用VC6.0实现亚像素边缘精密检测

文档字数:6895,页数:20 只有文档

摘 要
 在光测进行各种精密测量的领域中，提高光测系统测量精度最直接的方法就是提高 CCD摄像机分辨率，即增加像素点数。近二十年来，在光测数字图像处理领域，许多研究者试图利用软件处理的方法来解决图像中目标的高精度定位问题。如果能用软件方法将图像上的特征目标定位在亚象素级别，就相当于提高了系统精度。本文介绍了一种基于多项式插值法的改进的亚像素精度边缘检测算法及相应计算模板和公式.本算法应用在经典Sobel算子基础上构造出的方向模板,对灰度图像进行处理,得到梯度图像,然后在梯度图像上沿目标边缘的梯度方向进行多项式插值法亚像素细分计算,对目标边缘进行亚像素精确定位。并且用VC++实现该算法。
 关键词：亚像素，图像测量，精度边缘检测，图像分割
  

Abstract
 In the field of the flash ranging system carrying on each kind of precision measurement, improving the CCD camera resolution ratio is the most direct method to improve the measuring accuracy of the flash ranging system. In the flash ranging digital image processing domain, many researchers have attempted to solve the question of high accuracy localization in the image using the software for the recent 20 years. It is equal to increase the precision of the system, if we can locate the characteristic target of an image in the sub pixel rank using the software. An improved sub pixel edge detecting algorithm which is based on the polynomial interpolation, calculating template and formula have been introduced in this text. This algorithm applies to the direction template based on the classical Sobel operator. We can obtain the gradient image through processing the gray image and then carry on the sub pixel segmentation computation along the gradient direction of the target edge using the polynomial interpolation in the gradient image to fix the position of the target edge accurately. All these algorithms are realized with the VC++.
 
 Key words: sub pixel, image measure, and precision edge detection.
 

目   录
摘 要 I
Abstract II
第1章  前言 1
1.1 背景 1
1.2 亚像素概念 1
1.3 研究现状 2
1.4 本章小结 2
第2章  开发工具和数字图像基础知识 3
2.1 Visual C++6.0简介 3
2.1.1工具简介 3
2.1.2 MFC简介 4
2.2 数字图像其本概念 5
2.2.1 数字图像的表示 5
2.2.2 Sobel原理 5
2.3 本章小结 7
第3章 算法 8
3.1 算法原理 8
3.2 算法流程图 11
3.3 本章小结 11
第4章  程序结果与分析 12
4.1图片的边缘处理 12
4.2 亚像素精度 13
4.2.1精度分析 13
4.2.2亚像素精度实验结果 13
4.3 本章小结 15
第5章 总结 16
第6章 致谢 17
第7章　参考文献 18

相关论文