traffic-sign-recognition

利用奇异值分解（SUV）对图像信号分行降噪！(Singular value decomposition (SUV) of the image signal branch noise!)

图像420转换为422或444的算法。高斯加噪算法，脉冲加噪算法。(YCbCr 4:2:0 image is converted to 4:2:2 or 4:4:4 of the algorithm, Matlab program is)

叠加平均去噪过程的实现，信号产生，加噪，去噪效果的显示(Average denoising, signal generation, noise addition, display the denoising effects)

this is a program for image compression

改程序就代码可以找出从不同角度或区间的两幅遥感影像的同名点(Change the program code can find it from different angles of the same name point or interval of two remote sensing images)

多个函数的曲线拟合，里面统计有最小二乘、非线性回归(Curve fitting for multiple functions)

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)

采用的是基于弦中点的hough变换椭圆检测算法，大家可以在百度上搜这篇论文，由于不是用的标准hough变换的算法，所以运行速度还不错(Used is based on the midpoint of the chord hough transform ellipse detection algorithm, we can search on Baidu paper, because the algorithm is not used in standard hough transform, so the speed is also good)

说明：  快速的人脸特征提取算法KPCA，比普通的pca特征提取算法在效率上好了不少(Fast facial feature extraction algorithm KPCA, than ordinary PCA feature extraction algorithm in the efficiency of a good many)

Opencv处理图像，用ostu法实现图像的二值化(Opencv image processing, using ostu method to achieve image binarization)