TDSR-LATEST-CODE

S-curve simulation in matlab

针对基于小波变换的红外图像增强方法视觉效果不够理想的缺点, 提出了一种基于平稳小波变换和Retinex 的红外图像增强方法, 利用Retinex 增强算法增强图像的视觉效果, 并改善其亮度均匀性。首先, 对红外图像经平稳小波变换后的最大尺度低频子带图像进行多尺度Retine x 增强 然后, 利用贝叶斯萎缩阈值法对高频子带图像进行阈值去噪, 并根据低频子带图像的局部对比度和模糊规则计算高频子带的增益系数, 从而得到增强后的高频子带图像 最后, 由低频子带图像和高频子带图像重构得到增强后的图像。针对大量图像进行了实验和增强效果的定性与定量评价, 并与双向直方图均衡法、二代小波变换法、Curvelet 变换法和多尺度Retinex法作了比较。结果表明, 所提出的方法增强了图像细节, 抑制了噪声, 并明显改善了图像的整体视觉效果。(As infrared image enhancement method based on wavelet transformation has the problem of unperfect visual effect , and Retinex enhancement algorithm can enhance visual effect of the image by improving brightness uniformity , amethod based on stationary wavelet transformation and Retinexis proposed.Firstly , the infrared image is decomposed into high-frequency detail and low-frequency approximation components at various resolutions, and the low frequency subband image of the largest scale is enhanced by multiscale Retinex algorithm.Then, the highfrequency subband images are denoised by Bayesian shrinkage method, and the gain coefficients of high-frequency subbands are available by calculating the local contrast of the enhanced low-frequency subband based on fuzzy rules to get the enhanced high-frequency subband images .Finally, the enhanced image is reconstructed by the low-frequency subband and high-frequency subbands .Experiments with qualitative and quantitative uation are carried ou)

逆合成孔径雷达（ISAR）具有远距离、全天候、全天时工作等优点，目前通过ISAR成像来进行目标识别已得到广泛的应用。ISAR的观测对象是非合作目标，因此ISAR成像具有深远的国防应用背景，受到各国的重视，是竞争激烈、发展迅速的技术领域。(Inverse Synthetic Aperture Radar (ISAR) has a long-range, all-weather, all-time work, etc, currently conducted by ISAR imaging for target identification has been widely used. ISAR target non-cooperative target observation, it has far-reaching defense ISAR imaging application background, attention all, is highly competitive, rapidly developing technology.)

SOM Algorithm in Data Mining

CIC补偿滤波器的介绍和设计方法，包含MATLAB程序设计方法。(CIC compensation filter and design methods introduced, including MATLAB programming method.)

说明：  pso算法的matlab实现中有很多共同点，该工具箱对一些共性建立函数，以后可以直接调用(The Tool Box)

使用同态滤波对图像进行对比度增强的处理，本代码仅适用于灰度图像。(Homomorphic filtering using the contrast of the image enhancement processing, the code applies only to grayscale.)

利用MCMC来取柯西分布的随机数，使用的是M-H采样。(MCMC to take advantage of the Cauchy distribution random number, using MH sampling.)

短时傅里叶，matlab程序，经过检验可以正常使用，(stft short-time ft)

This program compress and recostruct using wavelets. We can select level of decomposition(here maximum 4 levels are given) of images using selected wavelet. For eg:-wavelets can be haar, db1, db2,dmey............... Decomposition can be viewed in figure. (Please note that select 256X256 image for better result.) Then compression can performed, PERFL2 give compression score. Then reconstruction can be performed. Each decompsition we can choose different threshold values. For each threshold value we can calculate mse,psnr,pq(picture quality), bit ratio etc. To get pq install pqs function . (This program compress and recostruct using wavelets. We can select level of decomposition (here maximum 4 levels are given) of images using selected wavelet.For eg:-wavelets can be haar, db1, db2, dmey ........ ....... Decomposition can be viewed in figure. (Please note that select 256X256 image for better result.) Then compression can performed, PERFL2 give compression score.Then reconstruction can be performed.Each decompsition we can choose different threshold values . For each threshold value we can calculate mse, psnr, pq (picture quality), bit ratio etc. To get pq install pqs function.)