Privacy-Preserving Method for Public Health Surveillance Data using Image Steganography

Abstract

An algorithm of bio-inspired black widow is being optimised and is employed for design privacy-preserving methods for public health surveillance data using image steganography. The algorithm of the Black Widow Optimization is completely dependent upon the mating behaviour of the black spiders. The algorithm contains three basic steps namely, procreate, cannibalism, and mutation. The cannibalism step removes the inadequate solutions while finding the optimal solution. Thus, the BWO algorithm provides early convergence to find optimal solutions as compared to the existing optimization algorithm. In the proposed method, BWO algorithm is used for random key generation and optimized data hiding to enhance the imperceptibility. Initially, secret data is read and pre-processing is done to split the data into sensitive and non-sensitive attributes. After that, sensitive data is encrypted by performing the XOR operation with the random key generated using the BWO algorithm. Next, pre-processing of the cover image is done to select the most appropriate plane for data hiding based on the pixel intensities of the planes. After choosing the appropriate plane, optimized data hiding is done using the BWO algorithm. The BWO algorithm searches the optimal starting pixel and secret data order in the cover image. The simulation evaluation is done on the standard dataset images. The results show that the proposed method achieves superior results over existing optimization methods, namely, GA and Artificial bee colony. Besides that, the appropriate plane, optimal starting pixel, and secret data order is different for different cover images. Thus, it enhances the security of the proposed method.