图像边缘检测器的FPGA实现  

The implementation of image edge detector based on FPGA

作　　者：覃永新[1] 田玉晓 陈文辉[1] 

机构地区：[1]广西科技大学电气与信息工程学院,广西柳州545006

出　　处：《广西科技大学学报》2017年第3期88-91,109,共5页Journal of Guangxi University of Science and Technology

基　　金：广西汽车零部件与整车技术重点实验室自主研究课题(15-A-02-02)资助

摘　　要：提高图像处理系统的处理速度,是当前图像处理技术的主要目标之一,其中纯硬件设计是一种越来越热门的手段.本文介绍了一种采用FPGA芯片实现Sobel算法来获取图像中的边缘像素的设计方法,图像在经过预处理得到灰度信息后,通过帧窗口处理、计算Sobel算子、与阈值比较等处理,判断出图像中的每一个像素是否为边缘像素,然后将边缘检测处理后的图像显示.系统采用50 MHz的时钟,经仿真波形可以看出,处理完一帧数据需要4.8 ms,相对于DSP系统来说,系统的处理速度得到极大的提高.Accelerating the image processing is one of the main goals of the image processing technology, which can be implemented by constructing the image processing system with pure hardware. A practical method based on Sobel algorithm is introduced, in which the FPGA chip is exploited to detect the edge pixel in the image. The gray level data of each pixel in the image is read by pre-processing module firstly, and then is extracted by module of frame-window to transform into the module of Sobel algorithm. After the results of Sobel algorithm are compared with a threshold, it can be determined that whether this pixel is an edge-pixel or not. After the processing of edge detection, the image is shown on the screen ultimately. It only takes 4.8 ms to process one frame of image data when the system clock frequency is 50 MHz. The result of the hardware testing shows that using a FPGA chip to process the image is faster than using a DSP processor.

关 键 词：图像处理 图像边缘检测 SOBEL算法 现场可编程门阵列 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]