compresion-dwt-sans-psychoachostique

针对高精度的畸变图像,提出了一种基于最小二乘影像匹配的高精度畸变图像矫正算 法. 算法首先利用特征提取与边缘检测对图像进行预处理,并且将特征匹配与最小二乘算法相 结合,从而实现了图像与模板之间精确的子像素定位与匹配. 实验表明,该算法较好的解决了目前高精度畸变图像矫正算法中普遍存在的定位和匹配精度较差的缺陷,图像矫正效果良好,是一种有效的畸变图像矫正算法.(Image distortion for high-precision, a least-squares image matching based on high-precision image distortion correction algorithm. First of all, the use of feature extraction algorithm for edge detection and image pre-processing, and will feature matching with a combination of least-squares algorithm in order to achieve the image and the template between the sub-pixel accurate positioning and matching. The experimental results show that the algorithm has better accuracy of the solution to the current distortion prevalent image correction algorithm to match the accuracy of the positioning and less defects, image correction effect well, is an effective algorithm to correct image distortion.)

二维方位角/仰角的music算法DOA估计程序(2D azimuth/elevation to the music algorithm DOA estimation procedures)

SMO算法的Matlab实现，花了很大的力气才弄出来的！和大家分享！更多就在我的主页访问吧：http://huangbo929.blog.edu.cn(SMO algorithm Matlab, spent a lot of time and energy only get out! And to share with you! More on the home page I visit it : http://huangbo929.blog.edu.cn)

《MATLABSimulink通信系统建模与仿真实例精讲》IS95前向链路系统(" MATLABSimulink communications system modeling and simulation examples succinctly" IS95 link to the system)

主路径信号的前导码，主要包含功能信号和基准时间信号。(The preamble of the main-path signal)

该书籍介绍了很好的统计学习方法，能为机器学习和数据挖掘的爱好者提供很好的理论依据。(The book describes the good statistical learning method that can provide a good theoretical basis for machine learning and data mining enthusiasts.)

介绍多任务学习MALSAR工具箱，包含function，example(Introduce multi-task learning MALSAR toolbox, including function, example)

MATLAB地震数据处理软件。主要设计线性Rp与非线性Rp 、多分量叠加速度成像分析，以及反射系数等代码，处理占一部分(Seismic processing software)

基本的调制解调程序集(包括BPSK,QPSK,QAM,GMSK等)(Modem basic assembly (including BPSK, QPSK, QAM, GMSK, etc.))

说明：  逼近线性模型中Ｍ估计分布的随机加权自助法，在一般的条件下证明了这种方法是强有效的。 (Approximate linear model estimated the distribution of M random weighted bootstrap, in general terms that this approach is strong and effective.)