利用串場效果偵測之實用棒球比賽精華擷取暨分類系統 ; A Practical Highlight Extraction and Classification Scheme in Baseball Videos Based on Transition Effect Detection

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/44754

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/44754

Title:	利用串場效果偵測之實用棒球比賽精華擷取暨分類系統;A Practical Highlight Extraction and Classification Scheme in Baseball Videos Based on Transition Effect Detection
Authors:	曾子欣;Zi-Xin Zeng
Contributors:	資訊工程研究所
Keywords:	棒球;視訊;串場效果;慢動作;SVD;MPEG;HMM;Baseball;Highlight;transition effect;slow motion;MPEG;SVD;HMM
Date:	2010-07-28
Issue Date:	2010-12-09 13:54:29 (UTC+8)
Publisher:	國立中央大學
Abstract:	棒球比賽轉播在目前商用電視台十分常見，一場比賽的完整視訊內容通常為九局上下，然而隨著實際比賽的變化以及球團戰術的實行，其比賽時間之長往往超乎預期，對於觀看棒球比賽者而言，並不是每個人都有充裕的時間能將整場比賽瀏覽完畢而且並非所有內容觀看者都非常感興趣，因此，若是能事先將大量的比賽資訊透過系統做分析整理並擷取出不同的精華片段提供觀看者做選擇，則可讓觀看者節省大量的時間並得到想要的資訊。本論文提出一個在MPEG 壓縮域底下對輸入系統的視訊內容做分析並擷取出精華片段再加以分類的系統。由於精采畫面的前後通常都會有電視台加入的串場效果以通知觀看者，因此我們首先透過鏡頭變化偵測加上視訊內容分析找出串場效果畫面的資訊，並利用數位訊號處理的方式準確的建立出串場效果模板，以此模版我們可定位出所有串場效果出現的位置。再來，我們以SVD 對所有出現在串場效果前後的投捕畫面做特徵擷取並定出一個特徵的閥值範圍，可作為投捕畫面的定位判斷，投捕畫面至串場效果即為分析精華片段的開始與結束點。最後，在找到串場效果畫面以及投捕畫面之後，我們可將此段畫面以HMM加以分析與分類，可得知此片段是否為我們所定義的某種精華片段。由於系統是建立在MPEG視訊壓縮的資訊串流做計算，因此可節省大量複雜的運算。實驗結果顯示本系統在視訊畫面的分析上以及精華擷取的分類上具有相當高的準確度與運算效率。This research presents a system of analyzing video content in the MPEG compressed videos for classifying the highlights in baseball videos. The system makes use of the transition e ects inserted preceding and following the slow motion replays by the broadcaster, which demonstrate highlights of the game. First, we examine the characteristics of the tran- sition e ects via the camera changes and the video content analysis to construct the transition e ect template, which can help to locate all the appearances of transition e ects. Next, we search the pitching views which appear before the transition e ects and construct the pitching view model for this game. Finally, after we locate the highlight candidates, we will apply HMM (Hidden Markov Model) to analyze and classify the content to ensure that the extracted highlights match our de nitions of high-level highlight semantics. Because the system is based on MPEG compressed video data streams, it can save a large amount of computational complexity. The experimental results show the feasibility of the potential solution.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	837	View/Open

社群 sharing

Loading...