Study and Design of Image Pre-Processing System Based on FPGA

Abstract:

The high speed of image processing is required in the real-time image processing system. To improve the image processing speed, it is mainly implemented by hardware. This thesis constructs an image preprocessing system based on FPGA. Several typical image preprocessing algorithms, such as fast median filter, mean filter, Gaussian filter, Laplacian and Sobel operator edge filters, are implemented in the system. In the process of the system design, the characteristics and requirements of the image-processing implemented by hardware are fully considered. The parallel attribute insiding the image algorithm is effectively dug and the pipeline structure is adopt. All mentioned above are useful to optimize and improve the algorithms, reduce the hardware consumption and raise speed of image processing.

• Periodical: Advanced Materials Research (Volumes 482-484)
• Pages: 200-205
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.482-484.200
• Online since: February 2012
• Authors: Jin Tong Li, Xiao Jie Duan
• Keywords: Algorithm (GA), Field-Programmable Gate Array (FPGA), Image Preprocessing

References

[1] D. J. Allred, Yoo.Heejong, V.Krishnan. A novel high performance distributed arithmetic adaptive filter implementation on an FPGA. IEEE International Conference on, 2004, 5:17-21

DOI: 10.1109/icassp.2004.1327072

[2] Kenech R Castleman.Digital Image Processing Prentice Hall International, Inc, 1998:71~83

[3] Nicol N. Schraudolph, Richard K. Belew, Dynamic Parameter Encoding for Genetic Algorithms Machine Learning, 1992.9(1): 9-21

DOI: 10.1007/bf00993252

[4] C. Pilato, A. Tumeo. Improving evolutionary exploration to area-time optimization of FPGA designs.Journal of System Architecyure, 2008, 4:1046-10

DOI: 10.1016/j.sysarc.2008.04.010

[5] Mostafa, Khadragi. Hardware implementation of genetic algorithm on FPGA.Radio Scence Conference, 2004, C9:1-9.

DOI: 10.1109/nrsc.2004.240504