本文選題：Kinect + 深度信息��； 參考：《北京郵電大學(xué)》2016年碩士論文

【摘要】：隨著多媒體技術(shù)的發(fā)展和數(shù)字傳媒業(yè)的科技化,數(shù)字摳像和合成作為圖像處理領(lǐng)域炙手可熱的研究課題,在電影、游戲和廣告制作業(yè),以及醫(yī)療衛(wèi)生和空間探測等多領(lǐng)域都有廣泛的應(yīng)用。數(shù)字摳像技術(shù)是指通過算法在圖像和視頻中提取帶有透明度的感興趣前景。傳統(tǒng)的自然摳像算法大都需要用戶提供劃分好的三元圖或涂鴉信息作為提示信息,耗費(fèi)大量的時(shí)間和人力成本,且很難應(yīng)用于視頻摳像。隨著深度信息獲取技術(shù)的發(fā)展,一系列結(jié)合深度信息的RGBD摳像算法應(yīng)運(yùn)而出。本文主要研究基于Kinect深度信息的自然圖像摳像算法,最終實(shí)現(xiàn)無需人工輸入輔助信息的全自動摳像系統(tǒng)。論文主要完成工作如下:(1)研究了體感外設(shè)Kinect的起源、發(fā)展和應(yīng)用,比較并總結(jié)了一代和二代傳感器的硬件配置和深度圖像獲取技術(shù)。針對實(shí)驗(yàn)中Kinect二代的深度圖像誤差,將結(jié)合深度信息的RGB-D引導(dǎo)濾波引入到深度圖像的修復(fù)中,并將其進(jìn)行迭代計(jì)算以提高對深度圖邊緣和內(nèi)部空洞的修復(fù)精度。(2)分析及對比了幾種傳統(tǒng)自然圖像摳像技術(shù)的原理和應(yīng)用場景,并深入研究了基于TOF的深度摳像算法,將Kinect獲取的深度信息引入傳統(tǒng)的RGB摳像算法中。綜合考慮摳像的實(shí)時(shí)性和精度,選擇了 Shared Matting作為本文研究算法。(3)實(shí)現(xiàn)了三元圖的自動生成算法,并將深度信息引入Shared Matting,提出了基于深度域改進(jìn)的區(qū)域擴(kuò)張方法。最終實(shí)現(xiàn)了基于Kinect深度信息的全自動摳像算法,經(jīng)測試本算法具有較好的魯棒性和實(shí)時(shí)性。(4)設(shè)計(jì)并搭建了帶有可視化界面的Kinect實(shí)時(shí)摳像和合成系統(tǒng)。針對摳像算法中的人物前景與新的虛擬背景之間存在的色調(diào)和光照條件的不一致,在系統(tǒng)中引入保持色調(diào)一致性的Erik Reinhard色調(diào)遷移算法。最終實(shí)現(xiàn)了 Kinect彩色視頻流的實(shí)時(shí)人體摳像和合成,并能對前后景色調(diào)一致的新的“人景合一”的圖像進(jìn)行存儲。
[Abstract]:With the development of multimedia technology and the technology of digital media industry, digital matting and synthesis as a hot research topic in the field of image processing, in film, games and advertising industry, As well as medical and health and space exploration and other fields have a wide range of applications. Digital matting technology is used to extract the interesting foreground with transparency in image and video. Most of the traditional natural matting algorithms require users to provide divided ternary or graffiti information as prompt information, which cost a lot of time and manpower, and it is difficult to be applied to video matting. With the development of depth information acquisition technology, a series of RGBD matting algorithms combined with depth information should be carried out. This paper mainly studies the natural image matting algorithm based on Kinect depth information, and finally realizes the automatic matting system without manual input of auxiliary information. The main work of this paper is as follows: (1) the origin, development and application of Kinect are studied, and the hardware configuration and depth image acquisition techniques of generation and generation 2 sensors are compared and summarized. Aiming at the error of Kinect second generation depth image, the RGB-D guided filter combined with depth information is introduced into the depth image restoration. In order to improve the repairing accuracy of depth map edge and inner cavity, the principles and application scenes of several traditional natural image matting techniques are analyzed and compared, and the depth matting algorithm based on TOF is deeply studied. The depth information obtained by Kinect is introduced into the traditional RGB matting algorithm. Considering the real-time and precision of matting, this paper chooses shared matching as the algorithm to realize the automatic generation of ternary images, and introduces the depth information into shared tracking, and proposes an improved region expansion method based on depth domain. Finally, the automatic matting algorithm based on Kinect depth information is realized. The Kinect real-time matting and synthesis system with visual interface is designed and built by testing the algorithm has good robustness and real-time performance. Aiming at the inconsistency of hue and illumination conditions between the foreground of characters in matting algorithm and the new virtual background, Erik Reinhard hue migration algorithm is introduced in the system. Finally, the real-time human matting and synthesis of Kinect color video stream is realized, and the new "human scene in one" image which is consistent with the front and rear scenery can be stored.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 夏倩;許勇;夏玉勤;;基于Kinect的自動視頻摳像算法[J];計(jì)算機(jī)工程與設(shè)計(jì);2015年05期

2 張展鵬;朱青松;謝耀欽;;數(shù)字摳像的最新研究進(jìn)展[J];自動化學(xué)報(bào);2012年10期

3 何貝;王貴錦;林行剛;;結(jié)合Kinect深度圖的快速視頻摳圖算法[J];清華大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年04期

4 周星;高志軍;;立體視覺技術(shù)的應(yīng)用與發(fā)展[J];工程圖學(xué)學(xué)報(bào);2010年04期

5 林生佑;潘瑞芳;杜輝;石教英;;數(shù)字摳圖技術(shù)綜述[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2007年04期

6 鄭寧斐;姜光;;顏色遷移算法的研究[J];電視技術(shù);2006年11期

相關(guān)碩士學(xué)位論文 前4條

1 趙旭;Kinect深度圖像修復(fù)技術(shù)研究[D];大連理工大學(xué);2013年

2 陳理;Kinect深度圖像增強(qiáng)算法研究[D];湖南大學(xué);2013年

3 張約倫;基于Kinect的摳像算法研究[D];西安電子科技大學(xué);2013年

4 丁津津;TOF三維攝像機(jī)的誤差分析及補(bǔ)償方法研究[D];合肥工業(yè)大學(xué);2011年

，

本文編號：2003481