發(fā)布時間：2018-04-23 17:26

  本文選題：視頻摘要 + 自適應(yīng)　； 參考：《杭州電子科技大學(xué)》2017年碩士論文

【摘要】：近年來,飛速發(fā)展的互聯(lián)網(wǎng)和多媒體技術(shù)導(dǎo)致多媒體信息數(shù)據(jù)驟增。數(shù)字視頻,作為主要的多媒體信息載體,廣泛地應(yīng)用于生活的方方面面。大量的視頻一方面可以使人們以更豐富的形式獲取信息,給生活帶來了很大的便利。而另一方面,卻給視頻存儲、傳輸、歸檔和檢索帶來巨大壓力。因此,視頻摘要技術(shù)應(yīng)運(yùn)而生。類似于文本摘要,視頻摘要是對原始視頻內(nèi)容的總結(jié)和概括,通過分析原始視頻數(shù)據(jù)流,從中選擇有意義的視頻內(nèi)容組成緊湊的摘要。視頻摘要可以結(jié)合視頻標(biāo)注技術(shù)用于視頻檢索;還可以形成獨(dú)立的產(chǎn)品比如電影預(yù)告片,應(yīng)用在我們的日常生活中。技術(shù)是目前計(jì)算機(jī)視覺領(lǐng)域的研究熱點(diǎn),傳統(tǒng)的基于數(shù)據(jù)聚類的生成視頻摘要的算法存在以下兩點(diǎn)不足:第一,無法根據(jù)輸入的不同時長和不同類型的視頻自適應(yīng)的得到最優(yōu)的聚類數(shù)目。第二,在算法中僅提取了圖像的顏色特征,忽略了圖像的紋理和形狀特征,而單特征既不能全面的表達(dá)圖像的視覺信息,又不能有效的消除圖像的噪聲,導(dǎo)致生成的視頻摘要的質(zhì)量低下。針對以上不足,本文提出了一種基于自適應(yīng)最優(yōu)聚類與多特征融合的視頻摘要生成算法。把視頻分解為圖像序列后,首先做預(yù)采樣處理。然后提取預(yù)采樣幀的顏色、HOG和LBP特征,并把這三種特征融合在一起來表征一幅圖像。接著使用零均值歸一化互相關(guān)指標(biāo)作為幀與幀之間相似性度量的標(biāo)準(zhǔn),把連續(xù)相似的幀分割為若干個鏡頭,從而得到最優(yōu)的聚類數(shù)目。接著使用改進(jìn)的k-means++算法對所有的幀進(jìn)行聚類,選取距離聚類中心最近的幀作為關(guān)鍵幀。最后,分別計(jì)算所有關(guān)鍵幀的顏色直方圖和梯度方向直方圖的歸一化方差,過濾掉無意義的關(guān)鍵幀。本文以匹配率和錯誤率兩個相互補(bǔ)充的指標(biāo)來衡量視頻摘要質(zhì)量的好壞。TRECVID2007數(shù)據(jù)集上的實(shí)驗(yàn)結(jié)果表明,本文算法具有良好的魯棒性,進(jìn)一步提高了所生成的視頻摘要的質(zhì)量。
[Abstract]:In recent years, the rapid development of the Internet and multimedia technology led to a surge in multimedia information data. Digital video, as the main multimedia information carrier, is widely used in all aspects of life. On the one hand, a large number of videos can enable people to obtain information in a richer form, which brings great convenience to life. On the other hand, it puts great pressure on video storage, transmission, archiving and retrieval. Therefore, video summary technology emerged as the times require. Similar to text summary, video summary is a summary and summary of original video content. By analyzing the original video data flow, a compact summary is made up of meaningful video content. Video abstracts can be used in video retrieval in combination with video tagging technology, and can also form independent products such as movie trailers, which can be used in our daily life. The traditional algorithm of generating video summary based on data clustering has the following two shortcomings: first, The optimal number of clusters can not be obtained adaptively according to different input time and different types of video. Secondly, only the color features of the image are extracted in the algorithm, and the texture and shape features of the image are ignored, while the single feature can not only fully express the visual information of the image, but also can not effectively eliminate the noise of the image. The resulting video digest is of low quality. In this paper, a video summary generation algorithm based on adaptive optimal clustering and multi-feature fusion is proposed. After the video is decomposed into image sequences, presampling is done first. Then the color hog and LBP features of the presampled frames are extracted, and the three features are fused together to represent an image. Then the zero mean normalized cross-correlation index is used as the criterion of similarity measurement between frames. The continuous similar frames are divided into several shots and the optimal clustering number is obtained. Then the improved k-means algorithm is used to cluster all frames and the nearest frame is selected as the key frame. Finally, the normalized variance of color histogram and gradient histogram of all key frames are calculated, and the meaningless key frames are filtered out.
【學(xué)位授予單位】：杭州電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前1條

1 李紹華;馮晶瑩;樓偶俊;景雨;;基于Contourlet變換和神經(jīng)網(wǎng)絡(luò)的視頻水印算法[J];電腦知識與技術(shù);2014年11期

相關(guān)博士學(xué)位論文 前1條

1 伍博;基于顯著性的視覺目標(biāo)跟蹤研究[D];電子科技大學(xué);2017年

相關(guān)碩士學(xué)位論文 前7條

1 高鵬輝;基于視頻監(jiān)控的智慧幼兒園安全檢測關(guān)鍵技術(shù)研究[D];浙江大學(xué);2017年

2 劉凱倫;視頻監(jiān)控中實(shí)時人臉識別系統(tǒng)的研究與實(shí)現(xiàn)[D];鄭州大學(xué);2017年

3 陳一根;基于壓縮感知的視頻追蹤方法研究[D];上海師范大學(xué);2017年

4 賈偉;煤礦井下視頻多目標(biāo)軌跡跟蹤方法研究與應(yīng)用[D];西安科技大學(xué);2017年

5 吳雙;基于MFC+OpenCV的視頻監(jiān)控區(qū)域入侵檢測系統(tǒng)設(shè)計(jì)與實(shí)現(xiàn)[D];山東師范大學(xué);2017年

6 王海懿;基于張量分解的視頻序列分類算法研究[D];天津大學(xué);2016年

7 張少華;基于機(jī)器對機(jī)器通信的智能電網(wǎng)視頻安全系統(tǒng)設(shè)計(jì)與實(shí)現(xiàn)[D];天津大學(xué);2016年

，

本文編號：1792906