Otsu2d_Xiefen

使用区域生长法对灰度图像进行分割，显示原图及分割后的图像(Region growing method using gray-scale image segmentation, image display artwork and divided)

采用半隐式方案实现变分水平集图像分割方法(Active contour without edge, image segmentation)

通过计算机usb摄像头进行图像采集，然后实时处理，跟踪其中运动物体，并进行计数，采用中值滤波法得到背景，并根据需要进行更新，视频处理采用了动态差分与前景差分结合的方法，效果比较理想，但无法解决人物重叠的问题(Usb camera through the computer for image acquisition, and real-time processing and tracking moving objects in which, and the count median filter by using the background and update as needed, video processing using a dynamic difference with the prospect of differential combination of methods, results ideal, but can not solve the problem of overlapping characters)

C#+EmguCV机器视觉开发平台，功能：基本的图像处理工具、随意配置执行流程、二维码识别、文字识别等等。(C+EmguCV machine vision development platform, functions: basic image processing tools, arbitrary configuration of the execution process, two-dimensional code recognition, character recognition and so on.)

个人感觉最好用最小二乘支持向量机工具箱，特与大家共享！(Personal feeling is best to use least squares support vector machine toolbox, special to share with everyone!)

视觉引导车路识别：灰度AGV路径识别；彩色AGV路径识别；HSI彩色空间的AGV路径识别；路径中心线的定位；Radon变换的AGV路径偏差检测(Vision Guided Vehicle Road Identification: gray AGV path identification color AGV path identification HSI color space of the AGV path identification path centerline positioning Radon transform of the AGV path deviation Detected)

粒子群算法的Pareto多目标函数优化，适用于多目标问题的寻优(Particle swarm optimization Pareto multi-objective function optimization, suitable for multi-objective optimization.)

针对多聚焦图像融合，采用基于小波分解和点扩展函数模型PSF相结合的融合方法。(For multi-focus image fusion, based on wavelet decomposition and the point spread function PSF model combining fusion.)

能够对采集到的连续信号进行傅里叶变换，以进行后续处理(It can carry out Fourier transform to collect continuous signals for subsequent processing.)

三维激光点云配准是点云三维建模的关键问题之一。经典的 ICP 算法对点云初始位置要求较高且配准 效率较低，提出了一种改进的 ICP 点云配准算法。该算法首先利用主成分分析法实现点云的初始配准，获得较好 的点云初始位置，然后在经典 ICP 算法的基础上，采用 k － d tree 结构实现加速搜索，并利用方向向量夹角阈值去除 错误点对，提高算法的效率。实验表明，本算法流程在保证配准精度的前提下，显著提高了配准效率。 (Three-dimensional laser point cloud registration is one of the key three-dimensional point cloud model. High classical ICP algorithm to the initial position of the point cloud registration requirements and low efficiency, proposed an improved ICP point cloud registration algorithm. Firstly, the use of principal component analysis of the initial point cloud registration, get a better initial position of the point cloud, then the basis of classical ICP algorithm using k- d tree structure to achieve speed up the search, and using the direction vector angle the removal of the threshold point error and improve the efficiency of the algorithm. Experiments show that the algorithm processes to ensure the accuracy of registration under the premise, significantly improve the efficiency of registration.)