本文選題：視頻目標(biāo)分割 + 光流　； 參考：《中國(guó)科學(xué)技術(shù)大學(xué)》2017年碩士論文

【摘要】：隨著互聯(lián)網(wǎng)基礎(chǔ)設(shè)施的逐步升級(jí)以及移動(dòng)終端的快速普及,人們可以越來(lái)越方便地拍攝和觀看視頻。視頻由于其本身所攜帶信息的豐富性和生動(dòng)性,成為了人們生活中重要的信息傳播載體之一。不斷增長(zhǎng)的海量視頻數(shù)據(jù)也帶來(lái)了如何識(shí)別、檢索和理解視頻內(nèi)容的需求。如何降低視頻內(nèi)容理解難度,提煉出視頻中的關(guān)鍵信息成為當(dāng)前視頻處理領(lǐng)域的重要研究課題。由于視頻目標(biāo)分割的研究目標(biāo)是有效分割出具有顯著性特征的前景目標(biāo),所以它在視頻摘要、視頻檢索、動(dòng)作分析和視頻語(yǔ)義理解等領(lǐng)域擁有廣泛的應(yīng)用。當(dāng)前的視頻目標(biāo)分割算法大多屬于自底向上的方法,通過(guò)獲取并分析視頻中顏色和邊緣特征、運(yùn)動(dòng)信息等底層特征分割出具有顯著性特點(diǎn)的前景目標(biāo)。傳統(tǒng)基于人工標(biāo)注的算法已經(jīng)不能滿足當(dāng)前大規(guī)模視頻數(shù)據(jù)環(huán)境下的應(yīng)用需求。同時(shí),海量視頻中包含的場(chǎng)景和拍攝條件是復(fù)雜而多樣的,使得當(dāng)前的自動(dòng)化視頻目標(biāo)分割算法并不能在一些復(fù)雜場(chǎng)景中仍保持較好的魯棒性。針對(duì)上述問(wèn)題,本文提出了兩種適用于不同場(chǎng)景的視頻目標(biāo)自動(dòng)分割算法。主要研究工作和創(chuàng)新點(diǎn)如下:1.現(xiàn)有基于圖割的算法容易受到背景噪聲和像素點(diǎn)失配的干擾,在一些復(fù)雜場(chǎng)景下魯棒性不佳。本文提出了 一種基于光流場(chǎng)和圖割的視頻目標(biāo)自動(dòng)分割算法,針對(duì)上述問(wèn)題做了改進(jìn)。在對(duì)前景目標(biāo)分割前,該算法預(yù)先對(duì)視頻全局動(dòng)作特征進(jìn)行分析,獲得了前景目標(biāo)的先驗(yàn)知識(shí),減少了背景噪聲對(duì)算法的干擾。針對(duì)像素點(diǎn)失配問(wèn)題,該算法提出了動(dòng)態(tài)位置模型優(yōu)化機(jī)制,利用前景目標(biāo)的位置模型增強(qiáng)了分割結(jié)果的時(shí)域連續(xù)性。實(shí)驗(yàn)表明,該算法在鏡頭快速移動(dòng)、前景目標(biāo)運(yùn)動(dòng)特征不規(guī)律等場(chǎng)景下能夠獲得更加準(zhǔn)確和魯棒的分割結(jié)果。2.在一些復(fù)雜場(chǎng)景下,現(xiàn)有基到候選目標(biāo)的算法往往會(huì)出現(xiàn)分割結(jié)果部分缺失的問(wèn)題,這一問(wèn)題的根源在于候選目標(biāo)過(guò)于碎片化以及候選目標(biāo)間的時(shí)域映射關(guān)系不夠準(zhǔn)確。本文提出了一種基于候選目標(biāo)的改進(jìn)算法。該算法對(duì)原生候選目標(biāo)進(jìn)行了時(shí)域擴(kuò)展與合并,不僅改善了候選目標(biāo)碎片化的問(wèn)題,還提高了相鄰幀間候選目標(biāo)的時(shí)域連續(xù)性。為了進(jìn)一步增強(qiáng)模型時(shí)域映射關(guān)系的準(zhǔn)確性,該算法引入了更多圖像特征用于度量模型的邊權(quán)值。在多個(gè)基準(zhǔn)數(shù)據(jù)集上的實(shí)驗(yàn)表明,相較于現(xiàn)有同類算法,該算法對(duì)背景噪聲的抗噪能力更強(qiáng),在背景環(huán)境復(fù)雜、水面倒影等場(chǎng)景中分割結(jié)果更加完整。
[Abstract]:With the gradual upgrading of Internet infrastructure and the rapid popularity of mobile terminals, people can more and more easily shoot and watch video. Because of the richness and vividness of the information it carries, video has become one of the important carriers of information dissemination in people's life. The growing mass of video data also brings the demand of how to identify, retrieve and understand the video content. How to reduce the difficulty of video content understanding and extract the key information of video has become an important research topic in the field of video processing. Because the research goal of video target segmentation is to segment the foreground target with significant features, it has a wide range of applications in video summarization, video retrieval, action analysis and video semantic understanding. Most of the current video target segmentation algorithms belong to bottom-up methods. By obtaining and analyzing the bottom features such as color edge feature and motion information the foreground target with significant characteristics is segmented. The traditional algorithm based on manual annotation can not meet the needs of the current large-scale video data environment. At the same time, the scene and shooting conditions included in the massive video are complex and diverse, which makes the current automated video target segmentation algorithm can not maintain good robustness in some complex scenes. In order to solve the above problems, this paper proposes two automatic video object segmentation algorithms for different scenes. The main research work and innovation are as follows: 1. The existing algorithms based on graph cutting are vulnerable to background noise and pixel mismatch, and are not robust in some complex scenarios. In this paper, an automatic video object segmentation algorithm based on optical flow field and graph cutting is proposed and improved. Before segmenting the foreground target, the algorithm analyzes the global motion features of the video in advance, obtains the prior knowledge of the foreground target, and reduces the interference of the background noise to the algorithm. To solve the problem of pixel mismatch, the algorithm proposes a dynamic position model optimization mechanism, which enhances the continuity of segmentation results in time domain by using the position model of foreground target. Experimental results show that the proposed algorithm can obtain more accurate and robust segmentation results. In some complex scenarios, the existing algorithms based to candidate targets often have the problem of partial absence of segmentation results. The root of the problem lies in the fragmentation of candidate targets and the inaccuracy of time domain mapping relationship between candidate targets. This paper presents an improved algorithm based on candidate targets. The algorithm extends and combines the original candidate targets in time domain, which not only improves the fragmentation of candidate targets, but also improves the continuity of candidate targets between adjacent frames in time domain. In order to further enhance the accuracy of the temporal mapping of the model, the algorithm introduces more image features to measure the boundary weights of the model. Experiments on several datum datasets show that the proposed algorithm is more robust to background noise than the existing algorithms, and the segmentation results are more complete in the background environment, water surface reflection and other scenes.
【學(xué)位授予單位】：中國(guó)科學(xué)技術(shù)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前2條

1 周治平;施小鳳;;基于超像素的目標(biāo)跟蹤方法研究[J];光電工程;2013年12期

2 高尚兵;周靜波;嚴(yán)云洋;;一種新的基于超像素的譜聚類圖像分割算法[J];南京大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年02期

相關(guān)碩士學(xué)位論文 前1條

1 宋巖軍;變分光流計(jì)算的數(shù)學(xué)模型及相關(guān)數(shù)值方法[D];青島大學(xué);2007年

，

本文編號(hào)：1839541