本文關(guān)鍵詞： 矢量場(chǎng) 層次聚類 加權(quán) 異常檢測(cè)　出處：《昆明理工大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：軌跡分析指的是對(duì)運(yùn)動(dòng)目標(biāo)運(yùn)行軌跡進(jìn)行分析,以便獲取運(yùn)動(dòng)目標(biāo)的行為。軌跡異常檢測(cè)就是通過(guò)對(duì)軌跡進(jìn)行分析檢測(cè)出其中出現(xiàn)的目標(biāo)異常行為、異常事件。軌跡異常檢測(cè)可應(yīng)用于颶風(fēng)、動(dòng)物遷徙預(yù)測(cè),交通流監(jiān)測(cè)等方面。隨著衛(wèi)星定位數(shù)據(jù)、交通監(jiān)控視頻數(shù)據(jù)量的迅速增長(zhǎng),軌跡數(shù)據(jù)量與其包含的時(shí)空信息也迅速增長(zhǎng),然而通過(guò)人工分析數(shù)據(jù)的方式耗時(shí)耗力,且容易出現(xiàn)錯(cuò)誤。本文利用聚類的方式將時(shí)空軌跡數(shù)據(jù)劃分為不同的簇,通過(guò)計(jì)算聚類中心軌跡與待檢測(cè)軌跡之間的距離從而自動(dòng)判別時(shí)空軌跡正常與否,以便有效解決各類時(shí)空數(shù)據(jù)分析應(yīng)用。本文首先簡(jiǎn)要分析了從視頻數(shù)據(jù)中獲取運(yùn)動(dòng)目標(biāo)軌跡的幾種常見(jiàn)方法的優(yōu)缺點(diǎn)。其次,提出一種矢量場(chǎng)層次聚類的方法對(duì)軌跡數(shù)據(jù)進(jìn)行聚類,解決矢量場(chǎng)軌跡聚類不能自適應(yīng)聚類類別數(shù)的問(wèn)題,并且通過(guò)加權(quán)矢量場(chǎng)擬合解決噪聲軌跡點(diǎn)對(duì)聚類結(jié)果的干擾,增強(qiáng)了算法的魯棒性。最后,通過(guò)計(jì)算檢測(cè)數(shù)據(jù)矢量場(chǎng)與各聚類中心軌跡矢量場(chǎng)的相似度,判定待測(cè)試軌跡正常與否。通過(guò)對(duì)監(jiān)控視頻數(shù)據(jù)上進(jìn)行的實(shí)驗(yàn)表明,本文提出的軌跡聚類方法與傳統(tǒng)的軌跡聚類相比具有更高的類別適應(yīng)性與魯棒性,對(duì)異常軌跡檢出率達(dá)到90%以上。
[Abstract]:Trajectory analysis refers to the trajectory analysis of the moving target, in order to obtain the target behavior. Anomaly detection is based on the trajectory analysis to detect abnormal behavior, which targets abnormal events. Trajectory trajectory outlier detection can be applied to the hurricane, animal migration prediction, traffic flow monitoring. With the development of satellite positioning data the rapid growth of traffic monitoring, the amount of video data, and contains temporal information track data are also growing rapidly. However, through artificial way of analyzing data is time-consuming, error and easy to use. In this way the clustering of trajectory data into different clusters, by calculating the cluster center trajectory and trajectory to be detected between the distance to automatically determine the spatio-temporal trajectory is normal or not, in order to effectively solve various spatio-temporal data analysis applications. This paper briefly analyzes from the optic frequency According to the advantages and disadvantages of several common methods for moving target. Secondly, put forward a kind of vector field hierarchical clustering method to cluster the trajectory data, solve trajectory clustering is not adaptive clustering number vector field, and by weighted vector field fitting to solve noise track points on the clustering results, robustness the algorithm. Finally, by calculating the detection data of vector field and the cluster center trajectory vector field test trajectory similarity, determined to be normal or not. Based on the video surveillance data. Experimental results show that, compared with the trajectory clustering trajectory clustering method proposed in this paper with the traditional categories of adaptability and greater robustness, the abnormal the trajectory detection rate reached more than 90%.

【學(xué)位授予單位】：昆明理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：X924.2;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 王蒙;戴亞平;;基于FAST角點(diǎn)檢測(cè)的局部魯棒特征[J];北京理工大學(xué)學(xué)報(bào);2013年10期

2 潘奇明;周文輝;程詠梅;;運(yùn)動(dòng)目標(biāo)軌跡分類與識(shí)別[J];火力與指揮控制;2009年11期

3 曲琳;周凡;陳耀武;;基于Hausdorff距離的視覺(jué)監(jiān)控軌跡分類算法[J];吉林大學(xué)學(xué)報(bào)(工學(xué)版);2009年06期

4 劉良旭;喬少杰;劉賓;樂(lè)嘉錦;唐常杰;;基于R-Tree的高效異常軌跡檢測(cè)算法[J];軟件學(xué)報(bào);2009年09期

5 孫吉貴;劉杰;趙連宇;;聚類算法研究[J];軟件學(xué)報(bào);2008年01期

6 查宇飛;張育;畢篤彥;;基于區(qū)域活動(dòng)輪廓運(yùn)動(dòng)目標(biāo)跟蹤方法研究[J];中國(guó)圖象圖形學(xué)報(bào);2006年12期

7 林廣宇;魏朗;;基于數(shù)字圖像技術(shù)的汽車行駛軌跡狀態(tài)識(shí)別[J];交通運(yùn)輸工程學(xué)報(bào);2006年03期

8 胡衛(wèi)明,謝丹,譚鐵牛,沈俊;軌跡分布模式學(xué)習(xí)的層次自組織神經(jīng)網(wǎng)絡(luò)方法[J];計(jì)算機(jī)學(xué)報(bào);2003年04期

9 李培華,張?zhí)镂?主動(dòng)輪廓線模型(蛇模型)綜述[J];軟件學(xué)報(bào);2000年06期

，

本文編號(hào)：1480818