An optimized high-speed high-accuracy image matching system based on FPGA

Accuracy and real-time performance are the main factors that determine the practical value of image matching system. To implement a high-speed high-accuracy low-source consumption image matching system, the improved formula of gray-correlation-based algorithm is selected, a pipeline-based highly-parallel structure is carefully designed and many optimization methods are adopted. Experimental results on the Altera Stratix II EP2S130F780 platform show that the system is much faster than other similar systems while retaining the accuracy no less than 10−5 and supporting a maximum of 600*600 reference image.

[1]  Hong-yan Shi,et al.  Design and realization of image matching based on GMPT algorithm , 2008, 2008 Chinese Control and Decision Conference.

[2]  Chew Lim Tan,et al.  Matching of double-sided document images to remove interference , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[3]  Gu Hui,et al.  Image Match Algorithm Based on Feature Point with Bidirectional Threshold , 2005, 2005 International Conference on Neural Networks and Brain.

[4]  Tao Zhang,et al.  High Efficient Implementation of Image Matching Algorithm , 2009, 2009 2nd International Congress on Image and Signal Processing.

[5]  Yan Xue long,et al.  Improvement on Realization Structure of Block Matching Algorithm and its Performance Verification , 2007, 2007 International Conference on Communications, Circuits and Systems.

[6]  Wei Wang,et al.  Image Matching Algorithm Based on Subdivision Wavelet and Local Projection Entropy , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[7]  Avishek Saha,et al.  Speed-area optimized FPGA implementation for Full Search Block Matching , 2007, 2007 25th International Conference on Computer Design.

[8]  Wael Badawy,et al.  A fully parallel-pipelined architecture for full-search block-based motion estimation , 2002, The 14th International Conference on Microelectronics,.

[9]  Qiu Hui-juan.  Fast Image Matching Based on Correlation Coefficient , 2007 .

[10]  Xiaohang Hu,et al.  A New Image Matching Algorithm Based on the Relative Position of Intensity , 2007, 2007 10th IEEE International Conference on Computer-Aided Design and Computer Graphics.

[11]  Jane You,et al.  A wavelet-based coarse-to-fine image matching scheme in a parallel virtual machine environment , 2000, IEEE Trans. Image Process..

[12]  Guo Cheng-ming.  Implementation of Picture Product Correlation Algorithm Ba sed on TMS320C6000 DSP , 2003 .