摘 要
随着生产、生活需求的提高,目前，自动识别指针式仪表的读数成为人们的一个研究热点。基于数字图像处理技术的指针式仪表读数识别是指针式仪表自动检定的关键技术。Hough变换是一种用于区域边界形状描述的方法，经典Hough变换常常被用于直线段、圆和椭圆的检测等等。
本文利用Hough变换检测直线的基本思想和算法，研究并基本实现了基于数字图像处理的指针式仪表的读数自动识别。在获取指针式仪表的数字图像后，首先对图像进行预处理，然后通过Hough变换准确、快速地识别出指针所在直线的位置，最后利用指针位置和指示表读数的对应关系得到指示表的读数。实验结果表明，该算法实现了指针式仪表的读数，有较高的精度。克服了人眼读数不固定，易疲劳的的缺点，提高了效率。
关键字：数字图像处理 Hough变换 读数识别 指针式仪表
I
Title Research of pointer instrument reading recognition based on digital image processing
Abstract
With the increasing of production and life demands,at present, the automatic recognition of the pointer instrument become a hot topic in the research area. pointer meter reading recognition based on digital image processing technology is the key technology of the automatic verification of the pointer instrument . Hough transform is a method which used to describe the shape of a domain boundary. The classical Hough transform is most commonly used for the detection of regular curves such as lines, circles, ellipses, etc.
This paper describes the basic principle and algorithm of the Linear Hough transform,Research and basically achieved the automatic identification of pointer type instrument readings , which is based on digital image processing .After obtaining the pointer image , the computer first preprocess it, then recognizes the position of the pointer accurately and quickly by Hough transform, finally gets the reading according to the relation between the reading and the position of pointer. The experiment result proves that the operator has higher precision,the algorithm to realize the pointer type meter reading , the operator overcome eye reading is fixed improve efficiency.
Keywords Digital Image Processing Recognition of Reading Hough transform Pointer type instrument
II
目 次
摘 要.............................................................. I Abstract........................................................... II 1 绪论............................................................. 1 1.1 图像识别技术............................................... 1 1.2 MATLAB的简介 .............................................. 4 2 图像预处理....................................................... 7
2.1 图像灰度 ................................................... 7 2.2 图像平滑 ................................................... 9 2.3 图像二值化 ................................................ 12 2.4 边缘检测 .................................................. 13 3 Hough变换 ...................................................... 17
3.1 Hough变换简介............................................. 17 3.2 Hough变换的控制原理和算法................................. 17 3.3 基于Hough变换的指针式仪表识别算法 ........................ 19 3.4 Hough变换检测直线的优缺点................................. 21 4 总体实现和读数识别.............................................. 23
4.1 总体实现 .................................................. 23 4.2 读数识别 .................................................. 27 总 结............................................................. 28 致 谢............................................................. 29 主要参考文献....................................................... 30
III