發(fā)布時(shí)間：2018-05-16 12:41

  本文選題：鏡頭邊界檢測(cè) + 廣告單元邊界檢測(cè)��； 參考：《南京理工大學(xué)》2013年碩士論文

【摘要】：隨著視頻數(shù)據(jù)的快速增長(zhǎng),如何對(duì)視頻進(jìn)行有效地管理、存儲(chǔ)和檢索成為用戶十分關(guān)注的問題。本文以廣告視頻為研究對(duì)象,對(duì)廣告視頻檢索相關(guān)技術(shù)進(jìn)行了研究,設(shè)計(jì)并實(shí)現(xiàn)了一個(gè)廣告單元檢索系統(tǒng)。 對(duì)視頻進(jìn)行了結(jié)構(gòu)化的處理,采用基于運(yùn)動(dòng)補(bǔ)償?shù)淖赃m應(yīng)雙閾值方法來判定突變鏡頭和漸變鏡頭。在此基礎(chǔ)上,針對(duì)電視視頻流中的廣告板塊,對(duì)廣告板塊邊界檢測(cè)的算法進(jìn)行了改進(jìn)。首先提取并計(jì)算鏡頭特征作為基本單元,然后采用概率輸出的SVM模型對(duì)視頻進(jìn)行分類,最后在搜索窗口內(nèi)搜索概率極值判斷出極大似然切分點(diǎn)。 對(duì)分割出的廣告板塊,利用小波變換等提取圖像特征,對(duì)其進(jìn)行滑動(dòng)窗口分割,判斷出鏡頭組邊界信息。對(duì)廣告板塊中的音頻,提取底層音頻特征如Mel頻率倒譜系數(shù)等,采用基于GMM和分割熵的檢測(cè)算法,實(shí)現(xiàn)了廣告板塊中音頻跳變檢測(cè)。 在總結(jié)了現(xiàn)有的廣告單元分割方法的基礎(chǔ)上,利用鏡頭組邊界檢測(cè)結(jié)果、音頻跳變的檢測(cè)結(jié)果和鏡頭邊界檢測(cè)結(jié)果,設(shè)計(jì)了一種基于音視頻融合的廣告單元分割方法,實(shí)現(xiàn)了廣告單元邊界檢測(cè)。 在廣告視頻的檢索方法上,提出一種新的基于奇異值分解的SURF-DAISY關(guān)鍵幀匹配方法,對(duì)相似度模型的實(shí)際應(yīng)用進(jìn)行了研究,并提出了廣告視頻檢索的具體步驟。 本文采用Visual C++為開發(fā)平臺(tái),實(shí)現(xiàn)了一個(gè)廣告視頻檢索系統(tǒng),驗(yàn)證了上述方法。
[Abstract]:With the rapid growth of video data, how to effectively manage, store and retrieve video has become a very important issue for users. In this paper, the advertising video retrieval technology is studied, and an advertising unit retrieval system is designed and implemented. The video is processed structurally, and the adaptive double threshold method based on motion compensation is used to determine the abrupt shot and the gradient shot. On the basis of this, the algorithm of advertising block boundary detection in TV video stream is improved. Firstly the shot feature is extracted and calculated as the basic unit then the SVM model of probability output is used to classify the video. Finally the maximum likelihood segmentation point is determined by searching the maximum probability extremum in the search window. The image features are extracted by wavelet transform, and the moving window is segmented to judge the boundary information of shot group. The bottom audio features such as Mel frequency cepstrum coefficients are extracted for the audio in the advertising block. The detection algorithm based on GMM and segmentation entropy is used to realize the detection of audio jump in the advertising block. On the basis of summarizing the existing methods of advertising unit segmentation, using the results of shot group boundary detection, audio jump detection and shot boundary detection, a method of advertising unit segmentation based on audio and video fusion is designed. The advertisement unit boundary detection is realized. In this paper, a new SURF-DAISY key frame matching method based on singular value decomposition (SVD) is proposed. The practical application of similarity model is studied, and the concrete steps of advertising video retrieval are proposed. In this paper, an advertising video retrieval system based on Visual C is implemented, and the above method is verified.
【學(xué)位授予單位】：南京理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 丁南南;劉艷瀅;張葉;陳春寧;賀柏根;;基于SURF-DAISY算法和隨機(jī)kd樹的快速圖像配準(zhǔn)[J];光電子.激光;2012年07期

2 汪玉山;史萍;;基于音視頻特征的電視廣告單元分割技術(shù)[J];電視技術(shù);2012年16期

3 郭麗;劉磊;季鐸;蔡?hào)|風(fēng);;基于奇異值分解的關(guān)系相似度計(jì)算方法[J];沈陽(yáng)航空工業(yè)學(xué)院學(xué)報(bào);2009年03期

4 宣麗萍;;音頻跳變點(diǎn)的分割熵檢測(cè)算法[J];黑龍江科技學(xué)院學(xué)報(bào);2008年03期

5 任仙怡,張桂林,張?zhí)煨?廖云濤;基于奇異值分解的圖像匹配方法[J];紅外與激光工程;2001年04期

6 張春林,張鵬林,胡瑞敏;新聞視頻中基于主持人識(shí)別的新聞故事探測(cè)[J];計(jì)算機(jī)工程;2003年14期

7 權(quán)文;王曉丹;王堅(jiān);張玉璽;;基于SVM概率輸出與證據(jù)理論的多分類方法[J];計(jì)算機(jī)工程;2012年05期

8 李勇;劉雨;唐波;肖奇?zhèn)?;一種基于時(shí)空切片的鏡頭突變檢測(cè)方法[J];計(jì)算機(jī)工程與科學(xué);2006年03期

9 莊越挺,劉小明,吳翌,潘云鶴;通過例子視頻進(jìn)行視頻檢索的新方法[J];計(jì)算機(jī)學(xué)報(bào);2000年03期

10 王志明;張瑞杰;李弼程;;基于分層熵檢測(cè)的音頻分割算法[J];科學(xué)技術(shù)與工程;2009年17期

相關(guān)碩士學(xué)位論文 前1條

1 劉立巍;多模態(tài)融合的新聞故事單元分割[D];南京理工大學(xué);2010年

，

本文編號(hào)：1896849