Steganography Using Substitution Principle

Steganography Using Substitution Principle

DOI: 10.4018/978-1-5225-7516-0.ch002
OnDemand PDF Download:
No Current Special Offers


In this chapter, the author describes the various substitution-based image steganography techniques. Basically, there are four categories of substitution techniques: (1) least significant bit (LSB) substitution, (2) LSB array-based substitution, (3) group of bits substitution (GBS), and (4) adaptive LSB substitution. The LSB substitution-based techniques are very much insecure (detectable by RS analysis), so LSB array, GBS, and adaptive LSB substitution techniques are proposed in literature. If substitution principle is used in a steganography technique, then it must be evaluated by RS analysis.
Chapter Preview

Lsb Substitution

Simple LSB Substitution

In digital image steganography the simplest data hiding technique is least significant bit (LSB) substitution. For a gray image one pixel is represented in 8-bits. For a color image one pixel is represented in 24 bits. For every byte the right most bit is the LSB and left most bit is the most significant bit (MSB). LSB substitution means substitution of the LSBs by one data bit. For example, if my secret data is of 99 bits length then I can be able to hide in 99 pixels of a gray image or 33 pixels of a color image. When using a color image, the red, green and blue color components, each can hide 1 bit in their LSB position. For example, suppose in the following three pixels of a color image shown in Figure 1 (a), I want to hide the character A whose binary value is 10000001. Then the stego-pixels are given on Figure 1 (b).

Figure 1.

Original and stego pixels


In this example, we can see that out of 8 bytes we have touched; only 3 bits are changed from 0 to 1 or 1 to 0. In general, only 50% of the bits will be changed. In one-bit LSB substitution we hide only one bit in a byte. So, bit rate is 1. Bit rate means number of bits hidden per byte of the image. To improve the bit rate, substitution can be performed up to 3 least significant bits in a byte of the image, so that the bit rate can be increased up to 3.

Complete Chapter List

Search this Book: