denoiseImageDCT

利用非参数贝叶斯字典学习模型进行图像稀疏表示(use non-parametric-bayesian-dictionary-learning-for-sparse-image-representations)

说明：  本设计是基于MATLAB GUI界面的特效处理系统。配一篇配套论文。探讨了包括色彩调整、代数运算、几何运算、滤镜效果、艺术效果、扭曲效果和风格化七个模块的处理。色彩平衡可以按照特定要求改变图像中每个像素的亮度值；代数运算可以对两幅图象进行代数运算；几何运算可以对图象进行简单的形状处理；滤镜效果是针对相临像素间的关系来处理每个像素，达到一种特殊的效果。艺术效果是可以将图象加工成精美的“艺术品” ；扭曲效果可以对图像进行变形处理；风格化属于破坏性滤镜，通过置换像素生成绘画或印象派的效果。(This design is a special effect processing system based on Matlab GUI interface. With a supporting paper. The processing of seven modules including color adjustment, algebraic operation, geometric operation, filter effect, artistic effect, distortion effect and stylization are discussed. Color balance can change the brightness value of each pixel in the image according to specific requirements; algebraic operation can carry out algebraic operation on two images; geometric operation can carry out simple shape processing on the image; filter effect is to deal with each pixel according to the relationship between adjacent pixels to achieve a special effect. The artistic effect is that the image can be processed into exquisite "artworks"; the distortion effect can deform the image; stylization belongs to destructive filter, which can generate painting or Impressionist effect by replacing pixels.)

基于matlab的加窗傅里叶，使用了hanning窗(Matlab-based windowed Fourier, using the hanning window)

说明：  利用fista算法进行图像复原的matlab源码。(The matlab source code of image restoration using the fista algorithm.)

brox的光流算法第二部分，用于对两帧图像进行光流分析并检测运动物体(Brox optical flow algorithm of the second part, used in optical flow analysis was carried out on the two frames and detect moving objects)

经典的文本水印算法，实现了别人实现不了的功能，希望大家有所用！(The classic text watermarking algorithm to achieve the functions of others fail to realize, I hope you all have a use!)

说明：  对二维码识别进行预处理的二值化，平滑以及角度和畸变的校正等(qr code pre.)

WAV音频的LSB信息隐藏，隐藏文件为二值图像，Matlab编程(WAV audio LSB information hiding, hidden files as binary image, Matlab programming)

基于Curvelet变换的样本块图像修复算法提高现有样本块修复算法性能。首先利用Curvelet 变换估计待修复图像的4方向特征．然后利用颜色信息与方向信息共同衡量样本块间的相似度，在此基础上构造颜色-方向结构稀疏度函数．(Based on the sample block transform image restoration algorithm Curvelet enhance existing sample block repair algorithm performance. Firstly Curvelet transform estimated to be repaired 4 directional characteristics of the image. Then use the color information and direction information together measure the similarity between the sample block, constructed on the basis of color- direction sparsity structure function.)

基于感兴趣区域的图像检索。matlab的，希望对大家有所帮助。(Region of interest based image retrieval. matlab, and want to help everyone.)