基于形状特征的物体识别系统设计

基于形状特征的物体识别系统设计(任务书,开题报告,论文12800字)
摘  要
本文通过使用matlab识别了爱心、箭头和三角形，是通过傅里也描述子对这三类目标进行对比与分析，判断与分析物体形状特征的问题人们通常会使用傅里叶描述子进行适当的操作。一般都是使用傅里叶连续变换法，在傅立叶描述子，物体边缘的主点，物体边线的相似形状的相位效应，来说明傅里叶的平移，旋转和尺度不变这三方面的特性。和传统的归一化方法和傅里叶离散变换相互比较，在相位与模式特性这方面新的办法使之保留，并且帮助我们更好地分析形状。通过这次实验可以看出，新方法比传统方法更为准确合理有用。
本文的总体框架按照以下模块进行说明：第一章为绪论；第二章至第五章对上面探讨的问题进行了较为深入的剖析，第六章是总结。本文主要探讨了三个具有代表性的目标：爱心、箭头和三角形作为物体识别。总之，利用图像分割技术提取图像中的有用对象，并用傅里叶描述进行分析比较得出物体形状特征，并作识别。
本文特色：本文详细介绍了傅里叶描述子，较为深入地剖析了了现有文献中分析物体边缘的常用方法，即通过傅立叶描述子基于极坐标函数改进了矩不变量和矩序列，帮助我们更方便认识这门技术。
关键词：模式识别；图象分割；并行算法；傅立叶描述子

Abstract           
Through the use of MATLAB to identify love, arrow and triangle, this paper compares and analyzes the three kinds of targets through the Fourier descriptor. People usually use Fourier descriptors for proper operation to judge and analyze the shape characteristics of objects. In general, the Fu Liye continuous transformation method is used to describe the three aspects of Fu Liye's translation, rotation and scale invariance in Fu Liye descriptor, the main point of the edge of the object, and the phase effect of the similar shape of the edge of the object. Compared with the traditional normalization method and the Fu Liye discrete transformation, a new approach to the phase and pattern characteristics is preserved, and it helps us to better analyze the shape. Through this experiment, we can see that the new method is more accurate, reasonable and useful than traditional methods.           
The overall framework of this paper is explained in accordance with the following modules: the first chapter is the introduction; the second chapter to the fifth chapter makes a more in-depth analysis of the problems discussed above, and the sixth chapter is a summary. This paper mainly discusses three representative goals: love, arrowhead and triangle as object recognition. In a word, the image segmentation technology is used to extract useful objects in the image, and the shape features are identified and compared by Fu Liye description.          
 This paper introduces the Fu Liye descriptor in detail, and deeply analyzes the common methods to analyze the edge of the object in the existing literature, that is to improve the moment invariants and moment sequences by the polar coordinate function of the Fu Liye descriptor, which helps us to understand the technique more conveniently. 
Keywords:pattern recognition; image segmentation; parallel algorithm; Fu Liye Miao Shuzi.


目  录
第1章绪论…………………………………………………………………………1
1.1  研究的目的和意义…………………………………………………………1   
1.2  研究的历史和现状…………………………………………………………1
1.3  本课题的研究内容和安排…………………………………………………3
第2章  模式识别理论基础………………………………………………………4
2.1  模式识别的意义及用途……………………………………………………4
2.2  模式识别的基本方法………………………………………………………5
2.3本章小结………………………………………………………………………5
第3章  图像分割及目标边缘提取………………………………………………7
3.1   引言………………………………………………………………………7
3.2   本文图像分割及边缘提取算法…………………………………………8
3.3本章小结……………………………………………………………………8
第4章   傅里叶描述子…………………………………………………………10
4.1傅里叶变换…………………………………………………………………10
4.2描述子的定义………………………………………………………………10
4.3傅里叶描述子基本原理……………………………………………………11
4.4性质…………………………………………………………………………12
4.5 采用傅里叶描述子的目标识别方法………………………………………14
4.6本章小结………………………………………………………………………14
第5章 仿真试验及结果分析……………………………………………………18
5.1仿真试验………………………………………………………………………18
5.2本章小结………………………………………………………………………22
第6章 总结………………………………………………………………………23
致谢………………………………………………………………………………26