Effective and Secure Transmission of Health Information Using Advanced Morphological Component Analysis and Image Hiding

Abstract

The using morphological component analysis, deep learning, and steganography, this study examines the secure transmission, identification, and validation of textual pictures via this Internet of Things-based channel. To extract characteristics from text-based pictures, morphological component analysis is utilised. Each of these traits has a distinct morphological component. Without losing visual quality, the morphological component technique lowers duplicative but also uncorrelated characteristics, as well as the size in bytes of a text-based picture. The capacity to obtain full text-based picture texture characteristics, rather than relying on a single spatial texture feature, is considered to be a significant aspect of the proposed approach. Before being transferred over Internet of Things-based networks, the morphological portions of a concealed text-based picture have been further divided and generally implanted interested in the least significant pieces of the cover pixels using spatial steganography. To retrieve the message from stored-images, the hidden key of an embedding algorithm can be exchanged with a text recovery approach on the recipient's side. To recognise an embedded text-based message, a hybrid convolution neural network technique was eventually employed. Aside from that an optimization approach is utilised to improve the hybrid convolution neural network’s performance (HCNN’s).