tuxiangchuli

VS 环境下利用opencv 实现图像的傅里叶变换，输入图片，保存输出结果。(Under VS environment use opencv to achieve image Fourier transform, the input image and save the output.)

滤波的方法有很多，其中小波滤波很实用，提出了一种二进小波变化的算法。(There are a lot of filtering method, the wavelet filter is very practical, this paper proposes a binary wavelet algorithm.)

使用tensorflow框架搭建神经网络，从单目图像中估计深度值(Using tensorflow framework to build a neural network to estimate depth from monocular images)

用分水岭算法对三维模型进行分割处理。可以实现对三维模型的有意义分割。(Using three-dimensional model of the watershed algorithm to be dealt with separately. Three-dimensional model can achieve meaningful partition.)

增强型Lee滤波，附有详细注释，简单易用，参数可调(enhanced LEE filter)

各种非监督分类遥感算法的IDL代码，包括K均值，ISO聚类(Remote sensing of various non-supervised classification algorithms IDL code, including K-means, ISO clustering)

：近距离目标超高分辨率微波成像技术在安全检测、非破坏性控制、生物医学等领域中有着非常重要的应用 价值和广阔的应用前景。该文利用基于窄脉冲形式的超宽带时域雷达系统研究近距离目标的高分辨率ISAR成像， 给出了仿真和实测结果。利用转台旋转目标，基于超宽带脉冲源和超宽带收发天线以及高性能取样示波器，配以同 步触发脉冲和自行研发的数据采集软件，搭建了该时域雷达实验系统，提出了准确获取目标回波信息的实验条件， 和实现回波延时精确校正的误差补偿方法。并针对采样时间窗内的杂波干扰，讨论了时域后向投影算法(Back Projection，BP)与背景对消技术相结合的成像算法，实现了分辨率为 8 mm 的近距离目标的成像，准确地反映了 目标的位置、形状和大小等信息。( The high resolution imaging of near field targets are interesting for security, non-destructive control and biomedical application. In this paper, high resolution near field ISAR imaging is implemented using Ultra Wide Band (UWB) time domain radar system. Simulation and experiment are both conducted. The target is rotated by turntable. The system is built based on the UWB pulse source, UWB antennas, high performance sampling oscillograph, synchronous trigger pulse and proper motion study data acquisition software. Experiment condition and delay errors compensation method are presented to obtain accurate echo information of the target. And Back Projection (BP) combined with background cancel algorithm is discussed considering the clutter interference in time window. )

说明：  计算两个离散变量x和y的联合熵z = H（x，y）。 输入：x，y：两个相同长度的整数向量输出：z：联合熵 可以使用，有利于学习目的(Compute joint entropy z=H(x,y) of two discrete variables x and y. Input: of x, y: two integer vector of the same length Output: z: joint entropy good for learning purposes)

使用adaboost方法进行人脸检测,图象库自己添加。(Face detection with adaboost method.)

说明：  实验室自己用的fbp重建算法，基于C语言，亲测可用。上有个人C盘名称信息请删除(Ct reconstruction fbp)