Noble Method for Data Hiding using Steganography Discrete Wavelet Transformation and Cryptography Triple Data Encryption Standard: DES

Noble method for data hiding using steganography Discrete Wavelet Transformation: DWT and cryptography triple Data Encryption Standard: DES is proposed. In the current era, information technology has become inseparable from human life, especially regarding the processing and dissemination of information. In line with advances in information technology, there are also parties who want to abuse such information by changing information or even damage it. To avoid that happening, then the data needs to be secured first into other media using the DWT method. The choosing of this method is because the image of the data insertion almost resembles the original image. Triple DES methods are also required to encode data and provide additional security so that hidden data will be difficult to solve. The choosing of this method is because it is resistant against brute force, chosen plaintext, and known plaintext attack. Based on the test, image insertion results in 100% immune to the image manipulation of brightness and contrast, but not so resistant to cropped, resized, and rotated image manipulation. Other tests also indicate that the data which is in the picture can be extracted again and will not undergo any changes. Keywords—Data hiding; steganography; DWT; cryptography; 3DES


I. INTRODUCTION
As the development of the era, the development of technology is also overgrowing, especially information technology.Information technology has become inseparable from human life and helps people in many ways, such as processing and disseminating information.In line with advances in information technology, there are also parties who want to abuse such information by changing information or even damage it.
Data in 2016 showed 1061 reports of cyber-attacks with cybercrime category of 77.4% and 950 reports in 2017 with cybercrime category of 72.1% [1].By looking at the data, we can know that cybercrime is a serious problem that we must handle and every year the number of cases of cybercrime will remain substantial.It indeed can be a nightmare for companies or even government agencies who want to send data containing confidential information to clients or other companies.The data needs to be secured first using cryptographic and steganographic methods before being sent to avoid that happening.
Data Hiding Based on Wavelet Multi-resolution Analysis is proposed [2] together with Data Hiding Based on Multiresolution Analysis Utilizing Information Content Concentrations using Eigen Value decomposition [3].Other than these, Information Hiding Method Based on Coordinate Conversion is proposed [4].Meanwhile, Data hiding based on Multi-Resolution Analysis: MRA taking into account scanning of the embedded image for improvement of invisibility is proposed [5].
Furthermore, Improvement of secret image invisibility in circulation image with Dyadic wavelet-based data hiding with run-length coding is also suggested [6] together with Method for data hiding based on Legal 5/2 (Cohen-DaubechiesFeauveau: CDF 5/3) wavelet with data compression and random scanning of secret imagery data [7] Moreover, Data Hiding Method Replacing LSB of Hidden Portion for Secrete Image with Run-Length Coded Image is proposed [8].
Steganography is a technique to hide messages or information into other media.Therefore, besides the intended person, they will not be aware of any messages or data in that other media and prevent the occurrence of suspicion.
One of the methods in steganography is the DWT method.DWT has advantages over other steganographic methods that is steganographic images almost resemble the original picture [9].The message needs to be encoded in advance into another form that can not be understood using a technique called cryptography to provide additional security for the method.
Cryptography is a technique to avoid information being known by unwanted parties and to convert it into an incomprehensible form.The primary purpose of cryptography is to protect data from unauthorized people [10].
In cryptography, there are many methods, and one of them is the Triple DES method (it is referred to 3DES hereafter)1 operates on 64-bit blocks and uses three keys, each of which is 56-bit in size [10].Those keys are the strengths of the 3DES method.This method has a reasonably fast processing time and has resistance to attacks such as brute force, chosen-plaintext, and known plaintext [11].
In this research, data security application is made using DWT of steganography method and 3DES cryptography www.ijacsa.thesai.orgmethod to secure data by inserting data into an image before sending the data.
In the next section, the proposed method is described followed by some experiments.Then, the conclusion is described together with some discussions and future works.

II. PROPOSED DATA HIDING METHOD
The proposed data hiding method is based on DWT based steganography and 3DES based cryptography.

A. 3DES
The DES algorithm has proved that highly competent algorithms can be considered uncomfortable and unreliable.Therefore, there is a search for a method for using it again by making it stronger and more secure than creating a new algorithm starting from the scratch.Two significant improvements result in double DES and triple DES algorithms (3DES).Double DES repeats the DES process twice using two keys.If the experiment to crack a key in a DES is 256, then the research to break into two different keys of n-bit is 22n.However, all of this is not entirely true since the introduction of the concept of a meet-in-the-middle attack [12].
Given the idea that double DES may not be strong enough to prevent meet-in-the-middle attacks has led to the development of 3DES algorithms developed by IBM in 1999 by a team headed by Walter Tuchman [13].This kind of attack is one of the main reasons why double DES is replaced by 3DES that DES operation which is repeated three times using three different keys.It's important to avoid using the same key for encryption, as it will only result in a DES process with slower processing time.3DES has two shapes, the first form uses three completely different keys and the second one uses two completely different keys [12].
According to [14], 3DES has advantages that are fast processing time and a reasonably reliable level of security.Also, 3DES has resistance to several attacks such as brute force, chosen-plaintext, and known plaintext.This method requires three keys that have a 56-bit length per core.The time needed to check all possible keys using 50 million keys per second for each 3DES key is 400 days [15].
The steps in the 3DES process according to [15] shown in Fig. 1 are as follows:  Encrypt a message block (plaintext) using a single DES with a K1 key.
 Decrypt the results obtained from step 1 using the K2 key.
 Encrypt the output from step 2 using the K3 key so that it produces ciphertext.
 The decryption process of ciphertext is the opposite of the 3DES Encryption process, which is decryption using K3 keys, encryption using K2 keys, and decryption using K1 keys.
By looking at the advantages of the 3DES algorithm and considering its weakness as well, in this paper, 3DES method is used by using different keys to provide stronger security on the document.

B. DWT
DWT is a method that can divide information from an image into the approach and signal detail.LL bands include low pass coefficients and procedures to a copy as well as more information of other sub-signals indicating vertical, horizontal, or diagonal information or changes in an image [16].The general equation for DWT can be seen in the equation below [17]. where,.
In the DWT method, there are several techniques for representing images to approach and signal details, one of them is wavelet Haar.Wavelet Haar can be used to describe a picture with a wavelet counting process.The equation of wavelet Haar transform is in the equation shown below [18]: , - The DWT method can represent images into approach and signal details.This method also has advantages compared to other steganography methods, namely the model of steganography results almost resemble the original image [19].Therefore, in this paper, the DWT method is used to hide the message into the approach and details of the signal, so that the changes that occur in the image will not be too visible in the human vision system.Haar-DWT is one of the most basic and straightforward transformations in the DWT family.This method reduces the calculation work.Haar-DWT decomposes each signal into two components.The first component is called the average and the second component is called the difference [10].This process is used to reduce memory requirements and the amount of inefficient Haar coefficient movements.Disadvantages in the sum and subtraction operations can be balanced by decreasing the number of division operations; especially when used at low bit rates, it introduces compression artifacts [19].A detailed procedure of Haar-DWT 2 dimensions according to [14] described as follows : Step 1: First, scan the pixels from left to right with the horizontal direction.Then, perform addition and subtraction operations on neighboring pixels.Save the amount on the left and the difference on the right as illustrated in Figure 2. Repeat this operation until the end of the rows.The total pixel represents the low-frequency part (denoted by the symbol L) while the pixel difference represents the high-frequency portion of the original image (indicated by the symbol H).
Step 2: scan the pixels from top to bottom in a vertical direction.Perform the addition and subtraction operations on the neighboring pixels and then store the sums at the top and the difference at the bottom as illustrated in Figure 3. Repeat this operation until the end of the columns.Finally, we will get four sub-bands each denoted as LL, LH, HL, and HH.Subband LL is a low-frequency part, so it looks very similar to the original image.Haar-DWT is a technique that can be readily applied to get four sub-bands LL, LH, HL, and HH.The processes in Haar DWT are not too complicated to do, but still, represent the frequency of the image.Therefore, in this paper, the Haar-DWT technique is used as an application of the DWT method to get sub-bands of the image and insert the document into the subband.

C. Process Flow of the Proposed Method
The flow of the embedding data and extraction process using the DWT and 3DES methods is explained as follows, Fig. 4: In the insertion process requires a cover image, document (.docx, .xlsx,.pdf,or .txt),and keys (K1, K2, and K3).The following are the steps in the insertion process : 1) Encrypt the document using 3DES with the key, so that it will obtain the ciphertext.

 Embedding Algorithm
In the extraction process requires a stego image and key (K1, K2, and K3).The following are the steps in the extraction process, Fig. 5: 2) Extract information in the HH sub-band on each sheet.
3) Combine the information that has been extracted to get a complete ciphertext.
4) Decrypt ciphertext using 3DES with the key to obtaining the document back.

III. EXPERIMENTS
In this research, the embedding process uses a few cover images shown in Fig. 6.
Those images are the images that will be inserted by the document.The types of support materials and cover images shown in Table 1.Before embedding the document into images, the calculation needs to be done to know the maximum integrated document size.The equation to calculate it shown in the equation below.After testing the image quality has been done, then the next step is insertion testing on the different frequencies shown in Table III.The next test is insertion testing using different insertion values as shown in Table IV.
After that the test of the stego image endurance against some image manipulation attacks shown in Table V.

3 ) 4 ) 5 )
Transform the cover image using DWT to get four subbands on the R, G, and B layers.Insert each ciphertext into the HH sub-band at each layer R, G, and B. Image reconstruction uses the IDWT process, resulting in a stego image.
size (KB), x = image width, y = image height The first test is the insertion test of the document.Any documents with different extensions shown in Table I will be inserted in each image to compare the time of insertion of each material in each image.The type of paper used in this test is a 12 KB docx, xlsx with a size of 9.67 KB, a pdf with a volume of 22.3 KB, and a txt with a capacity of 161 bytes.The results of this test shown in Fig. 7. Furthermore, the test results of document extraction from the image which contains the document in Fig. 8.

Fig. 8 .
Fig. 8. Graph of Document Extraction.The next test is testing the image quality on each image of the insertion shown in TableII.

TABLE II .
RESULTS OF IMAGE QUALITY TESTING

TABLE III .
INSERTION TESTING ON DIFFERENT FREQUENCY

TABLE IV .
INSERTION TESTING ON DIFFERENT INSERTION VALUE