Image Encryption using Edge Maps and Scrambling methods

Encryption is an effective way to avoid threats to data. Here we perform bit plane decomposition(e.g. Fibonacci P-code, Truncated Fibonacci P-code methods ) for the source image which is to be encrypted, where the number of bit-planes generated depends on the bit length of the pixel. Other than our source image we use another key image to which we perform edge mapping with randomly generated thresholds(e.g. Canny, Prewitt, Sobel, LoG, Roberts, Kirsh edge generators), where the number of edge maps generated depends on the number of bit planes. We perform a pixel scrambling method to the edge maps which generates scrambled edge maps. Then xor operation to each individual bit-plane with the scrambled edge map is done. These images are combined using a randomly generated sequence to form an encrypted image. The final encrypted image is then formed by scrambling the encrypted image. In this process the results of encrypted images are high with greater NPCR and UACI values and lower Correlation values, depicting the consistency of this algorithm and the security against differential attack for any image.