Image encryption can be achieved by utilizing several spatial domain techniques to manipulate the placements of image pixels. The image scrambling method based on recursive sequences is one example. Several recursive sequences, including the Fibonacci sequence, cellular automata, and chaotic maps, are used to jumble images. Another method for encrypting images is to jumble blocks or coefficient matrices in the transform domain. However, because there aren’t any security keys or there isn’t much key space, these methods have very poor security levels. Moreover, it is well known that encryption methods based solely on permutations are susceptible to plaintext assaults. Altering the values of image pixels using a combination of logic operations and image bit plane decomposition is another method for encrypting images. Because the outcomes of its logic operations and deconstruction process are predictable, this technique has a much lower security level. Attacks using plaintext are not unheard of. One way to attain higher security levels is to use different approaches to jumble the placements of image blocks or pixels while altering their values. This study presents a new edge mapped combined key generation (EM-CKG) algorithm for grayscale and RGB picture encryption and decryption. EM-CKG generates a binary image of the same size as the original image that needs to be encrypted.