jpg_encoder_and_decoder_matlab

评估指标为点目标距离向和方位向峰值坐标、峰值旁瓣比、一维积分旁瓣比、剖面图(The evaluation indexes are peak coordinates, peak sidelobe ratio, one-dimensional integral sidelobe ratio and profile of point targets in range and azimuth directions.)

图像边缘检测算法，采用小波图像边缘检测的算法实现边缘检测(Image edge detection algorithm, using wavelet edge detection)

Computer Vision:Algorithms and Applications，Richard Szeliski，适合图像处理人群(Computer Vision: Algorithms and Applications, Richard Szeliski, suitable for image processing crowd)

Jianchao Yang 的基于稀疏表示的单幅图像重建的原始代码，先将高低训练图像分块，再将块训练成高低字典，将测试图像映射到低字典上，得到系数，再乘以高子典就得到最后的图像。对学习超分辨率同学的参考作用很大。(This is the original matlab code for super resolution by Jianchao Yang 。The method is sparse represent based on the overcomplete dictionary。)

说明：  三维受激布里渊散射谱使用MATLAN仿真(Stimulated Brillouin Scattering)

Relief-F 算法是由 Kononenko 在 Relief 算法基础上扩展得到的，可以用来解决多分类与回归问题。 Relief-F 特征选择算法利用特征与类别间的相关程度判别特征的敏感程度，敏感的特征能使同类样本靠近，非同类样本远离。基本思想是： 先随机从总样本集 D中获取一个样本 R， 然后在 与 R 同类样本集中和与R 异类 样 本 集 中 各 获 取 k 个近 邻 样 本， 在 特 征 集 F中如果某个特征上异类样本的距离大于同类样本，则这个特征是有利于分类的，增加其权重，重复 n 次 取所有结果的均值作为每个特征的最终权重值.

压缩包中的算法使用moment matching和MLE估计矩阵的广义高斯分布参数，推荐一下。(Compression package using the moment matching algorithms and the estimated matrix of MLE parameters of generalized Gaussian distribution, recommend.)

图像检测的算法，Faster R-CNN算法，先对整张图像进行卷积计算，然后通过感兴趣区域池化层(RoI Pooling Layer)将选择性搜索算法推荐出来的候选区域和卷积网络计算出的特征映射图进行融合，得到候选区域对应的特征矢量，这种共享卷积计算的操作极大地减少了卷积计算的次数。而且这些特征矢量的维度统一，方便后续的分类工作。通过感兴趣区域池化层处理卷积特征，并将得到的特征送往两个并行计算任务进行训练，分类和定位回归。通过这些方法和改进的框架，Fast R-CNN 用更短的训练和测试时长，取得了比 R-CNN 更好的效果(Faster R-CNN algorithm first convolutes the whole image, then fuses the candidate regions recommended by the selective search algorithm and the feature mapping maps calculated by the convolution network through the RoI Pooling Layer to get the corresponding feature vectors of the candidate regions, which greatly reduces the number of convolution calculations. Moreover, the dimension of these feature vectors is unified, which facilitates the subsequent classification work. The convolution feature is processed by the pooling layer of the region of interest, and the obtained feature is sent to two parallel computing tasks for training, classification and positioning regression. Through these methods and improved framework, Fast R-CNN uses shorter training and testing time and achieves better results than R-CNN.)

用MATLAN实现了FLL（锁频环）的热噪声误差、动态应力误差、总的跟踪误差与带宽、预检积分时间等的关系图(MATLAN achieve FLL (frequency-locked loop) thermal noise error and dynamic stress error, the total tracking error and bandwidth, preflight integration time diagram)

利用维纳滤波实现噪声中的语音信号恢复，产生chirp信号，噪声为高斯白噪声，噪声叠加在chirp信号上，构成观测信号x(n),使用编程方法实现使用维纳滤波滤除高斯噪声。(Wiener filtering the voice signal recovery in the noise generated chirp signal, the noise is Gaussian white noise, noise superimposed on the chirp signal, constitute the observed signal x (n), using the programming method using the Wiener filter filter Gaussian noise.)