Hiding information by using Discrete Laguerre Wavelet Transform with new algorithms

In this work, the focus was on embedding images and hiding image information in another image by mixing the two images analyzes using the new Discrete Laguerre Wavelets Transform (DLWT) by Approximate and Details coefficients. This method is used for the first time in this field to embedding two images and hide submerged image features. The results were compared with other waveforms indicating that the new wavelet transform proved efficient during the application of two simple examples, including immersing two mysterious images, the other example being the mask and the face.


Asma Abdulelah Abdulrahman 1 , Fouad Shaker Tahir Al-azawi 2
A theoretical presentation of the theoretical framework in applications has been presented quickly 2D(DLWT) by clarifying the concepts of decomposition, approximation and detail of the image, which plays an important role in edge detection where its role is evident in the ability of the new wavelet to analyze local gray gradients and gradients through which edges can be isolated [1],[3], [9], [10].

Preliminaries
In this section all important information will benefit the recipient so that he can understand this paper

Discrete Laguerre wavelet transform
The construction of the new wave for use in this work has been derived from multiple borders and converted into intermittent waves because they have certain qualities that qualify them to be able to use in the image processing of these qualities they belong to space all square functions in  (2)

Activation of transfers in the Matlab program
It is natural that the Matlab program is equipped with many waves such as Haar, Daubechies, Symlets, Coiflets, etc. In order for our built-in guidance to be built through our previous work, we must create a program to allow the new wave to take its place with the waves supplied by the program.

The proposed theory in image processing
In our previous work the Multi Resolution Analysis (MRA) of   R L 2 was defined to Discrete Laguerre Wavelets Transform (DLWT), where it becomes clear to us that it is a family of decreasing sub-distances that is related to approximation and expansion of were defined in [4], [5], [6].

Multi Resolution Analysis (MRA)
The following points illustrate the process MRA in 1D the approximation space D j V 1 and the detail space From above we understand in 2D the two sort is find

Asma Abdulelah Abdulrahman 1 , Fouad Shaker Tahir Al-azawi 2
2-Reconstructed of approximation and details to projections on the same spaces.
The image is a matrix of integers as it is known to be three matrices RGB.
If signal is S then where S is the sum of orthogonal signals implies in 2D , , indicate respectively horizontal, vertical, diagonal details.
Using DLWT, will analyze the original image at the first level To implement the image immersion algorithm using the proposed theory It is completed in the following steps: Step 1: Load two original images for example load mask; X1 = X; load bust; X2 = X; Step 2: Embedding the two images from wavelet decompositions at level 2 using DLWT of two original images with approximations coefficients and details Step 3: Plot original and synthesized images. The above strategies are used to immerse and mix the two analyzes, after which many images were obtained as in Fig. 6

The results
1. Transactions (A,B) are pairs related to approximations and details.

A global strategy has been applied for the selection of original
transactions Through the previous analysis we conclude.
Let's L be the matrix of coefficients for decomposition, A, B are the matrices treated to decompose the two images In the matrix L we will get the   If linear group weights are adjusted we get a variety of merged images as shown in Fig.7

Example2
Merge of fuzzy images, the best example of this technique is the image of How this process?
The choice of a linear array is an example of a medium or a maximum.
To illustrate this strategy, for example, the maximum level 2, we will choose the approximate coefficients and the details, which are important so that the images are complementary, we will choose the type of local analysis.
The built-in image is invers conversion of IDLWT.
We get the new image after the merger and from the first look we find that the result is acceptable where the picture is free of ambiguous areas and any other defects or exist but are difficult to identify.

Conclusion
In this work, the process of immersion of images within each other was used conversion DLWT, which proved its efficiency for other transfers and through the above examples where they were used using other conversions and can be compared through Fig. 9. Fig.9: Shows the types of mergers for coefficients by using db wavelet In the above figure a different transformation was used and we can compare it with the wave used in the two examples above, the resulting images are using DLWT better than using db.