1、毕业设计 - I - 图像分割和特征提取技术研究图像分割和特征提取技术研究 摘摘 要要 图像分割是图像分析的第一步，是图像理解的重要组成部分，在有关图像处理的几乎 所有领域具有广泛的应用。因此，图像分割一直受到高度重视，对其研究具有十分重要的 意义。长期以来，研究人员提出了许多实用的分割算法。随着统计学理论，神经网络，小 波理论等在图像分割中的应用日益广泛，遗传算法、尺度空间、非线性扩散方程等近期涌 现的新方法和新思想也不断被用于解决分割问题，许多国内外学者也针对一些具体应用提 出了许多实用有效的方法。 本文介绍了数字图像处理技术中图像分割技术的基本理论和三种图像分割方法(1)基 于阈值图像分
2、割；(2)基于边缘检测及算子分割；(3)基于区域特性的图像分割。对基于点的 分割方法进行了较全面的叙述，主要研究了图像分割方法中的边缘检测法，区域提取法和 阈值分割法。通过大量的理论研习。并编写了 MATLAB 软件程序，对各分割方法进行了仿 真实验，得到分割图像。最后对于仿真进行了数据处理分析，验证了 Canny算子的整体效 果最好, Prewitt 算子分割细致。 但对于一幅图像仅仅只有只用一种方法达不到很好的效果， 而根据待分割图象的不同特点， 结合已知的先验知识， 研究符合具体图象特性的分割模型， 才是提高图象分割的重要手段。 关键词：图像分割；边缘法；区域法；阈值法；分水岭分割法 -
3、 II - Lmage Segmentation And Feature Extraction Technology Research Abstract Image segmentation is the first step in image analysis, image segmentation is an important component of image understanding, in almost all areas of the image processing has widely application. As a result, image segmentatio
4、n has been attached great importance to, its research has the very vital significance. For a long time,researchers put forward many practical segmentation algorithm. With statistics theory, the neural network, wavelet theory has been used increasingly in image segmentation, such as genetic algorithm
5、, scale space, and nonlinear diffusion equation with the recent emergence of new methods and new ideas are constantly being used to solve the segmentation problem, many scholars at home and abroad for some specific application put forward many practical and effective method. Digital image processing
6、 techniques were introduced in This paper introduces the digital image processing technology of image segmentation technology in basic theory and three methods of image segmentation. (1) based on threshold image segmentation. (2) segmentation based on edge detection and operator; (3) the image segmentation based on region feature. On the segmentation method based on the point of narrative, mainly studies the edge of image segmentation method, region extraction method and threshold segmentatio
