资源列表
NPDFaceDetector_Train
- NPD人脸检测训练demo，学习基于NPD特征的DQT，soft cascade结构用来训练和拒绝负样本-NPD face detection training demo, learning-based NPD feature DQT, soft cascade structure for training and rejected negative samples
123456
- 人脸识别，人脸重构，不同角度的人脸图像也能正确识别，包含测试图片-Face recognition, face reconstruction, face images different angles can also be correctly identified, including test images
Desktop
- 用matlab实现二维哈尔小波变换和逆变换程序-Using matlab to achieve two dimensional Haar wavelet transform and inverse transform procedures
m_imadjust
- imadjust 调整图像亮度函数 J IMADJUST(I,[LOW_IN HIGH_IN],[LOW_OUT HIGH_OUT],GAMMA) GAMMA默认值为1（线性映射），GAMMA<1——曲线上凸，GAMMA>1——曲线下凹 图像I与图像f的类型一致，任何类型的图像[LOW_IN HIGH_IN],[LOW_OUT HIGH_OUT]取值均在0—1范围内(计算时函数自己转换)-adjust image brightness function ima
otsu
- 图像二值化处理，利用大津法otsu求取图像的阈值-Binary image processing using the image obtaining Otsu otsu threshold
ASM
- ASM 主动形状模型 提取人脸特征点-ASM，auto shape model
RANSAC
- RANSAC 它是根据一组包含异常数据的样本数据集，计算出数据的数学模型参数，得到有效样本数据的算法-RANdom Sample Consensus
double_view_camera
- 双目视觉视频采集器,可以捕捉视频图像，用于图像处理等方面-double view video catch camera
objpose
- Fast and globally convergent pose estimation video imag这篇论文的算法进行人脸姿态计算-Fast and globally convergent pose estimation video imag.pdf
POINT_OF_PPICTURE
- 图像坐标标点工具，可以对图像进行标点，得出标点坐标-we can use this tool to get the coordinate of the 2D picture by manual pointing。
Traffic-Statistics-Based-on-Video
- 该程序基于视频统计车流量。该算法通过识别出交通场景图像中复合车辆特征的像素，进行图像分割，并依据提取出来的特征来匹配前后两幅图像中的车辆，从而计算车流量。-The program is based on the video statistics vehicle traffic. The algorithm calculates the traffic volume by recognizing the pixels of the composite vehicle feature in the
Digital-Watermark
- 嵌入水印可以看作是在强背景（原始图像）下叠加一个弱信号（水印）。该程序是基于小波变换的数字水印嵌入程序。-Embedded watermark can be seen as a superposition of a weak signal (watermark) in Strong (original image) under. The program is based on wavelet transform digital watermark embedding program.