Saliency-Detection

首先将二值水印图像经过降维操作转化为一维序列,再将一维序列与m 序列做扩频 调制,最后对音频信号进行分段离散余弦变换,把经过调制的水印信号经过量化处理过程嵌入到离 散余弦变换后的系数中 提取水印时对含水印信号的音频信号也进行分段离散余弦变换,然后利用 原始音频信号提取水印. 实验证明了该技术的鲁棒性和不可感知性.

外国人写的图像融合处理工具以及GUI界面 源代码 相信会对你有所帮助(foreigners write image-processing tools and the integration of the GUI interface believe that the source code will help you)

图像去雾c++版源代码，实现效果还不错，可直接运行。(Fog to remove the c++ version of the source code, the effect is not bad, can be run directly.)

本程序实现了基于m-PCNN的医学图像融合算法，即多通道PCNN图像融合。(use for fusion)

说明：  在用Mean-shift进行跟踪中，采用颜色与纹理相结合的特征空间(Mean-Shift using features of color combined with texture.)

利用极大似然估计进行盲反卷积复原，通过多帧图像约束，效果较好(Use of maximum likelihood estimates for blind deconvolution restoration by the multi-frame image constraint, the better)

叠加平均去噪过程的实现，信号产生，加噪，去噪效果的显示(Average denoising, signal generation, noise addition, display the denoising effects)

说明：  利用IDL编程，使用最小距离法对图像进行监督分类(Using IDL programming, using the minimum distance method to supervise the classification)

数字图像处理中的散度特征空间中的无监督的图像纹理分割(Digital image processing in the feature space of divergence Unsupervised texture segmentation of images)

提出一种基于小波变换的图像配准方法。该方法首先用小波变换提取边缘特征点, 同时构成小波金字塔影像。在最高层利 用角度相关系数确定基准图像和配准图像之间的旋转角度, 经多分辨分级搜索匹配得到最终匹配特征点对, 利用这些特征点对待 配准图像进行配准纠正。实验结果表明, 改进方法有较好的效果, 可用于广泛应用领域。(Proposed a wavelet-based image registration method. In this method, using wavelet transform to extract the edge feature points, the same time constitute wavelet pyramid image. Point at the highest level the correlation coefficient using the baseline image registration between image and the angle of rotation, the multi-resolution hierarchical searching and matching features of the final match point, the use of these features point towards registration image registration correction. Experimental results show that the improved method has better results, can be used for wide range of applications.)