液晶与显示, 2020, 35 (3): 250, 网络出版: 2020-05-12   

一种组合GAUSS-filter、SOBEL、NMS、OTSU 4种算法的图像边缘检测的FPGA实现

FPGA implementation of image edge detection based on four algorithms of GAUSS-filter, SOBEL, NMS and OTSU
作者单位
青岛大学 电子信息学院, 山东 青岛 266071
摘要
边缘检测是分离图像目标主体和背景的主要依据。通过检测图像上灰度显著变化的像素点集合来确定图像的边缘。目前边缘检测的方法大多综合应用两个算法来提高边缘检测质量。本文在FPGA上综合应用GAUSS-filter、SOBEL、NMS、OTSU等多种算法进行图像边缘检测。将原始采集到的彩色图像转换成灰度图像, 先对图像进行高斯平滑处理, 消除干扰性噪声; 应用SOBEL算法计算图像梯度值, 再应用NMS算法处理灰度, 挑出轮廓边缘梯度值最大的灰度像素集合; 最后应用OTSU算法自适应地确定图像门限阈值进行边缘检测。该方法不仅能刻画出图像的边缘, 而且还能对边缘进行细化, 实验结果显示对不同的图像都有很好的检测效果。本文所提出的在FPGA上实现的图像边缘检测方法具有准确、实时、高速等特点, 能够应用到许多领域中。
Abstract
Edge detection is the main basis for separating the target object and background of the image. The edges of the image are determined by detecting a set of pixel points whose gray level changes significantly on the image. At present, most of the methods of edge detection apply two algorithms to improve the quality of edge detection. In this paper, GAUSS-filter, SOBEL, NMS and OTSU algorithms are adopted and implemented on the FPGA to perform image edge detection. The original collected color image is converted into a grayscale image, and the image is firstly smoothed by GAUSS-filter to eliminate interfering noise. The SOBEL operator is used to calculate the image gradient value, then the NMS algorithm is applied to process the grayscale, and pick out a set of the grayscale pixels with the largest gradient value of the contour edge. Finally, the OTSU algorithm is used to adaptively determine the image threshold for edge detection. This method can not only describe the edge of the image, but also refine the edge. The experimental results show that the image has a good detection effect on different images. The method proposed in this paper that has been implemented on FPGA of the image edge detection has the characteristics of accurate, real-time, high-speed, embedded, etc., and can be applied to many fields.

张灏, 王素珍, 郑宇, 王伟, 李潍志, 王鹏, 任贵珊, 马家麟. 一种组合GAUSS-filter、SOBEL、NMS、OTSU 4种算法的图像边缘检测的FPGA实现[J]. 液晶与显示, 2020, 35(3): 250. ZHANG Hao, WANG Su-Zhen, ZHENG Yu, WANG Wei, LI Wei-zhi, WANG Peng, REN Gui-shan, MA Jia-lin. FPGA implementation of image edge detection based on four algorithms of GAUSS-filter, SOBEL, NMS and OTSU[J]. Chinese Journal of Liquid Crystals and Displays, 2020, 35(3): 250.

本文已被 2 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!