摘要
针对目前基于角点特征的视频分析系统前期特征提取速度较慢的问题,提出了一种利用FPGA来实现Harris角点提取的方法,并在充分考虑FPGA处理器结构以及Harris角点提取算法理论的情况下,对算法进行了进一步的改进,最后进行了仿真验证。结果表明,新算法在保证提取的角点完整准确的情况下,进一步提高了系统的实时性,同时大大简化了系统结构,减少了系统硬件开销,非常容易实现,具有较强的实用价值。
A method for applying FPGA to realize the Harris corner extraction is proposed,aiming to overcome its slow process rate in video analysis system based on corner feature,and by taking full account of the FPGA processors structure and Harris corner detection algorithm theory,the algorithm is further improved,and finally simulated.The simulation result shows that the novel algorithm could raise the real-time performance of the system under the conditions of ensuring the completeness and accuracy of corner extraction,and at the same time greatly simplify the system structure and reduce the overhead of system hardware resource.This novel algorithm is very easy in implementation and of fairly strong practical value.
出处
《通信技术》
2010年第4期158-160,共3页
Communications Technology
