A Remote Sensing Image Encryption Method Combining Chaotic Neuron and Tent Map

A Remote Sensing Image Encryption Method Combining Chaotic Neuron and Tent Map

Aiming at the characteristics of remote sensing images with large amount of data, and the shortcomings of low-dimensional chaotic systems such as small key space and poor randomness of generated chaotic sequences, chaotic neurons are introduced. Considering that the chaotic sequence generated by flat tent map is more uniform than other chaotic systems and conforms to the two-dimensional characteristics of images, a remote sensing image encryption method combining chaotic neuron and tent map is proposed. The initial key to this method is generated from the plaintext image hash value. This study takes GF-2 multi-spectral image as an example, and the research area selects the area around Songyuan city, covering a variety of ground object types. In the simulation results and comparative analysis, information entropy, gray histogram, correlation coefficient and other indicators are used for comparison and analysis. These operations have fully verified that the method which has good key sensitivity can fully resist differential attacks and other means of cracking, and effectively protect all kinds of information inside the remote sensing images.