stblpdf

Speedy GA Graphical analysis

将图片通过灰度值转化为文字符号，并排列为原来图像的样式。 (Convert image by gray scale value into text symbols, side by side as the style of the original image. )

基于半色调图像的数据隐藏即加数字水印，针对半色调图像，所以使用的是DHSPT算法，算法较容易理解，简单，嵌入效果相对较好(Hidden based on the halftone image data plus digital watermark for halftone images, use DHSPT algorithm, the algorithm is easy to understand, simple, embedded effect is relatively good)

说明：  自己写的位图三态按钮还添加了按钮声音,对自己制作图形界面有很大帮助(Write your own bitmap three-state buttons to add the button sounds right to produce their own graphical interface helps a lot)

W.T.Freeman 经典 基于例子的超分辨率重构matlab源码，图片自己可加，经测试可用(WTFreeman classic example of super-resolution reconstruction based on matlab source code, images themselves can be added, the test can be used)

KLT跟踪算法，附有程序，并附有两个例子，可以完美运行(KLT tracking algorithm, with the program, along with two examples of the perfect run)

粒子滤波算法的实现，其中重采样采用残差重采样算法。(Particle filter algorithm, in which the use of residual resampling resampling algorithm.)

运用于人脸图像预处理中，主要实现功能有：人脸检测，人脸对齐，人脸图像放缩，人脸剪切。(Applied to the face image preprocessing, the main features are: face detection, face alignment, face image zoom, face cuts.)

针对杂波环境下的多个机动目标跟踪问题, 本文将多模型概率假设密度 (Multiple-model probability hypothesis density, MM-PHD) 滤波器和平滑算法相结合, 提出了 MM-PHD 前向 – 后向平滑器. 为了避免引入复杂的随机有限集 (Random finite set, RFS) 理论, 本文根据 PHD 的物理空间 (Physical space) 描述法推导得到了 MM-PHD 平滑器的后向更 新公式. 由于 MM-PHD 前向–后向平滑器的递推公式中包含有多个积分(By integrating the multiple-model probability hypothesis density (MM-PHD) filter with the smoothing al- gorithms, an MM-PHD forward-backward smoother is proposed in this paper for tracking multiple maneuvering targets in clutter. To avoid use of complex random finite set (RFS) theory, the backward updated equation of the MM-PHD smoother can be derived according to the physical-space explanation of the PHD)

全卷积神经网络FCN用于图像分割的工具箱(FCN for image segmentation)