Securing the image through the combination of elliptic curve cryptosystem and hill cipher encryption

The advancement of communication technology helps users sharing the images through internet. However, the sharing through unsecure channels may expose the images to certain attack that will compromise their confidentiality. Image encryption is one of the methods that protect against this threat. Hill Cipher is being applied in image encryption because of its simple operation and fast computation, but it also possesses a weak security level which requires the sender and receiver to use and share the same private key within an unsecure channel. There is a proposed technique that combines elliptic curve cryptosystem together with Hill Cipher (ECCHC) to overcome the above mentioned weakness, however, the experiment only performed an analysis on four images which leads to inaccuracy of the results. There were two objectives in this research. The first objective was to redesign and redevelop the image encryption technique, a combination of Elliptic Curve Cryptosystem and Hill Cipher (ECCHC). The second objective was to test the ECCHC technique using 209 images from USC-SIPI database that consisted of 159 grayscale images and 50 colour (RGB) images in order to obtain reliable results. In ECCHC, the sender and receiver must agree on the elliptic curve function and share the domain parameter to initiate the key generation including private key, public key, and 4 x 4 self-invertible key matrix before sharing image. The image will split into size four (vector of 4 x 1) that later will be encrypted using Hill Cipher with 4 x 4 self-invertible key matrix. The generated 4 x 1 vector image will take modulo 256 to generate the cipher image. Once the receiver receives the cipher image, decryption can be done by using the 4 x 4 self-invertible key matrix, generated on receiver site. Matlab is a simulation environment to conduct the experiments. There were two experiments of this study. The first experiment was re-implementation of ECCHC technique and run the testing on the same four images that had been used by the based work paper. On the other hand, the second experiment was extensive analysis of ECCHC technique by using 209 images from USC SIPI database. The results from both experiments discussed on Entropy, Peak Signal to Noise Ratio (PSNR) and Unified Average Changing Intensity (UACI). For the second experiment, there were additional results of the analysis on the actual data of the encrypted images to ensure the confidentiality of the encrypted images.. From the obtained results, it has been proven that several images were not encrypted well. In conclusion, the technique did not guarantee a secure image encryption as it was verified that certain images were not successfully encrypted