本文選題：圖像分析 + 人臉識別　； 參考：《南京航空航天大學(xué)》2016年博士論文

【摘要】：隨著信息時(shí)代的到來,計(jì)算機(jī)成為了現(xiàn)代人類社會生活中必不可少的信息處理工具。由于現(xiàn)代通信技術(shù)的進(jìn)步和互聯(lián)網(wǎng)的普及,圖像日漸成為人們?nèi)粘Ｉ钪薪佑|最多的信息載體。相比于傳統(tǒng)的文字載體,圖像作為信息載體具有明顯的優(yōu)勢:直觀性——圖像能夠直接反應(yīng)現(xiàn)場的情景;全面性—-圖像能夠全面和細(xì)致的重現(xiàn)場景;通用性—-圖像不受國界和語言的影響;便捷性—-圖像中的內(nèi)容更加便于理解。因此,運(yùn)用計(jì)算機(jī)實(shí)現(xiàn)圖像的自動(dòng)分析和處理成為社會智能化發(fā)展的基礎(chǔ)。與文字的處理不同,圖像分析和處理的自動(dòng)化具有更大的挑戰(zhàn)性。首先,人們?nèi)狈\(yùn)用計(jì)算機(jī)實(shí)現(xiàn)圖像處理和分析的認(rèn)識:人類對于文字的使用和處理具有數(shù)千年的歷史,積累了豐富的經(jīng)驗(yàn),而數(shù)字圖像的出現(xiàn)和計(jì)算機(jī)科學(xué)的發(fā)展最多也只有一百年的歷史。如何運(yùn)用和發(fā)展計(jì)算機(jī)科學(xué)技術(shù),充分把握圖像的特點(diǎn)提取其蘊(yùn)含的信息,始終是計(jì)算機(jī)視覺發(fā)展的目標(biāo)和動(dòng)力。其次,人們?nèi)狈\(yùn)用計(jì)算機(jī)實(shí)現(xiàn)圖像處理和分析的方法指導(dǎo):目前,人們主要借鑒人類圖像認(rèn)知系統(tǒng)的運(yùn)行方式來實(shí)現(xiàn)圖像分析與處理的自動(dòng)化,但是,人類的圖像認(rèn)知系統(tǒng)經(jīng)過了幾百萬年的進(jìn)化,是一個(gè)非常復(fù)雜的系統(tǒng),而計(jì)算機(jī)科學(xué)的發(fā)展歷史與其相比只是滄海一粟。如何有效的把握人類圖像認(rèn)知系統(tǒng)的實(shí)質(zhì),將其用于圖像處理將會是一個(gè)長期擺在計(jì)算機(jī)視覺發(fā)展道路上的難題。從現(xiàn)有的圖像表示方法上來看,圖像特征表示的提取主要來自于兩個(gè)方面:模仿人類圖像認(rèn)知器官的圖像內(nèi)部結(jié)構(gòu)提取方法—-特征描述子,以及模仿人類神經(jīng)系統(tǒng)處理圖像的方式—-淺層和深層學(xué)習(xí)。以圖像分析的一個(gè)基本應(yīng)用—-人臉識別系統(tǒng)的設(shè)計(jì)為背景,從淺層學(xué)習(xí)入手,運(yùn)用線性重建的方式改進(jìn)了人臉識別系統(tǒng)中的人臉對齊和圖像表示。具體來說,貢獻(xiàn)如下:(1)提出了一種新的流形學(xué)習(xí)方法,并將其應(yīng)用于人臉對齊問題。目前,淺層學(xué)習(xí)方法通常假設(shè)樣本的空間結(jié)構(gòu)是線性的。這種方式雖然降低了數(shù)據(jù)處理的復(fù)雜度,但是數(shù)據(jù)間的拓?fù)浣Y(jié)構(gòu)卻被忽視了。事實(shí)上,高維數(shù)據(jù)通常具有一定流形結(jié)構(gòu),最為明顯的例子就是人臉形狀向量空間。然而,人臉對齊的參數(shù)模型中,形狀模型依然假設(shè)人臉形狀空間是線性的。流形學(xué)習(xí)作為一種非線性嵌入方法,能夠有效的將高維數(shù)據(jù)通過非線性降維嵌入到流形空間,從而得到線性結(jié)構(gòu)的數(shù)據(jù),但是需要估計(jì)數(shù)據(jù)流形空間的維度,因此計(jì)算復(fù)雜度較大而無法滿足實(shí)時(shí)性要求。通過平滑局部子流形,在局部切空間排列的基礎(chǔ)上提出了一種新的流形學(xué)習(xí)方法。由于其顯式的投影矩陣以及在原空間中的流形變換,使得它能夠很好的與人臉對齊方法中形狀模型相結(jié)合,從而將人臉形狀的流形結(jié)構(gòu)嵌入到模型中去。(2)提出了一種改進(jìn)的空間非負(fù)矩陣分解方法�；诰�性重建的表示學(xué)習(xí)方法中,非負(fù)矩陣分解是一種專門針對圖像數(shù)據(jù)的特征學(xué)習(xí)方法。與以往的表示方法相比,非負(fù)矩陣分解的基圖像具有更好的局部結(jié)構(gòu),因此非負(fù)矩陣分解作為一種基于局部的表示學(xué)習(xí)方法,其學(xué)到的圖像表示向量具有更好的魯棒性和可理解性。為了進(jìn)一步改進(jìn)基圖像的局部性,對于非負(fù)矩陣分解的改進(jìn)目前主要集中將圖像的空間信息嵌入到基圖像中。然而,這些空間信息通常來自于圖像的二維網(wǎng)絡(luò)結(jié)構(gòu),因此缺乏與數(shù)據(jù)內(nèi)容的聯(lián)系。對此,根據(jù)因子分析對圖像特征之間關(guān)系的提取,提出了一種結(jié)合數(shù)據(jù)特征分布與空間結(jié)構(gòu)信息的空間正則化方法,并將其與大間隔約束相結(jié)合,不但實(shí)現(xiàn)了空間結(jié)構(gòu)的嵌入,判別性與局部性的融合,還降低了判別性約束和局部性約束對數(shù)據(jù)表示產(chǎn)生的矛盾影響。(3)提出了一種新的屬性特征。與傳統(tǒng)的特征描述子抽取的特征相比,屬性是一種更高層次的特征,它所概括的不是圖像中蘊(yùn)含的某種幾何結(jié)構(gòu),而是圖像中某種語義信息的體現(xiàn)度。由于這種特點(diǎn),屬性特征對于人類來說具有更好的解釋性。然而,對于語義的定義各不相同,而且很多語義是相對抽象的概念,因此屬性的學(xué)習(xí)通常非常復(fù)雜和不準(zhǔn)確。具體來說,對于連續(xù)屬性的學(xué)習(xí),需要對每個(gè)屬性分別提取相應(yīng)的特征并學(xué)習(xí)各自的屬性分類器,以分類器的輸出作為每個(gè)屬性在樣本上的體現(xiàn)度。對于選取哪些屬性作為樣本的特征,哪些特征最能體現(xiàn)每個(gè)屬性以及屬性分類器的設(shè)計(jì),都會在不同程度上影響樣本的屬性質(zhì)量。于是,基于心理學(xué)中的原型理論,提出了一種類相對關(guān)系屬性—-原型相對屬性。其中,每個(gè)屬性分別體現(xiàn)了樣本與已知各類的相關(guān)度,而不必在屬性池中搜索問題相關(guān)的屬性,同時(shí),每個(gè)屬性都使用相同的特征表示樣本,因此在一定程度上簡化了屬性學(xué)習(xí)過程。
[Abstract]:With the arrival of the information age, computer has become an indispensable tool for information processing in modern human life. Because of the progress of modern communication technology and the popularity of the Internet, image has become the most important information carrier in people's daily life. Compared with the traditional carrier, the image is a carrier of information. Advantage: intuition - the image can react directly to the scene of the scene; comprehensiveness - the image can reproduce the scene in an all-round and meticulous way; universality - the image is not affected by the national boundary and the language; the convenience - the content in the image is more convenient to understand. Therefore, the automatic analysis and processing of the computer real image becomes social intelligentization. Different from the processing of words, the automation of image analysis and processing is more challenging. First, people lack the knowledge of computer image processing and analysis: the use and processing of words have a history of thousands of years, rich experience, and the appearance of digital images and computer science. It has a history of one hundred years at most. How to use and develop the computer science and technology to fully grasp the features of the image to extract the information contained in it is always the goal and motive of the development of computer vision. Secondly, people lack the guidance of computer image processing and analysis. At present, people are mainly drawing on human drawings. Like the operation mode of the cognitive system to automate the image analysis and processing, the human image cognition system has evolved for millions of years, it is a very complicated system, and the history of computer science is just a drop in the sea. How to effectively grasp the essence of human image cognitive system and use it Image processing will be a difficult problem on the road of computer vision for a long time. From the existing image representation method, the extraction of image feature representation mainly comes from two aspects: image internal structure extraction method imitating human image cognitive organs - feature descriptors, and imitation of human neural system processing map Based on the design of face recognition system, a basic application of image analysis, which is based on the design of face recognition system, improves face alignment and image representation in face recognition system by means of linear reconstruction. In particular, the contribution is as follows: (1) a new manifold learning method is proposed. At present, the shallow layer learning method is usually assumed that the spatial structure of the sample is linear. This method reduces the complexity of data processing, but the topology structure of the data is ignored. In fact, the high dimensional data usually has a certain manifold structure, and the most obvious example is the shape direction of the face. However, in the parameter model of face alignment, the shape model still assumes that the shape space of the face is linear. As a nonlinear embedding method, manifold learning can effectively embed the high dimensional data into the manifold space through the nonlinear dimensionality, thus obtaining the data of the linear structure, but it is necessary to estimate the dimension of the data manifold space. By smoothing the local submanifolds, a new manifold learning method is proposed on the basis of the local tangent space arrangement by smoothing the local submanifolds. Because of the explicit projection matrix and the manifold transformation in the original space, it can make it good for the shape model in the face alignment method. In addition, the manifold structure of the face shape is embedded into the model. (2) an improved spatial nonnegative matrix decomposition method is proposed. In the representation learning method of linear reconstruction, the nonnegative matrix decomposition is a characteristic learning method for image data. Compared with the previous representation method, the basic image of the nonnegative matrix decomposition is compared. It has better local structure, so the non negative matrix decomposition is a local representation learning method, and the image representation vector has better robustness and comprehensibility. In order to further improve the locality of the base image, the spatial information of the image is embedded in the base for the improvement of the non negative matrix decomposition. However, the spatial information is usually derived from the two-dimensional network structure of the image, and therefore lacks the connection with the data content. According to the factor analysis, the extraction of the relationship between image features is extracted. A spatial regularization method is proposed, which combines the feature distribution of the data and the spatial structure information, and combines it with the large interval constraints. It not only realizes the embedding of spatial structure, the fusion of discriminability and locality, but also reduces the conflicting effects of discriminative and local constraints on data representation. (3) a new attribute feature is proposed. Compared with the characteristics of traditional feature descriptor extraction, the attribute is a higher level feature, and it is not a graph. A certain geometric structure contained in the image, but a representation of some semantic information in the image. Because of this, attribute features have better interpretative properties for human beings. However, the definition of the semantics is different, and many semantics are relatively abstract concepts, so the learning of attributes is usually very complex and inaccurate. For the learning of continuous attributes, it is necessary to extract the corresponding characteristics of each attribute and learn their respective attribute classifiers, with the output of the classifier as the embodiment of each attribute on the sample. For which attributes are selected as the characteristics of the sample, which features are most capable of each attribute and the design of the attribute classifier, The quality of the samples is influenced to different degrees. Based on the prototype theory in psychology, a relative attribute of relative relation - the relative attribute of the prototype is proposed, in which each attribute reflects the correlation between the samples and the known types, and does not have to search the related attributes in the attribute pool, and each attribute uses the phase. The same features represent samples, thus simplifying the attribute learning process to a certain extent.

【學(xué)位授予單位】：南京航空航天大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2016
【分類號】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 賈宇峰,劉少君,郭堯君;雙向電泳圖像分析軟件[J];現(xiàn)代科學(xué)儀器;2000年05期

2 徐培渝,張立實(shí);圖像分析系統(tǒng)在研究生畢業(yè)課題中的應(yīng)用淺析[J];現(xiàn)代預(yù)防醫(yī)學(xué);2001年02期

3 薄立華,郭志良,崔亞南;生物醫(yī)學(xué)圖像分析工作者應(yīng)具備的計(jì)算機(jī)基本技能[J];中國體視學(xué)與圖像分析;2004年02期

4 ;2009圖像分析與信號處理國際會議4月召開[J];中國印刷與包裝研究;2009年01期

5 陳武凡;秦安;江少峰;馮前進(jìn);郝立巍;;醫(yī)學(xué)圖像分析的現(xiàn)狀與展望[J];中國生物醫(yī)學(xué)工程學(xué)報(bào);2008年02期

6 ;2010圖像分析與信號處理國際會議(英文)[J];智能系統(tǒng)學(xué)報(bào);2009年04期

7 張寧;劉文萍;;基于圖像分析的植物葉片識別技術(shù)綜述[J];計(jì)算機(jī)應(yīng)用研究;2011年11期

8 王剛;潘亮星;劉星舟;;利用公共圖像分析軟件計(jì)算金屬晶粒大小的方法[J];現(xiàn)代商貿(mào)工業(yè);2012年01期

9 邢彩虹,李桂蘭,紀(jì)之瑩,高耘,尹松年;單細(xì)胞凝膠電泳圖像分析軟件的比較[J];衛(wèi)生毒理學(xué)雜志;2005年02期

10 溫宏愿;趙琦;陳延如;周木春;張猛;許凌飛;;光譜圖像分析用于轉(zhuǎn)爐終點(diǎn)實(shí)時(shí)預(yù)測[J];光電工程;2008年05期

相關(guān)會議論文 前10條

1 劉國權(quán);;體視學(xué)與圖像分析基本原理及若干應(yīng)用問題之討論[A];第九屆中國體視學(xué)與圖像分析學(xué)術(shù)會議論文集[C];2001年

2 羅永剛;;醫(yī)學(xué)圖像分析的未來趨勢[A];山東省計(jì)算機(jī)學(xué)會2005年信息技術(shù)與信息化研討會論文集（二）[C];2005年

3 Roscoe Atkinson;;法醫(yī)學(xué)與病理學(xué)中圖像分析及病理信息學(xué)的潛能[A];法庭科學(xué)最新技術(shù)研討培訓(xùn)班專家講課提綱[C];2006年

4 張維琴;王淑云;李京晉;魏靈哲;張文義;;數(shù)據(jù)處理技術(shù)在粒度圖像分析中的應(yīng)用[A];第二屆全國信息獲取與處理學(xué)術(shù)會議論文集[C];2004年

5 李劍鋒;李乃民;王寬全;張宏志;;面向舌圖像分析的改進(jìn)色度計(jì)算公式[A];第四次全國中西醫(yī)結(jié)合診斷學(xué)術(shù)研討會論文集[C];2010年

6 王曉民;胡文華;;生物醫(yī)學(xué)圖像分析視頻采集設(shè)備的發(fā)展概況[A];第九屆中國體視學(xué)與圖像分析學(xué)術(shù)會議論文集[C];2001年

7 黎軍英;徐穎;洪健;;JD-801形態(tài)學(xué)圖像分析系統(tǒng)在TEM圖像分析中的應(yīng)用[A];2006年全國電子顯微學(xué)會議論文集[C];2006年

8 李l勆，

本文編號：1884198