Mutual Authentication Using Image Processing and Visual Cryptography Protocol for Patient Database

Online Patient database is a particular type of service rendered by a chain of intercommunicated hospitals. Medical practitioners and clients are able to access their records or look into status of their diagnosis from any of the other or branch hospitals. The principle concern in online patient database is the authenticity of both the doctor and the client. Due to ineluctable intruding of the database on the Internet, it is very hard to rely on any kind of data on the Internet. The propounded approach of ours, based on Image processing technique and Visual Cryptography, the authentication problem is resolved. This particular paper proffers a technique of processing the Endoscopic Ulcer images based on Grow-cut and then divides it into Secret shares based on Random Grid Visual cryptographic Techniques. The total number of shares to be created depends on the strategy selected by the medical practitioner. If two shares are prepared, one of the shares is reposited in the Global Hospital Server, and the other is retained by the client. The client has to produce his/her secret share during the consultation. Client's share is then stacked with the Hospital Server share to obtain the original Endoscopic image. This method of correlation aids in concluding the decision based on espousal or denial of the output and hence confirms the client.