Description: 智能图像/视频处理中，复杂背景环境（比如室外环境、机场、车站等）下，人脸识别的第一步是人脸的检测。它的精确度直接影响到后期识别的结果。不过，领域内的科学家们基本上很难有足够的精力和时间开发优化的C++代码，使其用于商业用途，而一般都是只在Matlab中进行模拟。
本文的目的是提供一个我开发的SSE优化的，C++库，用于人脸检测，你可以马上把它用于你的视频监控系统中。文章中的分类器的训练数据来自与我的 webcam图像，它们被采集于不同时间，不同光照，不同背景环境下，它几乎可以实时地检测出我（的脸：）。训练的非人脸数据来自对不同背景的采集，用的是同一个webcam。被提取出的人脸区域，已经经过下面的处理：高斯滤波，直方图均衡化。
如果你需要更精确的结果，请从internet上下载更多不同的人脸集合，然后从新训练分类器。和我的库中一样尺寸的公共库是CBCL，其库超过100MB，所以，请大家自己下载楼-Intelligent image/video processing, complex background environment (such as an outdoor environment, airports, stations, etc.), the first step is the recognition of face detection. It directly affects the accuracy of the latter part of the identification results. However, scientists in the field are basically difficult to have enough energy and time to develop optimized C++ code to be used for commercial purposes, and are generally only be simulated in Matlab. The purpose of this paper is to provide an optimized SSE my development, C++ library for face detection, you can immediately use it for your video surveillance system. Face article classifier training data with my webcam images, which are collected at different times, in different lighting, different background environment, it is almost real-time detection of me (a :). Training of non-face data collection different backgrounds, with the same webcam. Was extracted face region, has been subjected to the following treatment: Gaussi Platform: |
Size: 15348736 |
Author:周文活 |
Hits:
Description: 为了实现复杂环境下的人脸特征有效表达，提出一种改进的梯度方向直方图（ＨＯＧ）人脸识别方法．首先以人脸图像网格作为采样窗口并在其上提取 ＨＯＧ特征；然后将所有网格 ＨＯＧ特征向量进行组合，实现整个人脸特
征表达；最后采用最近邻分类器进行识别．另外，比较了该方法与Ｇａｂｏｒ小波和局部二值模式（ＬＢＰ）２种著名的人脸
局部特征表示方法的优劣．实验结果表明，在调优的 ＨＯＧ参数下，在具有光照和时间环境等复杂变化的ＦＥＲＥＴ人
脸库中，较少维数的 ＨＯＧ特征比ＬＢＰ特征有更好的表现，而且 ＨＯＧ特征提取时间和特征向量维数比Ｇａｂｏｒ小波方法更具有优势-In order to achieve facial features in complex environments valid expression, an improved gradient direction histogram (HOG) face recognition method. Firstly face image and extract the grid as a sample window HOG features on it then all mesh HOG feature vector combination, realize the whole people express facial feature Finally, nearest neighbor classifier to identify. In addition, the comparison of the method with Gabor wavelet and local binary pattern (LBP) 2 famous facial features indicate the quality of the local approach. Experimental results show that HOG parameter tuning in FERET face with complex changes in the environment of light and time, the characteristic dimension of less than HOG LBP features better performance, and feature extraction time and HOG dimension of feature vectors have an advantage over Gabor wavelet method Platform: |
Size: 1274880 |
Author:wang |
Hits:
Description: 基于opencv3.0的SVM分类器源码（可直接运行），在人脸识别类的领域。-Based on the SVM classifier opencv3.0 source (can be run directly), in face recognition field. Platform: |
Size: 1024 |
Author:以然 |
Hits:
Description: 基于opencv3.0的非线性SVM分类器源码（可直接运行），在人脸识别类的领域。-Based on nonlinear SVM classifier opencv3.0 source (can be run directly), in face recognition field. Platform: |
Size: 2048 |
Author:以然 |
Hits:
Description: 基于opencv的使用python写成的人脸识别程序，里面包含各种haar分类器-Based on the use of Python opencv written in the face recognition program, which contains a variety of Haar classifier Platform: |
Size: 1774592 |
Author:ivan |
Hits:
Description: 可以demo的KNN分类器，对模式识别十分重要的作用，有着较好的分类效果，可以帮助新手更好的理解KNN原理，对人脸识别有着很好的演示作用。-Can demo KNN classifier, the pattern recognition is an important role, has a good classification effect, can help beginners a better understanding of KNN principle, has a very good demonstration effect on face recognition. Platform: |
Size: 3072 |
Author:赵礼 |
Hits:
Description: 可以分别设置度量距离的KNN分类器，有欧式和马氏距离。对模式识别十分重要的作用，有着较好的分类效果，可以帮助新手更好的理解KNN原理，对人脸识别有着很好的演示作用。-Distance can be set respectively KNN classifier, style and markov distance. For pattern recognition is an important role, has a good classification effect, can help beginners a better understanding of KNN principle, has a very good demonstration effect on face recognition. Platform: |
Size: 5120 |
Author:赵礼 |
Hits:
Description: 可以分别介绍度量距离的KNN分类器英文论文，有欧式和马氏距离。对模式识别十分重要的作用，可以帮助新手更好的理解KNN原理，对人脸识别有着很好的演示作用。-KNN classifier can respectively introduces measure distance English papers, style and markov distance. For pattern recognition is an important role, can help beginners a better understanding of KNN principle, has a very good demonstration effect on face recognition. Platform: |
Size: 194560 |
Author:赵礼 |
Hits:
Description: 基于opencv自带的Haar特征分类器的人脸识别，可识别完整的人脸并用圆圈圈出。-Based opencv comes Haar classifier recognition feature, you can identify the full face and a circled. Platform: |
Size: 1024 |
Author:bangbang |
Hits:
Description: 基于LDA线性辨别分析的人脸识别算法，采用KNN分类器，可直接运行，自带数据库，识别率有88 。-LDA face recognition algorithm based on linear discriminant analysis, using KNN classifier, can be directly run, comes with a , the recognition rate of 88 . Platform: |
Size: 4245504 |
Author: |
Hits:
Description: 这是一个SVM分类器，可以实现对人脸的识别，里面有数据库。用的时候要先添加路径。-This is a SVM classifier, can be achieved on the face recognition, which has a . Use the time to add the path. Platform: |
Size: 9829376 |
Author:HUAILI |
Hits:
Description: 基于dsp的人脸识别，使用harr特征和adaboost级联分类器(Based on dsp face recognition, use the harr feature and the adaboost cascade classifier) Platform: |
Size: 3505152 |
Author:devil00
|
Hits:
Description: 实现人脸识别，分类器用svm,功能非常好，改名易于理解，适合初学者(The realization of face recognition, the classifier uses SVM, the function is very good, the Rename is easy to understand, suitable for the beginner.) Platform: |
Size: 145408 |
Author:lyq0209
|
Hits:
Description: 利用KNN分类器在FERET人脸库上进行单样本人脸识别(Single sample face recognition on FERET face database using KNN classifier) Platform: |
Size: 32055296 |
Author:wenjiali123
|
Hits:
Description: 主要功能为脸部识别，有相关分类器标记特征点之间的距离(The main function is face recognition, and the distance between feature points with related classifier marks.) Platform: |
Size: 71682048 |
Author:muzine1995 |
Hits:
Search - face recognition classifier