图像去噪

本文提出了基于二维混沌映射的数字图像水印算法，混沌具有随机性、似噪声及对初始条件的极端敏感性等特点。将经过二维混沌映射置乱后的数字水印信号嵌入图像小波域的低频系数，实现了数字水印的隐蔽性、保密性和稳固性；利用二维混沌映射Arnold变换对水印信号进行置乱，不仅增强了水印信号保密性，同时有效提高了视觉上抵抗图像剪切攻击的能力。(In this paper, two-dimensional chaotic map based on digital image watermarking algorithm, Chaos with random noise and the likelihood of extreme sensitivity to initial conditions and so on. Two-dimensional chaotic map will be scrambling after the digital watermark embedded image signal of the low-frequency coefficients in wavelet domain, the realization of a digital watermark hidden, security and stability the use of two-dimensional chaotic map on Arnold transform scrambling watermark signal, not only enhances the confidentiality of the watermark signal at the same time improve the visual image shear resistance attacks.)

利用R语言编程实现高光谱遥感的随机森林分类(Random Forest Classification Based on Hyperspectral Remote Sensing Using R Language Programming)

网格简化， 将复杂的带有很多细节的物体的三维立体网格进行任意比例的简化，使得其减少点和面片的数量并保持物体的基本特征，以达到加快图像显示及压缩文件的效果。(Mesh simplification, the complex three-dimensional grid with many of the details of the object simplified in any proportion, such that it reduce the number of points and surfaces of the sheet and keep the basic characteristics of the object, in order to achieve the speed up of the image display and the effect of the compressed file .)

影像仿射变换特征提取，结合 sift 等局部特征可以处理较大视角变化下的图像匹配。(This is an implementation of Hessian-Affine detector. )

基于主动学习的高光谱遥感图像分类，主动学习方法也比较流行，新颖。(An Active Learning Approach toHyperspectral Data Classification)

说明：  Arnold 变换与 Gray 码变换相融合的双置乱算法程序，包括了加密和解密程序(The program of double scrambling algorithm combining Arnold transform and gray code transform, including encryption and decryption program)

说明：  基于灰度矩的亚像素边缘检测，零件边缘提取(Subpixel edge contour extraction)

基于颜色的图像检索，将RGB转化成ＨＳＶ的方法(Color-based image retrieval, the method of RGB into HSV)

本程序能够综合地实现图像边缘细化，而且程序够复杂。(This procedure can be integrated to achieve the Image Edge thinning, and the procedures complex enough.)

改程序就代码可以找出从不同角度或区间的两幅遥感影像的同名点(Change the program code can find it from different angles of the same name point or interval of two remote sensing images)