Three-improved-image-enhancement-

介绍了不同非线性模型在随机共振中的应用，像饱和非线性，阈值非线性等(Describes the different nonlinear model of stochastic resonance in the application, such as nonlinear saturation threshold nonlinearity. . . .)

CVPR2007，论文“Saliency detection: A spectral residual approach”的源码(a code for "Saliency detection: A spectral residual approach,"CVPR,2007.)

基于卡尔曼滤波器的监控系统的设计，实现了对运动目标进行识别，并对其运动轨迹进行预测的功能；当二值化图像超过一定阈值时，判断为有外人入侵，实现报警功能。(Kalman filter based on the design of monitoring system to achieve the recognition of moving objects, and its motion trajectory prediction function when the binary image exceeds a certain threshold, to determine the invasion of outsiders, to achieve alarm function.)

基于彩色信息的道路图象分割,可以有效的去除道路上的阴影.(Based on color information of the road image segmentation, can effectively remove the shadow on the road.)

说明：  简单的数字图像处理程序，实现灰度化，二值化，边缘检测等(Simple digital image processing program, graying, binarization, edge detection, etc....)

表面张力CSF模型，用于计算微重力条件下的两相流问题。(Comtinum Saface tension Force)

说明：  基于BP神经网络的鼠标手势识别算法与实验程序GestureStudy(BP neural network based on the mouse gesture recognition algorithm and experimental procedures GestureStudy)

基于多帧图像插值（Interpolation）技术的方法是SR恢复技术当中最直观 的方法。这类方法首先估计各帧图像之间的相对运动信息，获得HR图像在非均 匀间距采样点上的象素值，接着通过非均匀插值得到HR栅格上的象素值，最后 采用图像恢复技术来去除模糊和降低噪声（运动估计!非均匀插值!去模糊和 噪声）。(In this paper, we propose a novel method for solv- ing single-image super-resolution problems. Given a low-resolution image as input, we recover its high- resolution counterpart using a set of training exam- ples. While this formulation resembles other learning- based methods for super-resolution, our method has been inspired by recent manifold learning methods, par- ticularly locally linear embedding (LLE). Speci?cally, small image patches in the low- and high-resolution images form manifolds with similar local geometry in two distinct feature spaces. As in LLE, local geometry is characterized by how a feature vector correspond- ing to a patch can be reconstructed by its neighbors in the feature space. Besides using the training image pairs to estimate the high-resolution embedding, we also enforce local compatibility and smoothness con- straints between patches in the target high-resolution image through overlapping. Experiments show that our method is very ?exible )

本算法实现了dct基字典的构造并用其进行对图像的消噪。(Algorithm DCT-based dictionary structure and use it for image de-noising.)

目标跟踪的两个算法，分别是运动变化和基于高斯权值的颜色空间HSI的meanshift算法(Two target tracking algorithm, namely, Gaussian-based movement and the right to change the value of the HSI color space meanshift algorithm)