搜索资源列表
UFR2.0
- 通过人脸图像归一化,人脸检测,特征提取等处理后,实现人脸识别功能-Through the Human face image normalization, face detection, feature extraction, such as treatment, the realization of face recognition feature
face
- 完整的表情识别系统一般包括人脸表情图像捕获、预处理、人脸检测与定位、 人脸分割与归一化、人脸表情特征提取、人脸表情识别。本文着重研究了人脸表 情特征提取、特征选择及表情分类等关键问题,并提出了一些改进的方法,同时 进行了仿真实验-Complete expression recognition systems typically include facial expression image capture, preprocessing, face detection and loca
Face_recognition
- 人脸识别程序。算法部分目前分为4个模块:人脸对齐、光照归一化、特征提取和选择、子空间降维,每个模块是一个项目,每个项目生成一个dll供功能部分隐式调用-Recognition program. Part of the algorithm is currently divided into four modules: face alignment, illumination normalization, feature extraction and selection, subspace dime
nensai
- 随机调制信号下的模拟ppm,数据模型归一化,模态振动,小波包分析提取振动信号中的特征频率。- Random ppm modulated analog signal under Normalized data model, modal vibration, Wavelet packet analysis to extract vibration signal characteristic frequency.
nanfai
- 包含优化类的几个简单示例程序,用于信号特征提取、信号消噪,数据模型归一化,模态振动。- Optimization class contains several simple sample programs, For feature extraction, signal de-noising, Normalized data model, modal vibration.
6582
- 用于信号特征提取、信号消噪,数据模型归一化,模态振动,插值与拟合的matlab实现。- For feature extraction, signal de-noising, Normalized data model, modal vibration, Interpolation and fitting matlab implementation.
qi114
- 数据模型归一化,模态振动,用于信号特征提取、信号消噪,基于混沌的模拟退火算法。- Normalized data model, modal vibration, For feature extraction, signal de-noising, Chaos-based simulated annealing algorithm.
NormalizeFea
- 特征归一化matlab程序,可行归一和列归一(Characteristics of Matlab procedures, and the column belongs to a possible normalization)
pfh
- 使用pfh算法,计算点云特征,计算pfh特征集的最大值,将pfh值进行归一化处理(The PFH algorithm is used to compute the feature of point cloud, and the maximum value of the PFH feature set is calculated, and the PFH value is normalized)
Source code
- 在opencv上实现双目测距主要步骤是: 1.双目校正和标定,获得摄像头的参数矩阵: 进行标定得出俩摄像头的参数矩阵 cvStereoRectify 执行双目校正 initUndistortRectifyMap 分别生成两个图像校正所需的像素映射矩阵 cvremap 分别对两个图像进行校正 2.立体匹配,获得视差图: stereoBM生成视差图 预处理: 图像归一化,减少亮度差别,增强纹理 匹配过程: 滑动sad窗口,沿着水平线进行匹配搜索,由于校正后左右图片平行,左图
AllFeatureAllFeature
- 提取图像的梯度特征,计算梯度边缘的数字,经归一化后进行特征提取(Gradient feature of image extraction)
红酒分类-点-2归一化后
- 将三类不同红酒进行自动分类,其中每种酒具有13类特征(Three kinds of different red wine are classified automatically, each of which has 13 types.)
iris code
- 利用线性Hough变换和Canny边缘探测得到的所输入虹膜图像上各点的坐标,最终通过归一化产生一个虹膜的特征模板。(The coordinates of each point on the iris image are obtained by using the linear Hough transform and the Canny edge detection. Finally, a feature template of the iris is generated by the norma
Harris
- 基于离散分数布朗随机场模型的水下图像目标检测方法。该方法根据分形理论和水下图像的特点,以图像中每 个像素点为中心取窗口,计算在该窗口内的分形维数均值,将该均值作为中心像素的分形特征,然后根据分形维 数分布图确定分割阈值,从而实现对水下图像分割,并且通过将目标表面不同尺度下的灰度差分平均值进行归一 化处理,减少了用于表示不同尺度下的平均绝对值灰度差分的数据,从而提高算法检测效率(Underwater target detection method based on discrete frac