0975-766X CODEN: IJPTFI Available Online through

Aditya Kumar Sahu*et al. /International Journal of Pharmacy & Technology

ISSN: 0975-766X CODEN: IJPTFI Review Article

Available Online through www.ijptonline.com

DIGITAL IMAGE STEGANOGRAPHY TECHNIQUES IN SPATIAL DOMAIN: A STUDY Aditya Kumar Sahu, Monalisa Sahu Assistant Professor, CSE, K L University, Guntur, Andhra Pradesh, India. Received on 14-11-2016 Accepted on: 25-11-2016 Abstract Secure digital data communication is always a concern. Cryptography and Steganography are the prominent fields in secure digital data communication. By seeing an innocent image it is very difficult to imagine that, the image is doing the task of a messenger. When an image carries information without the knowledge of an outsider, is called as Image steganography. This paper discusses about various image steganographic techniques in spatial domain such as least significant bit (LSB), pixel value differencing (PVD), Combination of LSB and PVD, Modulus function etc. The comparison among various parameters has been made to determine the efficient technique. Keyword: Least significant bit (LSB), Pixel value differencing (PVD), Capacity, Stego-image. 1. Introduction: Internet is playing a major role in digital data communication [2]. Cryptography and Steganography are the most common platforms used to maintain secrecy to the data in transit [24]. In case of Cryptography the actual data will be replaced into encrypted data by using various transposition and substitution techniques [37]. Steganography is the composition of 2 Greek words steganos and graphia [36]. It means “cover writing” or “hidden writing”. Steganography hides the data by using some other media such as image, text, video etc. [37] as shown in fig-1. Types of Steganography

Image Steganography

Network Steganography

Video Steganography

Text Steganography

Fig-1. - Digital Mediums to Achieve Steganography. IJPT| Dec-2016 | Vol. 8 | Issue No.4 | 5205-5217

Page 5205

Aditya Kumar Sahu*et al. /International Journal of Pharmacy & Technology 2. Image steganography: When image is used to carry the message to the receiver, then it is called image steganography [37]. This works in either spatial or transform domain. Spatial domains are those domains, where the secret data is embedded directly to pixels. Where as in case of transform domain the information are transformed to frequency distribution. Again after processing and inverse transformation the image is retrieved in spatial domain. The various parameters to find the efficiency of any image steganographic algorithm are Capacity: The hiding capacity is maximum amount of data an image can hide. It is represented in bits per byte, or bits per pixel. Security: It is the ability to survive from different attacks. The more the security the better is the algorithm. Invisibility – The first and foremost requirement for any steganographic algorithm is the invisibility that is the ability to be unnoticed by the human eye. Tamper resistance: The steganographic algorithms should be robust. Imperceptibility: Any visual artifacts in the stego-image should not be noticeable to human eye. Peak Signal-to-Noise Ratio (PSNR), is used to find out whether the stego-image quality is acceptable or not. The PSNR can be calculated by the following equations. PSNR  10 log 2 (255 2 MSE )db P 1 Q 1

MSE  (1 P * Q) ( P( x, y )  P' ( x, y )) 2 m 0 n 0

Where, P and Q are image size. In the formula, P(x,y) is stands for the original pixel value, and P ' ( x, y ) 2 pixel values of stego-image. The greater the PSNR the lesser the distortion.

2.1.

Least significant bit (LSB) [1]:

Least Significant Bit Steganography is a simple way of embedding data in image. LSB technique directly embeds the secret data into the least significant bits of the pixel. Usually there are eight bit are there in a grayscale image. If you take a pixel value Pi = 127, the eight bit binary value of 126 is 01111110. The most significant bit (MSB) i.e. b7 of 01111110 is 0 and LSB i.e. b0 is 0, as shown in the figure 2.

IJPT| Dec-2016 | Vol. 8 | Issue No.4 | 5205-5217

Page 5206

Aditya Kumar Sahu*et al. /International Journal of Pharmacy & Technology Pi = (b7 b6 b5 b4 b3 b2 b1 b0)2 126 = 0 1 1 1 1 1 1 0

Most significant bit (MSB)

Least Significant Bit (LSB)

Fig-2. MSB & LSB of a pixel If we will change the MSB bit to hide another bit, i.e. we can hide either 0 or 1. Suppose we want to hide a 0 bit, then there I no change to the pixel value but if you hide a 1 bit, then the new value becomes 11111110 i.e. 254. It means there is a change in 254-126=128 of the pixel value. This much of change to the pixel values can easily identified to a normal eye. So data should not be inserted or embedded in the MSB of any pixel. In the otherside if we embed 0 in the LSB of the pixel then there is no change to the pixel value but if you hide a 1 bit, then the new value becomes 01111111 i.e. 127, and change in only 1 to the pixel value. This small change to the pixel value will not be noticed by a normal human eye. This technique is called simple 1-LSB technique. The LSB embedding approach has become the basis of many techniques to hide secret data. LSB such as 1-LSB, 2-LSB, 3-LSB and combining LSB with other image steganographic techniques such as Pixel value differencing (PVD) are exists in literature. 95(01011111) 1111

Original pixel

001

Secret data

89(01011001)

Stego pixel

Fig-3. Simple 3-LSBs technique. 2.2. Optimal LSB [1] To further increase the stego-image quality, in [1], introduced an improved technique called as optimal LSBs method. One of the three new formed pixels is called as optimal pixel or near to original pixel having secret data in it. The embedding process is as follows: Step1 - Let Py be the original pixel and the secret message to be embedded is of length k-bits. Step2 – The stego-image Px can be found by using simple LSB technique, then find another two pixel Pm,Pn by modifying to the k+1 bit of the stego image (Px). Where Pm= Px + 2K and Pn= Px - 2K. The three newly formed pixels are Pm, Px and Pn. The pixel which is close to the original pixel Py will be considered as optimal pixel. To find the optimal pixel (Po), Po = Pm, if |Py - Pm | df , 

IJPT| Dec-2016 | Vol. 8 | Issue No.4 | 5205-5217

Page 5208

Aditya Kumar Sahu*et al. /International Journal of Pharmacy & Technology   df' - df  px     2 

   df' - df   ,  px  1      2

   , if px ≥ px+1, df'