First , we must filter the image to reduce noises , this process includes two phase : one is using the nearest neighbor middle filter ( knnmf ) to reduce the isolated noises , the other is using gauss filter to remove white noises 首先对图像进行滤波去除噪声。滤波包括两个阶段：一是进行最近邻中值滤波( knnmf ) ，以去除孤立点噪声，再进行高斯平滑，去除白噪声。
Finally we analyze common data association algorithm such as nearest neighbor algorithm 、 probability data association algorithm and joint probability data association algorithm and adopt the nearest neighbor algorithm in our simulation system 然后对当前比较常用的数据关联算法最近邻法、概率数据关联( pda )算法以及联合概率数据( jpda )算法进行了分析选择。
The low frequency sub - images are obtained by utilizing two - dimensional wavelet transform for several times . the features are extracted by applying lda to the sub - images . the nearest - neighbor classifier is designed to recognize the face 首先对原图像进行若干层二维小波变换，得到低频子图像，然后利用lda进行特征提取，最后利用最近邻分类器进行分类识别。
When the discount coefficient is 1 and all weights of the nearest neighbor sample points are the same , the k - nn classification method based on evidence reasoning model will become the k - nn classification method based on evidence theory 并且当折扣系数为1 ，且给定所有最近邻样本点权重相等时，基于证据推理模型的k - nn分类方法就成为基于证据理论的k - nn分类方法。
For all doping concentrations , the coupling between cr and the nearest neighbor n is found to be antiferromagnetic , and the cr 3d states hybridize strongly with n 2p states , which are in agreement with the band calculations 对于不同的掺杂浓度, cr原子与最近邻n原子之间均为反铁磁偶合, cr原子的3d电子与n原子的2p电子之间有很强的杂化,这和晶体的能带计算方法得到的结果一致。
This study employed six data mining methods , including logistic regression , discriminant analysis , artificial neural networks , k - nearest - neighbors , na ? ve bayes classifier , and classification trees , to find the most important factors of earthquake - caused landslide 本研究利用六种资料探勘方法，包括逻辑回归、判别分析、类神经网络、最近邻法、贝氏分类器、分类树，探讨影响地震引起山崩的重要因子。
Based on the vq - based indexing method , a hierarchical indexing scheme is proposed for higher performance . this approach integrates vq - based indexing structures with approximate nn searches and performs probabilistic approximate nn searches on approximate vectors 该方法在矢量量化索引机制的基础上使用了概率近似最近邻的方式进行检索，这种两种索引机制相结合的方法取得了比单独的索引机制更好的性能。
During the former classification method , the classification expert gives the weights of the nearest neighbor sample points of the sample point to be classified , then defines the key sample point and non - key sample point , furthermore gives their support degree , discount coefficient 在前一种分类方法中，分类专家对待分类样本点的最近邻样本点给出权重，从而定义关键样本点及非关键样本点，进而给出它们的支持度、折扣系数。
A variety of methods including the tabular comparison of data , the tabular comparison of similarity coefficient , the nearest neighbor method and the group - average method of hierarchical agglomerative classification were applied to investigate the forest communities in meizi lake area 森林植被样地中以乔木层树种的重要值为指标，采用纸条排队法、群落相似系数分类法、最近邻体法、组平均法对梅子湖森林植被样地进行数量分类。
The fast nearest - neighbor search algorithm was proposed to accelerate the process of anomaly detection . it can reduce the unnecessary vector similarity measures , producing a large computation time saving and the high real - time performance of anomaly detection system 网络异常检测的可用性最为重要,为了提高算法分析的效率和系统的实用性,针对码书的结构,设计实现了一种基于向量约减的快速最近邻搜索算法来实现高效的网络异常检测。
