摘要
研究线阵CCD采集一维图像亚像素级别边缘检测方法;该方法经过图像平滑滤波、基于梯度算子边缘粗定位、拟合区间搜索、最小二乘直线拟合边缘精确定位几个步骤,定位出线阵CCD图像边缘位置;基于FPGA的图像采集卡接收基于DSP的图像处理卡下发的控制指令,接收CCD图像数据,并且负责图像数据的平滑滤波和梯度值计算,粗定位出边缘位置,然后把边缘粗位置坐标,以及截取粗定位的边缘点附近段数据上传给图像处理卡,以供后者进行下一步拟合区间搜索和最小二乘拟合精确定位出边缘位置;通过图像采集卡的前期处理,可以大大减少其与图像处理卡之间的通讯数据量,FPGA+DSP这种分工处理模式发挥了各自处理芯片的优势,大大提高了运算处理效率;实验结果表明,该算法检测精度达到亚像素级别,3σ小于0.5像素,处理时间小于90μs。
A subpixel-level edge detection method for linear CCD images is studied in this paper.The edge position of the array CCD image is detected by several steps of image smooth filtering,edge coarse orientation based on gradient operator,fitting region searching and edge precise orientation based on least square straight-line fitting.The image acquisition card based on FPGA receives the CCD image data after receiving the control command,which is sent by the image processing card based on DSP,and is responsible for the smooth filtering and gradient value calculation,the rough edge coordinate position,and then the coordinate and the data around it are passed to the image processing card for the next fitting interval search and the fine edge position.Through the early processing of the image acquisition card,the communication data between it and the image processing card is greatly reduced.This division of processing mode between FPGA and DSP takes the advantages of their respective processing chips and greatly improve the processing efficiency.The experimental results show the edge detecting precision achieves the sub-pixel level,the precision of 3σis less than 0.5 pixel,the processing time is less than 90μs.
作者
黄春霞
HUANG Chunxia(School of Optoelectronic and Communication Engineering,Xiamen University of Technology,Xiamen 361024,China)
出处
《计算机测量与控制》
2022年第10期45-50,共6页
Computer Measurement &Control
基金
福建省中青年教师教育科研项目(JAT190670)。
关键词
线阵CCD
边缘检测
平滑滤波
梯度算子
最小二乘直线拟合
linear CCD
edge detection
smooth filtering
gradient operator
least square straight-line fitting
