首页 考试资料幻灯片工程技术公务员考试小学教学中学教学大学教学外语资料
97哈尔滨理工大学本科毕业设计(论文)模板2015修改版


哈尔滨理工大学学士学位论文

基于机器学习的人脸检测系统 摘 要

随着互联网在当今世界的应用越来越广泛,各类人机交互的智能方向 层出不穷,机器学习应运而生。机器学习,是以一种方式来赋予机器以学 习的能力,它的实践意义是通过训练数据,建立模型,然后使用模型进行 预测的一种方法。人脸检测是机器学习的重要应用之一,近年来,基于机 器学习的人脸检测系统逐渐成为了研究热点,受到了广泛的关注。 在人脸检测方面,通过使用支持向量机和模板匹配模型相结合的方法 实现人脸检测。支持向量机算法是一个比较经典的分类算法,其广泛的应 用于回归分析和统计分类中,是一种基于机器学习的方法。支持向量机算 法首先将人脸像素作为分类器的输入,其次进行判断,判断区域内是否存 在人脸,但是由于已知训练图像的尺度往往是固定的,可是检测过程做不 到尺度不变,所以需要对多种尺度的图像进行检测,从而导致训练规模过 大。模板匹配模型的基本思想,在成功检测到人脸并准确定位面部特征点 之后,人脸的大致区域就可以被裁剪出来。准确提取人脸面部特征点之 后,与库存的已知人脸进行对比,根据一个已知的人脸模型与待检测人脸 图像匹配,寻找到正确的匹配位置,完成最终的分类。 另外,在人脸检测的特征提取过程中,本文采用了 PCA 和 LPP 降维 提取人脸图像的代数特征的算法,实现了对人脸图像数据的训练及分类, 这是本文的重点研究内容之一。 本文采用了 java 语言编写、OpenCV 框架搭建了人脸检测实验系统,实 现了人脸检测技术,为 SVM 分类器和 PCA、LPP 算法提供了技术支持。 关键词 机器学习;人脸检测;人脸识别;PCA;LPP

-I-

哈尔滨理工大学学士学位论文

Face Detection System Based on Machaine Learning Abstract
With the application of Internet in the world more and more widely, the intelligent directions of human-computer interaction emerge in endlessly, and machine learning emerges as the times require. Machine learning is a way of giving machines the ability to learn, and its practical significance is through training data, building models, and then using a model to predict. Face detection is one of the important applications of machine learning. In recent years, face detection based on machine learning has become a hot research topic and has attracted extensive attention. In face detection, face detection is achieved by combining support vector machine and template matching model. Support vector machines (SVM) is a classical classification algorithm, which is widely used in regression analysis and statistical classification. It is a method based on machine learning. Support vector machine algorithm will first face pixels as the input of the classifier, then judge whether there is a face judgment within the region, but due to the known training image scale is always fixed, but the detection process do not scale invariant, so it is necessary to detect the image of a variety of scales, resulting in large scale training. The basic idea of template matching model is that after the successful detection of the face and accurate location of facial feature points, the approximate area of the face can be cut out. After the accurate extraction of facial feature points, compared with the known face inventory, according to a known face model and the detected face image matching, matching to find the correct position, the completion of the final classification.
- II -

哈尔滨理工大学学士学位论文

In addition, in the face detection feature extraction process, this paper uses PCA and LPP dimensionality reduction to extract algebraic features of face image algorithm, the training and classification of face image data, which is one of the key research content of this paper. In this paper, we use java language and OpenCV framework to build the human face detection experiment system, and realize the face detection technology, which provides technical support for SVM classifier and PCA and LPP algorithm. Keywords machaine learning, face detection, face recognition, PCA, LPP

- III -

哈尔滨理工大学学士学位论文

摘要 ...................................................................................................................... I Abstract ............................................................................................................... II 第 1 章 绪论 ........................................................................................................ 1 1.1 课题背景 ................................................................................................... 1 1.1.1 研究意义 ............................................................................................ 1 1.1.2 国内外研究现状分析 ........................................................................ 3 1.1.3 论文研究的主要内容 ........................................................................ 5 第 2 章[单击此处输入标题] ........................................... 错误!未定义书签。 2.1 基本结构 ................................................................. 错误!未定义书签。 2.1.1 原理分析 .......................................................................................... 10 2.2 本章小结 ................................................................................................. 11 第 3 章[单击此处输入标题] ........................................................................... 12 3.1 方案对比研究 ......................................................................................... 12 3.1.1 方案一 .............................................................................................. 12 3.2 本章小结 ................................................................................................. 12 第 4 章[单击此处输入标题] ........................................................................... 13 4.1 结果分析 ................................................................................................. 13 4.1.1 方案一 .............................................................................................. 13 4.2 本章小结 ................................................................................................. 13 结论 .................................................................................................................... 14 致谢 .................................................................................................................... 15 参考文献 ............................................................................................................ 16 附录 A ................................................................................................................ 18 附录 B ................................................................................................................ 19 附录 C ................................................................................................................ 21

12345

 


 

  【Top

最新搜索

 

哈尔滨理工大学本科毕业设计(论文)模板2015修改版 - 哈尔滨理工大学学士学位论文 基于机器学习的人脸检测系统 摘要 随着互联网在当今世界的应用越来越广泛,各类人机交互...

哈尔滨理工大学本科毕业设计(论文)模板2017 - 哈尔滨理工大学学士学位论文 基于 GPS 的智能公交报站系统 摘 [单击此处输入中文摘要] 关键词 电机;电动机;磁场;故障...