CVA

基于梯度相似度的图像质量评价方法，发表在2012年TIP国际期刊上(A. Liu, W. Lin, M. Narwaria, “Image Quality Assessment Based on Gradient Similarity,” IEEE Transactions on Image Processing, vol. 21(4), pp. 1500-1512, 2012.)

说明：  干涉SAR(InSAR)相位解缠处理代码，包括枝切法及质量图指导法，代码为matlab(Interference SAR(InSAR) phase unwrapping code, including the branch cut method and quality chart guidance method, the code is MATLAB)

使用二维双密度小波2-D Double-Density DWT Thresholding Method，决定图像滤波门限值，实现图像去噪处理。(use of two-dimensional density Wavelet 2-D Double-Density DWT Thresho lding Method, the decision threshold image filtering, image denoising.)

1、读取显示所给PET图像（SE2IM210）。 2、学习imhist函数，画出所给PET图像的直方图。 3、根据直方图为该PET图像选择合适的显示窗（imshow参数），使胸 部轮廓等解剖结构可见。 4、根据直方图选择合适阈值，分割肿瘤区域与正常区域，计算肿瘤区 域内图像灰度均值、方差、最大值及最小值。(1 give PET images displayed, read (SE2 IM210). 2, learning, imhist function, draw the give PET image histogram. 3, according to the histogram for the PET image to choose the appropriate display window (imshow parameters), make bosom Outline of anatomical structures such as visible. 4, according to the histogram, choose appropriate threshold segmentation tumor area and normal area, calculating the tumor area In the domain of image grayscale average and variance, the maximum and minimum values.)

基于图像对比度和波段相关性的波段选择算法，适用于高光谱图像的特征选择(Bands selection algorithm based on image contrast and correlation of bands for hyperspectral image feature selection)

说明：  Hu矩定位图像，可以提取图像的特征，然后进行匹配定位(Hu Moment positioning images, can be extracted image features, and then proceed to match the positioning)

说明：  基于场景的RLS的非均匀校正算法，不同均值滤波器比较。(Scene based RLS nonuniformity correction algorithm, different mean filter comparison.)

迭代收缩阈值算法和快速收缩阈值迭代算法求解Lasso问题(Iterative Shrinkage Thresholding Algorithm (ISTA) for LASSO problem)

说明：  计算机图形学中应用VC++6.0和DirectX 8.0 SDK制作彩色旋转的金字塔和正方体的小程序(computer graphics application VC 6.0 and DirectX 8.0 SDK produced color rotating cube pyramid and the small program)

图像SURF特征提取，用于图像拼接和篡改检测，注释比较详细，程序很好理解(SURF feature extraction, image is used for image matching and tamper detection, comments are detailed, the program is easy to understand)