圖像視頻中運(yùn)動(dòng)估計(jì)和分析
發(fā)布時(shí)間：2018-11-16 18:16
【摘要】：隨著多媒體技術(shù)和傳感器技術(shù)的飛速發(fā)展,圖像和視頻作為記錄視覺的載體,對(duì)人們的生活、生產(chǎn)等產(chǎn)生越來越重要的影響。近十年來,在各個(gè)方面的記錄視覺信息的需求,產(chǎn)生了爆炸式的圖像和視頻。在這些數(shù)據(jù)中,運(yùn)動(dòng)信息在圖像和視頻的生成和記錄過程中都扮演著重要的角色,可以說覆蓋了圖像和視頻處理的各個(gè)領(lǐng)域,而在語義上也涵蓋了從低層的運(yùn)動(dòng)軌跡估計(jì)到高層的運(yùn)動(dòng)理解。具體來說,在圖像生成過程中,相機(jī)運(yùn)動(dòng)會(huì)帶來圖像的模糊,而使得圖像的信息熵增大,為了能夠恢復(fù)出清楚的尖銳的圖像,我們往往需要精確估計(jì)相機(jī)的運(yùn)動(dòng)。在圖像的記錄過程中,有可能因?yàn)槲矬w或人的快速運(yùn)動(dòng)造成了圖像內(nèi)容的運(yùn)動(dòng)模糊。在記錄過程中產(chǎn)生的運(yùn)動(dòng)模糊往往比清楚圖像提供的信息更多因?yàn)槠洳蹲降搅藙?dòng)態(tài)的物體疊加。視頻通�？梢钥醋魇菆D像在時(shí)序上的疊加,類似地,在視頻產(chǎn)生過程中,相機(jī)的運(yùn)動(dòng)往往會(huì)導(dǎo)致畫面的晃動(dòng)和模糊,無論在圖像質(zhì)量上和視頻可觀賞性上都有一定的影響。而在視頻內(nèi)容記錄中,運(yùn)動(dòng)信息往往是視頻存在的理由,分析其運(yùn)動(dòng)往往更關(guān)注高層的語義。本課題以運(yùn)動(dòng)為核心,對(duì)圖像和視頻中的運(yùn)動(dòng)估計(jì)和分析展開了深入研究,具體包括:圖像生成過程中相機(jī)運(yùn)動(dòng)的建模和表征,模糊圖像中的相機(jī)運(yùn)動(dòng)估計(jì)和圖像復(fù)原,視頻內(nèi)容記錄中的運(yùn)動(dòng)的多層次表征學(xué)習(xí)以及運(yùn)動(dòng)的快速分析。本論文的主要工作和創(chuàng)新點(diǎn)可以總結(jié)為以下幾點(diǎn):1.在圖像成像過程中對(duì)相機(jī)運(yùn)動(dòng)的深層分析,我們提出了對(duì)運(yùn)動(dòng)核進(jìn)行分解并獨(dú)自優(yōu)化的模型。該分解模式能夠揭示相機(jī)成像的內(nèi)在特性,從而以一種全新的角度觀測(cè)經(jīng)典的圖像去模糊問題。為了展示該表征的優(yōu)勢(shì),我們提出了軌跡隨機(jī)擾動(dòng)算法來優(yōu)化運(yùn)動(dòng)核。在很多例子當(dāng)中,我們發(fā)現(xiàn)現(xiàn)有的去模糊算法落入局部極值的時(shí)候,我們的算法通過獨(dú)立優(yōu)化相機(jī)軌跡能夠取得較好的去模糊效果以及正確的模糊核。2.在圖像成像過程中利用高亮區(qū)域能夠較準(zhǔn)確地刻畫相機(jī)運(yùn)動(dòng)這一特點(diǎn),我們針對(duì)夜景這一非常具有挑戰(zhàn)性的場(chǎng)景結(jié)合高亮區(qū)域把該問題變得可行。我們提出了一個(gè)全新的框架有機(jī)地把從高亮區(qū)域中推斷出的運(yùn)動(dòng)核和非高亮區(qū)域結(jié)合求解更準(zhǔn)確的運(yùn)動(dòng)核,除此之外,我們提出了一個(gè)全新的函數(shù)化運(yùn)動(dòng)核表征從而較準(zhǔn)確地從高亮區(qū)域推斷運(yùn)動(dòng)核,我們提出了一個(gè)新的能量最小化方程能夠自動(dòng)地把提取的運(yùn)動(dòng)核分配給不同的區(qū)域以便進(jìn)行非均勻去模糊。3.在視頻內(nèi)容記錄中,我們側(cè)重分析了視頻內(nèi)容中的重要的運(yùn)動(dòng)信息:摔倒動(dòng)作檢測(cè),為了適應(yīng)視頻流場(chǎng)景下實(shí)時(shí)的運(yùn)動(dòng)分析,即實(shí)時(shí)的摔倒檢測(cè),我們把視頻內(nèi)容中運(yùn)動(dòng)信息按照"難易程度"分層,通過級(jí)聯(lián)的方式進(jìn)行動(dòng)作檢測(cè),不同于傳統(tǒng)的級(jí)聯(lián)框架,該級(jí)聯(lián)框架能夠支持不同復(fù)雜度的特征。通過這種混合特征的級(jí)聯(lián)框架,我們的系統(tǒng)在精確度和效率上能夠達(dá)到較好的折中。除此之外,我們精細(xì)地設(shè)計(jì)了我們采用的特征,支持特征復(fù)用以及增量式更新從而能夠?qū)σ曨l流場(chǎng)景具有較好地支持。最后,在摔倒動(dòng)作檢測(cè)的基礎(chǔ)之上,我們進(jìn)行了拓展從而能夠支持一般種類的動(dòng)作檢測(cè)以及引入了更多種類的特征從而在精確度上有一個(gè)更好的提升。本文針對(duì)圖像生成過程中和視頻內(nèi)容記錄中的運(yùn)動(dòng)進(jìn)行了深入的分析。大量的實(shí)驗(yàn)結(jié)果表明了我們對(duì)相機(jī)運(yùn)動(dòng)建模的有效性以及對(duì)視頻內(nèi)容中運(yùn)動(dòng)分層而快速檢測(cè)的高效性。
[Abstract]:With the rapid development of multimedia technology and sensor technology, image and video, as the carrier of recording vision, have become more and more important to people's life, production and so on. In the last decade, the demand for recording visual information in various aspects has resulted in an explosive image and video. In these data, motion information plays an important role in both the generation and recording of images and videos, and can be said to cover various fields of image and video processing, and in the semanteme it also covers the motion understanding from the low-level motion track to the high-level motion. In particular, in the image generation process, the camera motion causes the blurring of the image, and the information entropy of the image is increased, and in order to be able to recover a clear sharp image, it is often necessary to accurately estimate the motion of the camera. during the recording of the image it is possible that the motion blur of the image content is caused by the rapid movement of the object or person. The motion blur generated in the recording process tends to be more information than the information provided by the clear image because it captures the dynamic object superposition. The video is generally considered to be a superposition of the images in time series, similarly, during the production of the video, the motion of the camera often results in the shaking and blurring of the picture, whether in the image quality and in the video. In the video content record, the motion information is often the reason of the existence of the video, and the analysis of its motion tends to pay more attention to the semantics of the higher layer. This paper studies the motion estimation and analysis of the image and video with the motion as the core, including: the modeling and characterization of the camera motion in the process of image generation, the motion estimation of the camera in the blurred image, and the image restoration. The multi-level representation of the motion in the video content record and the rapid analysis of the motion. The main work and innovation points of this paper can be summarized as follows: 1. In the process of image imaging, the deep analysis of the motion of the camera is carried out, and the model of the motion kernel is decomposed and optimized by itself. The decomposition model can reveal the intrinsic characteristics of the camera imaging, so as to observe the classical image de-blurring problem with a brand-new angle. In order to demonstrate the advantages of the characterization, we propose a trajectory random perturbation algorithm to optimize the motion kernel. In many examples, when we find that the existing deblurring algorithm falls into the local extreme value, our algorithm can obtain better de-blurring effect and correct blur kernel by independently optimizing the camera track. This feature of the camera motion can be more accurately described by using the high-bright area in the image-forming process, and it becomes feasible to combine the problem with the high-bright area for the very challenging scene of the night scene. In addition, we put forward a new frame that organically combines the motion kernel and the non-high bright region inferred from the high bright region to find a more accurate motion kernel. In addition, we propose a new functional motion kernel representation to accurately infer the motion kernel from the high bright region, We propose a new energy minimization equation that can automatically assign the extracted motion kernel to different areas for non-uniform deblurring. In the video content record, we focus on the analysis of the important motion information in the video content: the fall motion detection, in order to adapt to the real-time motion analysis in the video stream scene, that is, the real-time fall detection, we divide the motion information in the video content according to the "degree of difficulty", The action detection is carried out in a cascade manner, which is different from the traditional cascade framework, which can support the characteristics of different complexity. Through the cascading frame of this kind of mixing characteristic, our system can achieve a good compromise on the accuracy and efficiency. In addition, we designed the features we adopted, support the feature multiplexing and incremental updating, so that the video stream scene can be well supported. Finally, on the basis of the detection of the fall motion, we have developed so as to be able to support the general kinds of motion detection and to introduce more kinds of features so as to have a better improvement in the accuracy. This paper makes an in-depth analysis of the motion in the process of image generation and in the video content record. A large number of experiments show the effectiveness of the motion modeling of the camera and the high efficiency of rapid detection of motion layering in the video content.
【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】：博士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41
[Abstract]:With the rapid development of multimedia technology and sensor technology, image and video, as the carrier of recording vision, have become more and more important to people's life, production and so on. In the last decade, the demand for recording visual information in various aspects has resulted in an explosive image and video. In these data, motion information plays an important role in both the generation and recording of images and videos, and can be said to cover various fields of image and video processing, and in the semanteme it also covers the motion understanding from the low-level motion track to the high-level motion. In particular, in the image generation process, the camera motion causes the blurring of the image, and the information entropy of the image is increased, and in order to be able to recover a clear sharp image, it is often necessary to accurately estimate the motion of the camera. during the recording of the image it is possible that the motion blur of the image content is caused by the rapid movement of the object or person. The motion blur generated in the recording process tends to be more information than the information provided by the clear image because it captures the dynamic object superposition. The video is generally considered to be a superposition of the images in time series, similarly, during the production of the video, the motion of the camera often results in the shaking and blurring of the picture, whether in the image quality and in the video. In the video content record, the motion information is often the reason of the existence of the video, and the analysis of its motion tends to pay more attention to the semantics of the higher layer. This paper studies the motion estimation and analysis of the image and video with the motion as the core, including: the modeling and characterization of the camera motion in the process of image generation, the motion estimation of the camera in the blurred image, and the image restoration. The multi-level representation of the motion in the video content record and the rapid analysis of the motion. The main work and innovation points of this paper can be summarized as follows: 1. In the process of image imaging, the deep analysis of the motion of the camera is carried out, and the model of the motion kernel is decomposed and optimized by itself. The decomposition model can reveal the intrinsic characteristics of the camera imaging, so as to observe the classical image de-blurring problem with a brand-new angle. In order to demonstrate the advantages of the characterization, we propose a trajectory random perturbation algorithm to optimize the motion kernel. In many examples, when we find that the existing deblurring algorithm falls into the local extreme value, our algorithm can obtain better de-blurring effect and correct blur kernel by independently optimizing the camera track. This feature of the camera motion can be more accurately described by using the high-bright area in the image-forming process, and it becomes feasible to combine the problem with the high-bright area for the very challenging scene of the night scene. In addition, we put forward a new frame that organically combines the motion kernel and the non-high bright region inferred from the high bright region to find a more accurate motion kernel. In addition, we propose a new functional motion kernel representation to accurately infer the motion kernel from the high bright region, We propose a new energy minimization equation that can automatically assign the extracted motion kernel to different areas for non-uniform deblurring. In the video content record, we focus on the analysis of the important motion information in the video content: the fall motion detection, in order to adapt to the real-time motion analysis in the video stream scene, that is, the real-time fall detection, we divide the motion information in the video content according to the "degree of difficulty", The action detection is carried out in a cascade manner, which is different from the traditional cascade framework, which can support the characteristics of different complexity. Through the cascading frame of this kind of mixing characteristic, our system can achieve a good compromise on the accuracy and efficiency. In addition, we designed the features we adopted, support the feature multiplexing and incremental updating, so that the video stream scene can be well supported. Finally, on the basis of the detection of the fall motion, we have developed so as to be able to support the general kinds of motion detection and to introduce more kinds of features so as to have a better improvement in the accuracy. This paper makes an in-depth analysis of the motion in the process of image generation and in the video content record. A large number of experiments show the effectiveness of the motion modeling of the camera and the high efficiency of rapid detection of motion layering in the video content.
【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】：博士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 孫英皓;唐棣;;一種有效的交互式圖像局部?jī)?nèi)容放大方法[J];計(jì)算機(jī)工程與應(yīng)用;2012年12期
2 李瑞帥;余淑華;潘凱;王圓;;圖像特征金字塔快速計(jì)算方法[J];電子世界;2014年07期
3 曾喜良;王金娟;;指紋分割的塊圖像梯度因子聚類法[J];計(jì)算機(jī)與數(shù)字工程;2008年07期
4 劉陳;王欣欣;李鳳霞;趙相坤;;一種快速保邊的圖像對(duì)象分割方法[J];北京理工大學(xué)學(xué)報(bào);2010年02期
5 朱薇;劉利剛;;圖像適應(yīng)算法中非冗余顯著圖的計(jì)算[J];中國(guó)圖象圖形學(xué)報(bào);2011年08期
6 柳有權(quán);吳宗勝;韓紅雷;吳恩華;;線條增強(qiáng)的建筑物圖像抽象畫生成[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2013年09期
7 吳駿,唐紅梅,肖志濤,賈志成;一種基于相位信息的圖像對(duì)稱性檢測(cè)方法[J];信號(hào)處理;2004年01期
8 邵靜;高雋;趙瑩;張旭東;;一種基于圖像固有維度的感知物體檢測(cè)方法[J];儀器儀表學(xué)報(bào);2008年04期
9 潘如如;高衛(wèi)東;;高緊度機(jī)織物圖像傾斜的自動(dòng)糾正[J];紡織學(xué)報(bào);2009年10期
10 劉貴喜,趙曙光,楊萬海;基于梯度塔形分解的多傳感器圖像融合[J];光電子·激光;2001年03期
相關(guān)會(huì)議論文 前5條
1 張一鳴;劉亞t，
本文編號(hào)：2336262
本文鏈接：http://sikaile.net/shoufeilunwen/xxkjbs/2336262.html
最近更新
教材專著
