Infrared Image Edge Detection Technology And Fpga

Cellular neural networks (CNN) are a kind of recurrent neural networks proposed based on Hopfield neural networks and cellular automata, and constitute a class of recurrent and locally coupled arrays of identical dynamical cells, which can be implemented by VLSI easily and be applied to the signal processing. CNN are also a kind of nonlinear system, and with proper parameters, the dynamical behavior of a simple CNN system will show interesting bifurcation and complex chaos.In this thesis, we lay primary emphasis on the template design and application in image processing of cellular neural networks (CNN). Mainly we design template of CNN with something about edge detecting of image, with which we run simulations in PC. At the same time, the virtual CNN which is constructed by FPGA is used to process Real-time image. In the thesis, we study and discuss key algorithms and technique about system realization.This thesis proposes an efficient digital architecture for the cellular neural networks (CNN).The presented architecture is based on the combination of the bit-serial computation of distributed arithmetic (DA) with the characteristics of the CNN: the local connectivity and the translation invariance in the templates. Implementation of the CNN with the proposed architecture requires a low hardware complexity and a small number of bus lines. It consumes less silicon area because of the bit serial computation of DA and offers higher speed operation than the analog implementations of the CNN. As the key problem of image preprocessing with CNN is the design of template.Based on the study of PSO, this paper proposes a new method by using PSO's rapid convergence. The designed template is applied to edge detecting of image and the computer simulation results show that this method is better than the traditional algorithm results.A blue print of FPGA-based IRFPA video image digital processing system is proposed for testing the system scheme.The hardware of the system includes these function blocks as following: video A/D, data buffer FIFO, FPGA, data memory and colour space converter etc. The system can accurately read out the video signal of IRFPA, and then convert the analog signal into digital signal with A/D, and then the digital signal will be input into internal memory through FIFO, and CNN module carry out the processings such as edge dection. At last the system outputs the standard VGA analog video signal and propagates to LCD. According to the practical measurement of this system, the system performance meets with such requirements as stability, reliability and real time etc. And functionally, it is general to some extent.