【摘要】：如今社會,人們不滿足于僅僅播放多媒體信息,轉(zhuǎn)向基于視頻對象的訪問、檢索和操作,于是基于視頻的運(yùn)動分割技術(shù)成為了研究重點。運(yùn)動分割是將視頻中有著不同運(yùn)動的物體分開,是基于對象的視頻編碼、視頻檢索、多媒體交互的基石。傳統(tǒng)的運(yùn)動分割算法采用運(yùn)動目標(biāo)檢測和目標(biāo)跟蹤,在利用幀差法和光流法對運(yùn)動目標(biāo)檢測時,極易受到噪聲的影響,目標(biāo)跟蹤又涉及目標(biāo)的遮擋、扭曲和變形等問題,于是復(fù)雜場景下進(jìn)行運(yùn)動分割很難得到理想效果。通過轉(zhuǎn)換問題的角度,采用稀疏子空間聚類算法,避開運(yùn)動檢測和目標(biāo)跟蹤遇到的難題,來實現(xiàn)復(fù)雜場景下的運(yùn)動分割。基于同一運(yùn)動的特征點軌跡在同一線性流形上,于是可以利用稀疏子空間聚類算法對特征點軌跡進(jìn)行聚類來實現(xiàn)運(yùn)動分割。稀疏子空間聚類算法在處理高維數(shù)據(jù)時,能夠?qū)⒏呔S數(shù)據(jù)分割到所屬的低維子空間中去,揭示高維數(shù)據(jù)所在本質(zhì)子空間,算法可以同時處理奇異點和噪聲對聚類的影響,針對稀疏子空間算法的研究,本文做了如下工作:(1)通過對比k-means算法,深入研究自適應(yīng)譜聚類算法。由于稀疏子空間聚類算法是基于譜聚類,對譜聚類的相關(guān)基礎(chǔ)和理論知識做了深入研究,分析譜聚類的研究成果和應(yīng)用現(xiàn)狀,針對譜聚類需要手動輸入聚類數(shù)目的缺點,本文依據(jù)矩陣的擾動理論,同時計算矩陣的特征間隙,從而實現(xiàn)聚類算法自動確定聚類數(shù)目。為了證明譜聚類算法能處理任意樣本形狀的數(shù)據(jù)集,而且不陷入局部最優(yōu),本文選取各種形狀的樣本集進(jìn)行實驗,同時用k-means算法處理這些樣本集,通過實驗對比,發(fā)現(xiàn)自適應(yīng)譜聚類算法在處理樣本集上的優(yōu)勢。(2)提出混合最小二乘回歸的稀疏子空間聚類算法。針對稀疏子空間聚類算法如何構(gòu)造真實合理反映數(shù)據(jù)集的相似度矩陣的問題,相似度矩陣既要類間稀疏又要內(nèi)類均勻,這樣才能保證屬于同一個類的數(shù)據(jù)點相似度最大,屬于不同類的數(shù)據(jù)點相似度最小,對于樣本集存在各種噪聲點、奇異樣本點和孤立點,本文采用數(shù)據(jù)項矩陣來處理噪聲的影響,通過分析稀疏子空間聚類專注于每一個數(shù)據(jù)表示系數(shù)的最大稀疏性,缺乏對數(shù)據(jù)集全局結(jié)構(gòu)的描述;低秩子空間聚類算法保證了同一類數(shù)據(jù)的結(jié)構(gòu)相關(guān)性,但是不夠稀疏。本文決定將最小二乘回歸引入稀疏子空間聚類算法中,從而保證數(shù)據(jù)的相似度矩陣兼具稀疏性和分組效應(yīng),并用數(shù)據(jù)集驗證改進(jìn)算法性能。(3)研究改進(jìn)稀疏子空間聚類算法在運(yùn)動分割中的應(yīng)用。將稀疏子空間聚類算法應(yīng)用于視頻對象處理中,建立運(yùn)動分割模型,進(jìn)行運(yùn)動分割實驗,實驗結(jié)果表明,改進(jìn)的算法在保證時間復(fù)雜度的情況下,提高了運(yùn)動分割的準(zhǔn)確率。
文內(nèi)圖片：
圖片說明：無向圖G
[Abstract]:Nowadays, people are not satisfied with just playing multimedia information, but turn to video object-based access, retrieval and operation, so video-based motion segmentation technology has become the focus of research. Motion segmentation is the cornerstone of object coding, video retrieval and multimedia interaction, which separates objects with different motion in video. The traditional motion segmentation algorithm adopts moving target detection and target tracking. When using frame difference method and optical flow method to detect moving target, it is easy to be affected by noise. Target tracking also involves the occlusion, distortion and deformation of the target, so it is difficult to get the ideal effect of motion segmentation in complex scene. From the point of view of the problem, sparse subspace clustering algorithm is used to avoid the problems encountered in motion detection and target tracking, so as to realize the motion segmentation in complex scenes. The feature point trajectory based on the same motion is on the same linear manifold, so the sparse subspace clustering algorithm can be used to cluster the feature point trajectory to realize motion segmentation. When dealing with high-dimensional data, sparse subspace clustering algorithm can segment high-dimensional data into its own low-dimensional subspace, reveal the local proton space of high-dimensional data, and the algorithm can deal with the influence of singularity and noise on clustering at the same time. aiming at the research of sparse subspace algorithm, this paper does the following work: (1) by comparing k-means algorithm, the adaptive spectral clustering algorithm is deeply studied. Because the sparse subspace clustering algorithm is based on spectral clustering, the related basic and theoretical knowledge of spectral clustering is deeply studied, and the research results and application status of spectral clustering are analyzed. Aiming at the disadvantage that spectral clustering needs to manually input the number of clustering, this paper calculates the characteristic gap of matrix according to the perturbation theory of matrix, so as to realize the automatic determination of clustering number by clustering algorithm. In order to prove that spectral clustering algorithm can deal with arbitrary sample shape data sets, and does not fall into local optimization, this paper selects various shapes of sample sets to carry out experiments, and uses k-means algorithm to deal with these sample sets. Through experimental comparison, the advantages of adaptive spectral clustering algorithm in dealing with sample sets are found. (2) mixed least square regression sparse subspace clustering algorithm is proposed. In order to solve the problem of how to construct the similarity matrix which truly and reasonably reflects the dataset, the similarity matrix should be sparse between classes and uniform within classes, so as to ensure that the similarity of data points belonging to the same class is the largest and the similarity of data points belonging to different classes is the smallest. For the sample set, there are all kinds of noise points, singular sample points and isolated points. In this paper, the data item matrix is used to deal with the influence of noise. By analyzing the sparse subspace clustering, it focuses on the maximum sparsity of each data representation coefficient, and lacks the description of the global structure of the data set. The low rank subspace clustering algorithm ensures the structural correlation of the same kind of data, but it is not sparse enough. In this paper, we decide to introduce least square regression into sparse subspace clustering algorithm, so as to ensure that the similarity matrix of data has both sparsity and grouping effect, and the performance of the improved algorithm is verified by data set. (3) the application of improved sparse subspace clustering algorithm in motion segmentation is studied. The sparse subspace clustering algorithm is applied to video object processing, and the motion segmentation model is established and the motion segmentation experiment is carried out. the experimental results show that the improved algorithm improves the accuracy of motion segmentation under the condition of ensuring the time complexity.
【學(xué)位授予單位】：重慶理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41;TP311.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 楊歡;劉小玲;;虛擬現(xiàn)實系統(tǒng)綜述[J];軟件導(dǎo)刊;2016年04期

2 劉建華;;基于隱空間的低秩稀疏子空間聚類[J];西北師范大學(xué)學(xué)報(自然科學(xué)版);2015年03期

3 王衛(wèi)衛(wèi);李小平;馮象初;王斯琪;;稀疏子空間聚類綜述[J];自動化學(xué)報;2015年08期

4 許凱;吳小俊;;基于重建系數(shù)的子空間聚類融合算法[J];計算機(jī)應(yīng)用研究;2015年11期

5 賈璦瑋;;基于劃分的聚類算法研究綜述[J];電子設(shè)計工程;2014年23期

6 歐陽佩佩;趙志剛;劉桂峰;;一種改進(jìn)的稀疏子空間聚類算法[J];青島大學(xué)學(xué)報(自然科學(xué)版);2014年03期

7 姚剛;楊敏;;稀疏子空間聚類的懲罰參數(shù)自調(diào)整交替方向法[J];計算機(jī)技術(shù)與發(fā)展;2014年11期

8 高文;朱明;賀柏根;吳笑天;;目標(biāo)跟蹤技術(shù)綜述[J];中國光學(xué);2014年03期

9 張權(quán);胡玉蘭;;譜聚類圖像分割算法研究[J];沈陽理工大學(xué)學(xué)報;2012年06期

10 王駿;王士同;鄧趙紅;;聚類分析研究中的若干問題[J];控制與決策;2012年03期

相關(guān)博士學(xué)位論文 前2條

1 陳黎飛;高維數(shù)據(jù)的聚類方法研究與應(yīng)用[D];廈門大學(xué);2008年

2 姜志俠;數(shù)學(xué)規(guī)劃中的原始對偶內(nèi)點方法[D];吉林大學(xué);2008年

相關(guān)碩士學(xué)位論文 前10條

1 謝浪雄;稀疏表示理論及其應(yīng)用研究[D];廣東工業(yè)大學(xué);2015年

2 楊陽;數(shù)據(jù)挖掘K-means聚類算法的研究[D];湖南師范大學(xué);2015年

3 管春苗;基于機(jī)器視覺的運(yùn)動目標(biāo)軌跡跟蹤技術(shù)研究[D];沈陽理工大學(xué);2015年

4 周成舉;基于約束稀疏表達(dá)的視頻人臉聚類[D];天津大學(xué);2014年

5 王云峰;視頻對象分割技術(shù)研究[D];廣東工業(yè)大學(xué);2014年

6 張亞平;譜聚類算法及其應(yīng)用研究[D];中北大學(xué);2014年

7 陸洪濤;偏最小二乘回歸數(shù)學(xué)模型及其算法研究[D];華北電力大學(xué);2014年

8 黎蕾;求解凸最優(yōu)化問題的近似交替方向法[D];重慶師范大學(xué);2013年

9 萬海霞;圖與混合圖的特征值問題研究[D];鄭州大學(xué);2013年

10 羅懷金;基于近鄰路徑的自適應(yīng)尺度譜聚類算法研究[D];哈爾濱工程大學(xué);2012年

，

本文編號：2511109