一种多运动目标检测方法在FPGA上的实现

运动目标的检测已在众多的领域得到了广泛的应用，但是由于嵌入式处理器自身的速度限制，该应用主要集中在PC机上。使用一种改进的基于纹理的算法用于背景提取。采用高速的FPGA实现，通过对基于局部二元模式直方图算法的改进，使该算法适合硬件实现，通过一个图像块硬件结构的实现，在FPGA上同时实现12个图像块并行处理，使系统处理速度有了很大的提高。

Implementation of Moving Objects Detection Method on FPGA

Motion target detection and tracking has been widely applied in many fields. But due to the limitation of the embedded processor's speed, this application mostly concentrates on personal computers. This paper uses an improved texture-based algorithm for the background subtraction. The method is achieved with high-speed FPGA. After improving local binary pattern histogram-based algorithm, the algorithm becomes suitable for hardware implementation. In a hardware structure of the image block, the FPGA can process 12 image blocks simutaneously, which shows that the system's processing speed is greatly improved.