Article Preview
Top1. Introduction
Nowadays, nowadays, information sharing is now the most required necessity for everybody, and also, to fulfill one such urge, researchers enable the medium whereby information can be carried on. As the importance of a communication channel grows, it becomes vulnerable to a wide range of threats. The information is moved all over nodes inside a computer network composed of data packets of variable or fixed size. In most cases, a few secure methodologies are applied to data just at the application layer, and thus the securely encrypted information is packetized and then sent at lower levels in the OSI architecture. If an attacker obtains all of the packets, he or she would then acquire encoded information by adequately ordering the collected data of all of these packets. After that, efforts have been made to shatter a sender's secure method. If it is possible to avoid data transfer about the structure of the data inside the data packet during the transfer of data, an attacker would not understand the nature of the data being transmitted or even the ordering of information from different packets. As a result, the primary task of protecting the transmission path from all types of threats and securely transferring information to the receiver is met. That's what the suggested methodology based on Morphological Component Analysis and steganography should accomplish.
Morphological component analysis is a process that helps us to segregate features from the image that have different morphological facets. Morphological component analysis can be used for image painting as well as image segregation and is very useful for breaking down images into texture and smooth components (Aujol et al., 2003). The Morphological component analysis accompanied by the total variation regularization is a really effective method for separating an image into piecewise smooth content but also texture. Use of the curvelet dictionary throughout morphological component analysis causes ringing artefacts inside the piecewise smooth content part. To erase ringing artefacts from the piecewise smooth part, a total variation regularization scheme was used. The cartoon portion of the image is transformed using the daubechies wavelet transform, which is supported by thresholding of a coefficient, and the image is recreated at the receiving end using the inverse daubechies wavelet transform. Wavelets have become commonly utilized not just by computer scientists in functional and computational research, as well as by researchers in sciences such as astronomy, chemistry, and biology, and also in applied disciplines such as computer science, technology, and microeconomics. Wavelet implementations have recently been used by researchers in fields like signal voice and speech recognition.
The usage of wavelets for signal processing in particular, including image recognition and dimensionality reduction, has proven to be of great benefit. Ogden provides a detailed survey of wavelet applications in statistics. Mostly, during collection or transmission of an image, distortion frequently corrupts it. The aim is to eliminate any distortion while preserving the image's essential features. Wavelet reduction through a non-linear approach suggested by Donoho and John Stone (1994, 1995) is perhaps the most widely employed technique for distortion removal. This method has been commonly used throughout statistics, especially in signal processing and image processing.
In a mathematical sense, this would be known as estimating a real curve through data polluted with interference, which is typically assumed to really be Gaussian noise. Three measures are involved in estimating the actual curve. Apply a Discrete Wavelet Transform (DWT) that converts spatial information from the time domain to the time-frequency domain. The coefficients are indeed the values of the data sets in the time-frequency domain. Disturbance occupied the coefficients of tiny actual values, whereas the coefficients of high absolute values produced further data knowledge than disturbance. If a wavelet coefficient does not reach a certain threshold stage, it is reduced to 0 (hard threshold rule) or shortened (soft threshold rule) as in the second stage. The final move would be to use the Reverse Discrete Wavelet Transform and recreate that signal from the resulting coefficient (IDWT). There's been a lot of research about determining the threshold meaning since the study of "Donoho and Johnstone (1994, 1995)."