子像素边缘检测技术.doc

摘 要 边缘指的是图像中像素值有突变的地方。边缘检测是图像处理的重要的一部分。边缘往往携带着一幅图像的大部分信息。在分析对比已有边缘检测算法的基础上，设计了两种边缘检测方法。第一种方法先用Sobel算子粗定位，然后用三次样条插值函数对灰度图像进行插值，使目标达到亚像素级，对插值后的灰度图像，利用最大类间方差确定阈值，实现亚像素级的边缘检测。另一种方法是根据灰度矩算子在目标成像前后的矩不变特性，利用Tabatabai等人提出的前三阶灰度矩，实现了亚像素边缘检测。通过实验对算法有效性和检测精度进行了研究和验证，给出了工件的实测尺寸对比结果。实验表明，基于灰度矩的亚像素边缘检测算法和基于Sobel算子的亚像素级边缘检测法比传统的边缘检测算子具有更高的定位精度。 关键词：亚像素级边缘检测；基于插值法的边缘检测；灰度矩 ABSTRACT Edge refers to the value of the pixel in the image mutations. Edge detection is an important part of image processing. The edges tend to carry most of the information of an image. In this paper, the design two edge detection methods, the first method first used Sobel operator rough location, and then grayscale image interpolation, cubic spline interpolation function so that the target to achieve sub-pixel level grayscale images after interpolation, the use of the maximum variance between the threshold is determined to achieve sub-pixel edge detection. Another method is the gray moment operator in the moments before and after the target imaging invariant Tabatabai, who proposed the first three gray moment, to achieve sub-pixel edge detection, through experiments the effectiveness of the algorithm and testing the accuracy of research and validation, given the comparison of measured dimensions of the workpiece. The experiments show that, based on gray-scale operator sub-pixel edge detection algorithm has a higher positioning and accuracy than the traditional operator to meet the image target to achieve sub-pixel edge detection. Key words：Sub-pixel edge detection；Edge Detection based interpolation； Gray Moment； 目 录 TOC \o 1-3 \h \z \u HYPERLINK \l _Toc327518299 第1章 绪论 PAGEREF _Toc327518299 \h 1 HYPERLINK \l _Toc327518300 1.1 课题研究背景及意义 PAGEREF _Toc327518300 \h 1 HYPERLINK \l _Toc327518301 1.2 边缘检测的研究历史和发展趋势 PAGEREF _Toc327518301 \h 2 HYPERLINK \l _Toc327518302 1.3 论文研究内容，解决问题和章节安排 PAGEREF _Toc327518302 \h 4 HYPERLINK \l _Toc327518303 第2章 经典边缘检测算子 PAGEREF _Toc327518303 \h 5 HYPERLINK