conference Proceeding
• Language: English
• Subjects: Bars; block average; Communication systems; data hiding; image security; logistic map; Logistics; Measurement; Receivers; tamper detection; Watermarking
• ISSN: 2575-7288
• DOI: 10.1109/ICACCS57279.2023.10112748

This proposed image watermarking scheme works on 4×4 pixel blocks. The least significant bits (LSBs) of all the 16 pixels are set to zeroes. After setting all the LSBs to zeroes, the mean pixel value of the block is calculated. It is expressed in 8 bits. A logistic map (LM) is used to create 8-bit key. The 8-bit mean value and 8-bit key are input into exclusive or (XOR) function to generate watermark bits (WBs). The block is divided into 2 sub-blocks. The first 8 pixels in upper 2 rows of the block are known as upper half block (UHB) and the next 8 pixels in lower 2 rows of the block are known as lower half block (LHB). The 8 bits of block average are concealed in LSBs of UHB pixels, and the 8 WBs are concealed in LSBs of LHB pixels. At the receiver end the 8 bits from the pixels of UHB are extracted and XORed with the 8-bit key generated from LM. These are known as test bits. The LSBs of pixels of LHB are extracted, these are the extracted WBs. If the extracted WBs are same as the test bits, the block is earmarked as not tampered. The quality metrics values are satisfactory and better than the existing techniques.