This Demonstration shows how a character can be hidden in an image with least significant bit substitution. You can see that the encrypted image is indistinguishable from the original. The "adjusted subtraction" image shows the difference between the images.
Steganography is the science of hiding messages so that only a person who knows the existence of the message can obtain the hidden information. The least significant bit substitution method takes the information to be hidden (in this case, the ASCII value of the character) and converts it to a binary representation. Then, every pixel of the image is also converted to a binary representation, and the least significant bit of every pixel is modified with the value of a bit corresponding to the information that the sender wants to hide. In this way, the image is not perceptibly modified because the least significant bits are the ones that least affect the image.
 C. Paar, J. Pelzl, and B. Preneel, Understanding Cryptography: A Textbook for Students and Practitioners, Heidelberg, Germany: Springer-Verlag, 2010.
 P. Wayner, Disappearing Cryptography: Being and Nothingness on the Net, San Diego: Academic Press, 1996.
 M. Welschenbach, Cryptography in C and C++, 2nd ed. (D. Kramer, trans.), New York: Apress, Springer-Verlag, 2001.