Detailed introducing image enhancement , edge detection , thresholding , thinning etc . and many arithmetic such as median filter , sobel arithmetic operator , betterment gray scale histogram etc . applied or betterment applied it to the graduating project 图像增强、边缘检测、图像二值化、图像细化等,算法如中值滤波、边缘检测算子,改进的灰度直方图法、骨架抽取做了重要探讨,并应用或改进应用于本人的设计中。
Median filtering is chosen to modify the fuzzy edge by using the smoothing arithmetic , sobel operator is used to detect edge because of its filtering characteristic . image is changed from gray - scale to two - valued according to the threshold which is automaticly selected by otsu 选用中值滤波减小平滑算法导致的边缘模糊;利用有滤波特性的sobel算子进行边缘检测;使用最大类间方差自动确定阈值,将灰度图像转化为二值图像。
To develop the system to an automatic system , we have studied how to improve measurement precise by image processing after fast aiming . we use median filter to filtrate noise , use sobel arithmetic to detect the edge , and use kitchen - rosenfeld method to detect corners 第一种方法着重研究了视频采集和屏幕悬浮透明瞄准坐标系的同步显示以及提高瞄准精度的途径;第二种方法则着重研究快速瞄准后通过图像处理技术自动获得误差补偿数据。
The method of sobel edge detection is used to get the initial position and orientation value of optimization in order to improve the convergence speed . the iteration numbers and feature abstraction effect are compared with the method in which the initial value is randomly created 算法使用sobel边缘检测方法,给出优化初始位置和方向,以提高优化的收敛速度,针对提取特征时的优化迭代次数和提取特征的效果与随机产生优化初始值的方法进行了比较。
This paper study the greedy snake algorithms for the clutter tracking . it introduce the energy function of basic snake model and analyze the different kind of snake discrete energy function . it study on the method which use the pretreatment combining the sobel and grads arithmetic operators . the astringency of snake external energy was satisfied . this paper analyze the influence of tracking precision and running time for the preliminary border width , it also study on the tracking effect of different distributing and intensity . it is important to the next work 针对杂波数据的强度起伏大的特点,提出了结合sobel算子和梯度算子的预处理方法,满足了贪婪snake外部能量的收敛要求,在真实数据和仿真样本条件下,分析了初始边框间隔对跟踪精度、运行时间的影响,以及不同分布、不同强度杂波样本的跟踪效果,为后续的工作打下基础。