Research on Image Information Security Algorithm Based on Wavelet Neural Network

Abstract

With the continuous emergence of digital images and digital videos, a large number of images containing sensitive information need to be kept secret, and a large number of digital media works need intellectual property protection. Image security and confidentiality technology has become another hot spot in the field of information security. In view of the shortcomings of the statistical processing algorithm for the verification results adopted by various enterprises in the network information security evaluation, through the research on the characteristics of computer security, this paper compares the selection of weights in the commonly used weighted arithmetic average method. Image information encryption originates from the early classical encryption theory. Its purpose is to transform a given image into a disordered image in the spatial or frequency domain according to certain transformation rules, so as to hide the real information of the image information itself. How to protect image information security has become a hot topic in the international research. In order to better adapt to the unique nature of image information and meet the needs of data security and copyright protection, scholars are actively seeking new technologies and methods. This paper studies the basic algorithm of wavelet neural network and applies it to image information security. The main idea of the algorithm is to combine the theory of wavelet transform with the algorithm of neural network, and then get the algorithm of wavelet neural network.