gkde

实现一个完整的OFDM系统，可以显示频谱图。(OFDM systems to achieve a complete, can be displayed spectrum.)

分为行人检测，特征提取，行人跟踪三步，其中最主要的代码也是这个代码的中心是行人跟踪这一部分，我采用的是kalman滤波器进行行人下一帧位置的预测，可以很好的提高行人跟踪的鲁棒性，跟踪结果比较准确。(Into pedestrian detection, feature extraction, pedestrian tracking steps, which is the center of the main code of the code is pedestrian tracking this section, I kalman filter is used to predict the position of the next frame pedestrians, can be a good improvement pedestrian tracking robustness, tracking results more accurate.)

matlab OTSU 简易的大津算法 求最佳阈值的。(matlab simple algorithm seeking Otsu threshold.)

以MSTAR为实验数据，对SAR影像进行杂波统计分析研究。利用高斯、瑞利、对数正态、LN、gamma函数模型对其拟合，通过K-S方法检验拟合精度(The MSTAR the experimental data, statistical analysis of SAR images of clutter. Gaussian, Rayleigh, lognormal, LN, gamma function model to its fitting, fitting by means of checking the accuracy of KS)

该matlab代码实现了plsa算法。plsa全称是概率隐语义分析，是一个非常重要的隐语义分析模型。(This matlab code implements plsa algorithm, which is the short of probability latent semantic analysis and is a very important latent semantic analysis model.)

用IDl编写的，用IDL图像处理。已调试好的程序。(Prepared using IDL with IDL image processing. Debugging process has been good.)

说明：  find_center:寻找图像的形心，即图像几何中心点的检测。 otsu:otsu分割 sobelNedge：sobel边缘检测 catch_point:全局阈值分割函数(Find center: find the centroid of the image, that is, the detection of the geometric center of the image. Otsu: Otsu segmentation Sobel edge: Sobel edge detection Catch point: global threshold segmentation function)

最新的用matlab实现的ＳＰＩＨＴ算法，实现对图像的高性能压缩。(Matlab using the latest achievement of SPIHT, the realization of high-performance image compression.)

说明：  matlab实现帧间差分提取运动目标，包含测试视频和运行效果。在背景固定的视频中效果较好。附赠lena图片以供CV学习。(Matlab achieves inter-frame differential extraction of moving objects, including testing video and running effects. It works well in video with fixed background. Enclosed Lena pictures for CV learning.)

改程序是本人编写的，可以运行，采用的是yale人脸数据，采用4*4分块，无权值，统一旋转不变模式(The program is I write, can run, using Yale face data, using the 4* 4 block, has no right to value, uniform &rotation invariant mode)