摘要
针对运动目标检测对实时性和准确性要求较高的问题,本文设计了一种基于FPGA的运动目标检测系统。首先,用CMOS摄像头采集视频图像,缓存入SDRAM存储器,对缓存的图像采用帧间差分法检测出运动目标;之后,对帧间差分法在画面中产生的空洞与噪声点,采用形态学滤波滤除;最后,将处理后的图像送入图像显示模块进行显示。在QuartusⅡ13.0环境下完成该算法的硬件设计。实验结果表明,设计的系统能够实时地检测出运动目标且准确性高。
For the problems of high demand for real-time and accuracy of moving target detection, a moving target detection system based on FPGA is designed in this paper. First, the video is collected with the CMOS camera and cached into the SDRAM, then using the frame difference method to detect the moving target in the video. After that, the voids and noise points produced by the inter frame difference method in the video are filtered by morphological filtering. Finally, the processed video is sent to the image display module for display. The hardware design of the algorithm is completed under the environment of Quartus II 13.0. The experimental results show that the designed system can detect the moving target in real time and the accuracy is high.
出处
《仪器仪表用户》
2018年第3期40-42,48,共4页
Instrumentation
基金
天津市应用基础与前沿技术研究计划项目(15JCYBJC51800)
关键词
FPGA
目标检测
形态学滤波
FPGA
object detection
morphological filtering