wavelet

压缩感知中利用增广拉格朗日方程解最小稀疏正则化的恢复算法(DAL solves the dual problem of (1) via the augmented Lagrangian method (see Bertsekas 82). It uses the analytic expression (and its derivatives) of the following soft-thresholding operation， which can be computed for L1 and grouped L1 (and many other) sparsity inducing regularizers. If you are interested in our algorithm please find more details in our technical report or in my talk at Optimization for Machine Learning Workshop (NIPS 2009).)

绘制电脑中原始人声信号的时域、频域图，并在其上加入高斯白噪声，之后设计巴特沃兹滤波器将其滤除。(The draw primitive people in the computer sound signal time domain, frequency domain, and join in on the white Gaussian noise, after design Butterworth filter to filter out.)

说明：  基于视觉特征的图像特征提取，在HSV颜色空间提取彩色纹理特征(Based on the visual characteristics of image feature extraction, in the HSV color space, color texture feature extraction)

实现基于迭代反投影算法的超分辨率重建，对于没有噪声的情况有较好的效果。(Iterative Back Projection super-resolution.)

基于图割的图像分割的实现代码，在linux下测试通过(efficient graph cut source code，run in linux)

张量分析(第二版)清晰版—黄克智，一本很好的对张量进行操作的很好的书籍(operation tensor book)

关于图像对称性修复的资料，通过图象结构的对称性、应用图象的全局信息。提出了一种简单的图象修复算法(About image symmetry repair material )

说明：  二维、三维图像分割工具箱，采用基于matlab的标记分水岭算法(Two-dimensional and three-dimensional image segmentation toolbox, mark watershed algorithm based on MATLAB)

说明：  本课题为基于MATLAB的小波变换dwt和离散余弦dct的多方法对比数字水印系统。带GUI交互界面。有一个主界面GUI，可以调用dwt方法的子界面和dct方法的子界面。流程包括，读取宿主图像和水印图像，嵌入，多种方法的攻击（剪切，加噪，旋转等），提取，最后利用psnr峰值信噪比进行评价不同攻击下，鲁棒性的好坏。(This project is based on MATLAB wavelet transform DWT and discrete cosine DCT multi method contrast digital watermarking system. With GUI interactive interface. There is a main interface GUI, which can call the sub interfaces of DWT method and DCT method. The process includes: reading host image and watermark image, embedding, attacks of various methods (cutting, adding noise, rotation, etc.), extracting. Finally, PSNR peak signal-to-noise ratio is used to evaluate the robustness of different attacks.)

包括canny，sobel，log等边缘检测的改进算法，使用效果良好(Including improved edge detection algorithms such as canny, sobel, log and so on, the application effect is good.)