Research Of Image Data Caching And Processing In Monitoring System

The embedded monitoring system has been widely researched and applied with national attention on the social security. In the embedded monitoring system, a large amount of image data need for high-speed transmission and processing, so how to improve the speed of transmission and processing in the system is a key factor to system’s real-time. When the image data in the monitoring system belongs to confidential information, the image needs to be encrypted. Because of the above reasons, based on DSP+FPGA embedded monitoring system, this paper studied on the system image data high-speed transmission and processing and the image data security encryption, completed the main work is as follows:1. According to the problem of appearing read/write waiting when the single-FIFO’s read/write frequency difference is very big, a high-speed transmission caching method is designed:the double-FIFO buffer is structured by using the FPGA, and the read/write waiting time is reduced when the double-FIFO alternately cached the image data. According to the problem of costing more CPU resources when DSP readed the data from external memory, designed a triple-buffer structure in the DSP’s internal RAM, the CPU dealt with the data in triple-buffer directly, the CPU’s processing cycle is saved and the processing speed is improved.2. According to the problem of the key space is smaller when the key parameters are selected directly in general chaos encryption algorithms, a multi-chaotic image encryption algorithm with external key is proposed. The algorithm used the Chebyshev’s chaotic sequence, which has characteristics of good balance and pseudo random, as external key, controlling the initial values and parameters of Tent map and Logistic map. Scrambled image pixel by using Tent chaotic map’s uniform distribution characteristics and diffused the image pixel by using logistic chaotic map’s high sensitivity to initial value. The simulation results show that this algorithm had a large key space, at least1085, and had high sensitivity to the key and good statistical properties, the algorithm can resist statistical attack and exhaustive statistical effectively.3. According to characteristics of the chaotic sequence which is produced by the hyperchaos system has higher randomness, an image encryption algorithm based on Tent map and hyperchaos Lorenz map is proposed. In this algorithm, image pixel is scrambled by using Tent chaotic map’s uniform distribution characteristics, the image pixel is diffused by using characteristics of hyperchaos Lorenz’s system structure complex and higher randomness of chaotic sequence. The simulation and safety analysis show that this algorithm had a large key space at least1096and high sensitivity to the key and and good statistical properties, the algorithm can resist statistical attack and exhaustive statistical effectively, it had high security and practical value.