Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/373627
Title: | Certain Investigations on Medical Image Steganography Based on Pixel Prediction Using Optimized Machine Learning Classifier |
Researcher: | Reshma V. K. |
Guide(s): | R. S. Vinod Kumar |
Keywords: | Computer Science Computer Science Theory and Methods Engineering and Technology |
University: | Noorul Islam Centre for Higher Education |
Completed Date: | 2021 |
Abstract: | The steganography using images is an emerging method for protecting confidential information. Here, the secret message is entrenched in medical image for maintaining confidentiality of the information. Steganography is the process of hiding secret message in image or video, which is the trending area of interest in the research domain at present due to its significance in covert communication. Image steganography is one of the popular and promising techniques used for maintaining the confidentiality of the secret message that is embedded in the image. Even though there are a number of techniques available in the literature, an approach providing accurate result is still a challenge. The existing medical image steganography techniques faced numerous issues while hiding the secret information. These issues are overcome in this research by developing efficient medical image steganography techniques. The three major contributions of research are as devised below. newlineThe first contribution is the development of the pixel prediction method using the Support Vector Neural Network (SVNN) classifier and Discrete Wavelet Transform (DWT). The prediction map is constructed using the SVNN classifier, whereas the DWT is used to extract secret message from medical image. newlineThe second contribution of research is the development of the pixel prediction method using the error dependent SVNN. The SVNN along with features, such as Pixel Coverage, Edge Information, Wavelet Energy, Scattering Features and Texture are used for the identification of the effective pixel. The Moth Search (MS) and the Genetic Algorithm (GA) are utilized to train SVNN, which is truly error dependent. The secret message is extracted from embedded image considering inverse Contourlet Transform (CT). newlineThe third contribution of the research is the development of the pixel prediction method using the error-dependent Chicken-Moth Search Optimization (CMSO) technique-based Deep Convolutional Neural Network (DCNN) classifier. The DCNN classifier along with t |
Pagination: | 6115kb |
URI: | http://hdl.handle.net/10603/373627 |
Appears in Departments: | Department of Computer Science and Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
80_recommendation.pdf | Attached File | 118.67 kB | Adobe PDF | View/Open |
certificate.pdf | 236.68 kB | Adobe PDF | View/Open | |
chapter 1.pdf | 190.73 kB | Adobe PDF | View/Open | |
chapter 2.pdf | 209.53 kB | Adobe PDF | View/Open | |
chapter 3.pdf | 793.29 kB | Adobe PDF | View/Open | |
chapter 4.pdf | 1.3 MB | Adobe PDF | View/Open | |
chapter 5.pdf | 3.77 MB | Adobe PDF | View/Open | |
chapter 6.pdf | 72.94 kB | Adobe PDF | View/Open | |
list of publications based on thesis.pdf | 92.01 kB | Adobe PDF | View/Open | |
references.pdf | 127.32 kB | Adobe PDF | View/Open | |
table of contents.pdf | 206.4 kB | Adobe PDF | View/Open | |
title page.pdf | 128.49 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: