摘要
针对目前使用CPU或者树莓派图像处理实时性差、边缘检测实际效果差等缺点,课题组提出了基于现场可编程门阵列的边缘检测系统。设计了基于现场可编程门阵列的边缘检测系统的硬件方案,介绍了边缘检测的原理;基于硬件设计了图像采集与处理方案;算法上先通过实验设定卷积核的大小,然后使用Sobel卷积核对获取的图像进行卷积,再使用Robert卷积核进行卷积,并在显示器上显示实时图像处理结果。实验表明该系统的检测效果更加清晰,断点更少。该系统能够并行处理数据,提高图像处理速度,且实际的边缘检测效果好。
Aiming at the shortcomings of current CPU or Raspberry Pi image processing such as poor real-time performance and poor edge detection,an edge detection system based on field programmable gate array was proposed.The hardware scheme of the edge detection system based on field programmable gate array was designed and the principle of edge detection was introduced.The image acquisition and processing scheme was designed based on hardware.In the algorithm,the size of the convolution kernel was set through experiments and the acquired image were convolved by Sobel convolution kernel.Afterwards,the convolution was performed by Robert convolution kernel and the real-time image processing results were displayed on the monitor.The experimental results show that the detection effect of the system is clearer and with fewer breakpoints.The system can process data in parallel and improve image processing speed with good actual edge detection effect.
作者
张阵委
章伟
颜晨航
ZHANG Zhenwei;ZHANG Wei;YANC henhang(Laboratory of Intelligent Control and Robotics,Shanghai University of Engineering Science,Shanghai 201620,China;School of Mechanical and Automotive Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《轻工机械》
CAS
2020年第2期64-68,共5页
Light Industry Machinery
