digital-image-homework.doc

用surf和ransac实现两幅图像配准(Registration of two images with surf and RANSAC)

由于前台开发人员经常需要通过RGB值指定各种各样的颜色，而通过这几个图片就可以马上找到某个颜色的RGB值，从而更快地编写网页(Foreground developers often need a variety of colors through RGB values ​ ​ specified by these pictures can immediately find the RGB value of a color, so to write Web pages faster)

光场图像渲染技术，利用稀疏分层算法对图像深度进行分层渲染，再利用颜色一致性约束进行去噪处理，通过所给出的图像阵列渲染出中间各个视点的图像(Optical field image rendering technology, the use of sparse hierarchical algorithm for image depth rendering, and then the use of color consistency constraints for noise processing, through the given image array to render the image of the middle of the various views)

图像处理中比较有用的二维图像光谱分析，Matlab源代码(Image processing more useful two-dimensional image spectra, Matlab source code)

粒子滤波算法，综述，学习文档以及算法程序，包括PF,EKPF,UKPF，以及不同的重采样方法(This file is about algorithm of particle filtering, including codes and tutorial.)

虹膜识别整个的程序源代码包括虹膜内外边界定位、虹膜特征提取等(Iris recognition throughout the program source code, including the inner and outer boundary positioning iris, iris feature extraction)

选择合适的样本特征点，然后可以将特征导入svm进行分类(After the image processing, the main information is obtained by PCA transform, and then the feature of texture information selection is put forward)

纯contourlet变换，已对工具包进行了修改，使之在分解重构时能显示子带系数。(Pure contourlet transformation toolkit has been modified so that when displayed in the decomposition and reconstruction sub-band coefficients.)

用c#和emgucv实现了1.显示图像的直方图，2.直方图均衡化，3.中值滤波，4.sobel算子，roberts算子和Prewitt算子的边缘提取。各算法均自己实现不是调用函数。但是读图片的信息是用的emgucv。希望和同样刚接触emgucv的同学共同进步(With c# and emgucv achieve a 1 shows a histogram of the image, 2. Histogram equalization, 3 median filter, 4.sobel operator, edge roberts operator and Prewitt operator extraction. Various algorithms to achieve their own instead of calling a function. But reading the image information is used emgucv. The same hopes and common progress of students new to emgucv)

LLC做的图像分类算法，非常经典的图像分类算法。属于Bag-of-Feature模型，采用的是SIFT特征描述组。(LLC do image classification algorithm, belonging to the Bag-of-Feature model using the SIFT descriptor group)