bianyuanjiance

《基于计算机视觉的普适性三维重建方法研究》源代码，实现了一种将二维图像重建成三维场景的方法。(Source code of <General Sense Three-Dimension Reconstruction Research Based on Computer Vision>, a technique which can convert 2D images into 3D sence.)

说明：  静止背景下运动目标检测，采用帧间差分的方法进行运动目标识别(Static background motion detection, frame difference methods using Moving Object Recognition)

SIFT算法，可以有效的找到不同画面中的对应特征点，在跟踪和图形识别方面有很好的用途(SIFT algorithm, can effectively find a different screen the corresponding feature points, in the tracking and pattern recognition as a very good use of)

代码使用全相位双正交变换（APBT）稀疏表示图像，主函数为test.(The code uses the all phase biorthogonal transform (APBT) sparse representation of the image, the main function for the test.)

利用Hough变换，可以检测图像中的多条线段，效果很好(using hough transform for line detection)

有关纹理的图像检索算法，利用双树复小波提取图像的纹理特征(Texture image retrieval algorithm using tree complex wavelet extraction image texture features)

基于图像特征的模板匹配算法的研究， 进行图像处理地子程序；(Template based on image feature matching algorithm for image processing to Subroutine )

跟踪目标的算法，camshift,meanshift.(Target tracking algorithm, camshift, meanshift.)

实现shift算法，进行精确的图像识别，可以抵抗各种旋转、平移、缩放等仿射变换的攻击(Realize shift algorithm, an accurate image recognition, you can resist all kinds of rotation, translation, scaling affine transformation attack)

一个基于Gabor滤波器组的多特征融合的图像检索算法（CBIR based on Multi-Features Combining Multi-parameters from Gabor Filter-bank, 简称multi法），融合了基于Gabor 小波系数的标准方差，峰态，能量，平滑度和一致性度量五个特征量度，实验表明，该算法是一种更加有效的图像检索方法。(we propose a CBIR algorithm based on Multi-Features Combining Multi-parameters from Gabor Filter-bank(called multi algorithm for short), which use Gabor filter banks and combine five features including standard deviation, kurtosis, energy, smoothness and uniformity together. The results of experiments show that this algorithm is a more effective method used for image retrieval.)