Actac

对于一幅输入图像，将产生一幅输出图像，输出图像的每个像素点的灰度值由输入像素点决定，将感兴趣特征的对比度扩展使之占据可显示灰度级的更大部分。使一输入图像转换为在每一灰度级上都有相同的像素点数，均衡化会减少灰度级数。大多数自然图像由于其灰度值分布集中在较窄的范围之内，引起图像细节不够清楚。采用直方图修正后可使图像的灰度间距拉大或使灰度均匀分布，从而增大了反差，使图像细节清楚，达到增强的目的。(For an input image, will produce an output image, the output image gray value of each pixel from the input pixel decided to expand interest in the contrast of features to make it account for more gray levels can be displayed most. Make an input image is converted to grayscale in each pixel have the same points, equalization will reduce the gray series. Most natural images because of their gray value distribution concentrated in a narrow range, causing the image detail is not clear enough. Make the image histogram revised by the gray space widening or so gray uniform distribution, thereby increasing the contrast between the image details clear objective to achieve enhanced.)

implementation of median filter in matlab

字典学习的各种方法，K-SVD,online等主要的学习算法(Various methods dictionary learning, K-SVD, online learning algorithms and other major)

xampling compressive sensing

Dynamic_Two Links Robot Manipulator_Matlab Simulink

粘弹性地基结构动力响应计算程序，虽然比较老，还是很有价值。(Visco-elastic foundation structural dynamic response calculation program, though relatively old, is still of great value.)

说明：  自己写的Canny算子的matlab实现和演示。程序虽然不大，但相信也能给大家一些参考。 (himself wrote the Canny Operator Matlab realization and demonstration. Procedures not much, but I believe that can give you some information.)

说明：  配电网规划必须满足辐射性和连通性，配电网扩展规划问题随机产生的初始群体中往往存在大量不可行解，采用根节点融合法可确保方案的可行性。(Distribution network planning must meet radiation and connectivity, distribution network expansion planning problem with randomly generated initial population there is often a large number of infeasible solutions, use of root fusion method to ensure feasibility.)

turbo 编解码 各种算法，包括sova，logmap等(turbo encoding and decoding various algorithms)

二维图像的读取，辨认，分析，并且用神经网络进行，判定哪些图片可以归为一类(Two-dimensional image reading, identifying, analyzing, and using neural network to determine which pictures can be classified as a class)