自动化毕业论文基于K均值的图像分类设计与实现--133011312.docVIP

3 - 南京邮电大学通达学院 毕 业 设 计（论 文） 题 目：基于K均值的图像分类设计与实现 专 业 自 动 化 学生姓名 施昊琛 班级学号 指导教师 朱松豪 指导单位 自动化学院 日期：2012年3月14日至2012年6月17日 摘 要 K均值算法在高光谱遥感影像的非监督分类中具有较强的实用性, 表现出了良好的优点。首先采用了最大最小选心法确定初始类别中心, 然后使用了K均值算法实现遥感影像的分类。在分类过程中采用了VC++2005作为开发平台, 极大地提高了遥感影像的分类速度, 同时还给出了实现K均值分类主要步骤的代码。最后在深入分析不同迭代次数下得到的不同分类图的基础上, 研究了迭代次数值对最后分类结果的影响。 关键词:非监督分类; K均值算法; VC++2005 ABSTRACT Human Identification System Based on Biometrics is a totally brand-new technique different from traditional methods because it adopts the inherent organisms characteristics of human body. More and more people pay great attention to it as it is safer, more reliable and effective. It begins entering every realm of our society and meeting the challenge of new era. This paper first introduces the research background of the computer face recognition technology and the main method, and then propose a discriminant analysis based on Fourier transform and Face Recognition. Linear discriminant analysis (LDA) in the PCA combined with Fisher on the basis of the principle criteria used to extract feature vectors, the proposed removal of the feature vector retained the original image space in the major categories of information, and can the minimum mean square error reconstruction of the sense of the original image. The proposed method first uses Fourier transform to extract the face image is relatively stable low-frequency sub-band, blurring the facial expression and posture influence, and also reached a dimension reduction of the effect of the image vector, and then use the linear discriminant analysis and identification feature extraction components. The algorithm is compared with the traditional LDA methods, which greatly reduces the computational complexity, the extracted features further reflects the difference between the face and improve the recognition rate and noise pe