An Adaptive Exponential Filter Used for Image Edge Detection 一种用于图像边缘检测的自适应指数滤波器
The typical filter technique in ABS wheel speed signals is the exponential smoothing data filter technique whose filter gain is constant. Because the variation characteristics of measured signal variance are not to be considered, its filter effect can not meet the requirements of ABS. 常用的ABS轮速信号的滤波技术是滤波增益为常数的指数平滑数据滤波技术,由于没有考虑到测量信号的方差变化特性,滤波效果不能满足ABS系统的要求。
The third chapter proposes two new filters, which are a piecewise degree n and an exponential function of λ, respectively, and we prove how they can be successfully used instead of the classical Donoho and Johnstone's Soft thresholding filter. 第三章提出了两个新滤波器,它们分别是参数λ的分段n次函数和指数函数,并证明了它们可以成功地用来代替常用的Donoho和Johnstone软阈值滤波器。
The reason results in that small fault is difficult to be detected lies in the large noise of measured data and Exponential Weighted Moving Average ( EWMA) filter is proposed to improve the ability of small fault detection. 分析认为造成传感器小故障难于检测的原因在于测量数据的噪声太大,并提出采用指数加权滑动平均滤波器来提高系统小故障检测能力的方法。
An edge detection approach via IIR exponential smoothing filtering and nonlinear Laplacian operator is proposed, which has the following advantages. At first, utilizing the exponential kind filter can greatly enhance the ability of smoothing noise, and can implement fast structure recurred filter. 本文提出了一种基于IIR指数平滑滤波与非线性Laplacian算子的边缘检测方法,具有以下特点:采用最佳指数平滑滤波器增强抗噪声能力,且可实现无限大窗口的快速递归运算;
The FLAC monitor if the residual is zero-mean white noise and adjust the exponential weighting of filter, then the filter perform optimally. The extended KALMAN filter ( EKF) is modified using the fuzzy logic Adaptive system, and compared with the performance of a regular EKF. FLAC通过监视残差是否为零均值白噪音,然后根据模糊规则调整滤波器的指数加权,从而使滤波器不断执行最优估计。
Sufficient conditions are proposed for the existence of desired exponential filter. 获得了理想的H∞指数滤波器存在的充分条件。
This paper presents an adaptive exponential filter used for edge detection. It can be designed to eliminate the noise of image and obtain good precision of edge localization. 本文提出一种用于图像边缘检测的自适应滤波器,利用它既可滤除图像噪声,又可获取精确度高的图像边缘。
The exponential filter parameter is discussed. 分析了指数滤波参数对相关结果的影响。
The input ( record) is filtered by the application of an optimum bilateral exponential filter, and the output model of gold signal for each line is produced. 利用最佳双边指数滤波器,输入信号(记录)进行滤被面得出各测线上金的信号输出模塑。
This paper adopts an exponential average low pass filter and a cascade form to effectively solve the conflict between the veracity of harmonic detection and real-time requirements and the finite word length of DSP design. 文中采用指数平均低通滤波器,并利用级联技术,解决了数字信号处理器有限字长与谐波提取的准确性和实时性之间的矛盾。
The third is rank-ordered absolute differences ( ROAD) detection based exponential similarity ordering filter. 第三种是基于绝对差值排序检测的指数相似度排序滤波器。
Another method is jitter buffer based fast convergence method that uses variation of jitter buffer fill level to get the description of the clock difference between the two end clocks, and the double EWMA ( Exponential Weight Moving Average) filter is used to carry out noise cancellation. 另一种方法则是基于去抖缓冲区的快速锁定的自适应时钟方法,其主要特征是基于缓冲区变化量得到时钟差异的描述信息,并利用两级EWMA滤波完成噪声的消除。