A Novel Cryptographic Key Generation Method Using Image Features

Several methods in steganography and cryptography have been proposed for secured communication. But many compromises are there in them. Steganography which deals with security using multimedia has a very big disadvantage of increase in the information size. Cryptography which deals with encryption using the keys generated by various algorithms has the disadvantage of difficulty in remembering the key which can be easily cracked. The main objective of this study is to increase the security in communication by encrypting the information with the key generated by using an image. This study proposes a novel algorithm for key generation using image features. It is a reliable and flexible method of key generation for information security. In this method the user need not remember the keys during encryption and decryption. This study uses the Gray Level Co-occurrence matrix of an image to extract the Gray Level Co-occurrence properties of the image. A 56-bit sub-key is generated from the extracted Gray Level Co-occurrence properties. Then the key for encryption and decryption is generated using the sub-key generated from the image. This proposed method is easy to implement. The key strength is much better than others.