摘要
针对当前采用的DSP+FPGA实时图像处理和控制系统的不足,提出了一种基于SOC架构的智能图像处理和控制系统,该系统硬件上使用SOC架构进行算法电路实现和外设控制,能够对所获取的图像进行跟踪、识别和匹配等图像处理,同时根据软件配置对外部器件进行控制和交互;通过对大尺寸卷积运算的验证试验,结果表明所设计的SOC架构具有通用性好、可靠性高、处理速度快和控制精准的特点,能够完全适应高复杂的卷积运算。
Aiming at the deficiency of DSP+FPGA real-time image processing and control system,an image processing and peripheral control system based on SOC structure is proposed.The system uses the SOC architecture for algorithmic circuit implementation and peripheral control.It can achieve image processing methods,such as object tracking,recognition,and template matching in obtained images.After the software configure its register,it can control and interact with the external devices.The verification results on large-scale convolution operations show that the SOC architecture has the characteristics of good versatility,high reliability,fast processing speed and precise control,and can fully adapt to highly complex convolution operations.
作者
王升哲
唐中和
郭航
刘书信
刘云峰
张承果
张枭
郑杰
Wang Shengzhe;Tang Zhonghe;Guo Hang;Liu Shuxin;Liu Yunfeng;Zhang Chengguo;Zhang Xiao;Zheng Jie(Department of Guidance and Control,Northern Laser Research Institute Co.Limited,Chengdu 610011,China;College of Computer Science,Sichuan University,Chengdu 610065,China;College of Optoelectronic Engineering,Chongqing University Chongqing 400040,China;4th Military Representative Office of Rocket Force Equipment Department in Chengdu Area,Chengdu 610052,China)
出处
《计算机测量与控制》
2021年第4期90-94,109,共6页
Computer Measurement &Control
基金
国防科技项目基金(KJ2017ZD42)。
