rat

图像识别，包括人脸识别和虹膜识别，在数字图像中虹膜位置的有效定位是虹膜识别的关键问题。用一种基于主动轮廓线模型的方法定位虹膜的位置，先用灰度投影法检测出瞳孔内的一点作为瞳孔的伪圆心，该圆心只要能落在瞳孔内部即可。(Image recognition, including face recognition and iris recognition, iris position in the digital image of the iris recognition and effective positioning is the key issue. Using a model based on active contour method of positioning the iris position, the first gray projection method used to detect a point within the pupil as a pupil of the pseudo-center of a circle, the center of a circle can be as long as it falls within the pupil.)

通过高斯核密度估计计算多元变量之间的互信息熵(The mutual information entropy between multivariate variables is calculated by Gaussian kernel density estimation)

基于人类视觉系统的小波域自适应图像水印算法研究(HVS DWT watermarking)

Code for make histogram of image

otsu s multilevel thresholding method

说明：  基于hessian矩阵的FrangiFilter，用于图像增强,常用于影像图像血管增强(FrangiFilter based on hessian matrix, which is used for image enhancement, is often used for image enhancement of blood vessels.)

一份retinex算法的code，其中利用pca 做 色度分解，利用改进的bilateral滤波器进行低频和高频分解(Retinex algorithm in a code, which made use of PCA colority decomposition, the use of improved bilateral filter for low frequency and high-frequency decomposition)

blkproc.m,DCHuffmanEncoding.m,ACHuffmanEncoding.m,zigzag.m,PSNR.m, 对灰度图像进行DCT变换,量化,ZigZag扫描,Huffman编解码,反量化,反DCT变换而重建图像。 其中，blkproc.m为分块DCT变换函数； DCHuffmanEncoding.m，ACHuffmanEncoding.m分别为DC和AC变换系数的Huffman码表函数； zigzag.m为ZigZag扫描函数；PSNR.m为求图像峰值信噪比函数。 Copyright 2008 Reserved @ Wang Chengyou @ Tianjin University, P.R.China. (This function tests the DCTJPEG codec)

自己用matlab写的灰度图像压缩解压程序 jpegencoder可以打开图像压缩后存储为c.jpg（只限于灰度图像） jpegdecoder只适用于把jpegencoder的c.jpg解压，还不可以对任意图像解压，有待以后完善 作者 ar_dong(jpegencoder jpegdecoder author ar_dong)

说明：  一个用VC++6.0编的程序例子，实现远程屏幕图像的传输 (a VC 6.0 Provision procedures example, remote screen image transmission)