【摘要】：人臉識(shí)別因其非接觸性、易采集等優(yōu)點(diǎn)而被廣泛研究,并主要應(yīng)用于考勤、門禁、監(jiān)控、公安等系統(tǒng)。雖然目前眾多種人臉識(shí)別算法已取得較好的識(shí)別性能,但人臉識(shí)別系統(tǒng)在實(shí)際應(yīng)用中仍面臨很多挑戰(zhàn),其中主要包括:由光照變化、飾品等引起的遮擋人臉識(shí)別問題;非可控條件下可采集到的樣本數(shù)少,即小樣本問題;姿勢(shì)和表情變換等,本文將這種情況下的人臉識(shí)別稱之為欠完備采樣人臉識(shí)別。欠完備采樣會(huì)造成人臉信息的缺失,降低已有算法的識(shí)別性能。為此提出面向數(shù)據(jù)的稀疏表示算法對(duì)該問題展開研究以提高人臉識(shí)別算法的魯棒性和實(shí)用性,具體研究?jī)?nèi)容如下:(1)根據(jù)近鄰表示和分辨性分解算法分別提出基于近鄰類加權(quán)結(jié)構(gòu)稀疏表示圖像識(shí)別算法和基于分辨性分解結(jié)構(gòu)稀疏表示遮擋人臉識(shí)別算法。字典中各類訓(xùn)練樣本對(duì)測(cè)試樣本分類的貢獻(xiàn)不同,一般近鄰樣本對(duì)測(cè)試樣本的正確分類具有較大的促進(jìn)作用,因此考慮選取近鄰類并加權(quán)進(jìn)行測(cè)試樣本分類,不僅可降低算法的計(jì)算復(fù)雜度,同時(shí)提高算法識(shí)別率。此外,為提高遮擋情況下的人臉識(shí)別性能,采用分辨性分解算法對(duì)遮擋部位進(jìn)行分離,并在分解得到的共同部分和低秩條件部分上分別進(jìn)行主成分分析,并計(jì)算投影矩陣,最后在投影空間上進(jìn)行結(jié)構(gòu)稀疏表示并分類。(2)為解決全局算法對(duì)遮擋的敏感性,并進(jìn)一步降低遮擋對(duì)識(shí)別性能的影響,對(duì)圖像進(jìn)行分塊局部處理,通過對(duì)遮擋模塊賦予低權(quán)值,干凈模塊賦予高權(quán)值來降低遮擋模塊對(duì)算法性能的影響。為此,本文提出幾種不同的模塊加權(quán)方案:首先,將圖像分割成多個(gè)有重疊的模塊,并利用Fisher率計(jì)算每個(gè)模塊的分辨性,據(jù)此對(duì)每個(gè)模塊加權(quán),保留分辨性高的模塊進(jìn)行后續(xù)的分類識(shí)別;其次,將圖像分割成4部分,并利用稀疏殘差對(duì)模塊加權(quán)進(jìn)而對(duì)遮擋部分進(jìn)行估計(jì),最后僅在非遮擋部位上進(jìn)行分類判別;最后,將以上兩種加權(quán)方案聯(lián)合,提出基于Fisher判別和稀疏殘差的模塊加權(quán)算法,該算法聯(lián)合了Fisher加權(quán)和殘差加權(quán)的優(yōu)勢(shì),以進(jìn)一步提高遮擋檢測(cè)性能。(3)為精確檢測(cè)遮擋部位,并實(shí)現(xiàn)無遮擋訓(xùn)練集上的遮擋人臉識(shí)別,提出兩種像素級(jí)上的遮擋檢測(cè)算法:基于稀疏表示的像素級(jí)遮擋檢測(cè)人臉識(shí)別及基于塊遞推殘差分析的雙層稀疏表示分類算法。基于稀疏表示的像素級(jí)遮擋檢測(cè)算法根據(jù)類殘差分析各類遮擋估計(jì)結(jié)果,并對(duì)結(jié)果進(jìn)行統(tǒng)計(jì)得出最終的像素遮擋估計(jì),最后僅在非遮擋像素集上進(jìn)行識(shí)別�；趬K遞推殘差分析的算法將遮擋樣本分成上下兩個(gè)模塊,利用稀疏度較高的模塊重構(gòu)整幅測(cè)試圖像,并根據(jù)殘差估計(jì)遮擋像素進(jìn)而對(duì)各像素進(jìn)行加權(quán)并分類,以此提高遮擋人臉的識(shí)別性能。像素級(jí)遮擋檢測(cè)可避免分塊遮擋檢測(cè)算法中模塊中同時(shí)含遮擋和非遮擋部分而造成的識(shí)別率低的問題。(4)利用核空間對(duì)塊稀疏表示算法進(jìn)行非線性擴(kuò)展,并提出核塊稀疏表示算法(KBSRC:Kernel Block Sparse Representation based Classification),該算法將樣本投影到降維的核空間,因而可將樣本的原非線性空間線性化,而在該空間上利用樣本的結(jié)構(gòu)信息分類可提高分類性能。
[Abstract]:Face recognition is widely studied because of its non-contact and easy acquisition. It is mainly applied to the system of attendance, entrance guard, monitoring and public security. Although a large number of face recognition algorithms have acquired better recognition performance, the face recognition system still faces many challenges in the practical application, which mainly comprises the problems of shielding face recognition caused by illumination changes, ornaments, and the like; and the number of samples that can be acquired under the non-controllable condition is small, In this paper, the face recognition in this case is called the under-complete sampling face recognition. Under-complete sampling can cause the loss of face information and reduce the recognition performance of the existing algorithm. To this end, a sparse representation algorithm for data is proposed to study the problem to improve the robustness and practicability of the face recognition algorithm. (1) based on the nearest neighbor representation and the resolution decomposition algorithm, a sparse representation image recognition algorithm based on a neighbor class weighting structure and a sparse representation shielding face recognition algorithm based on the resolution decomposition structure are respectively proposed. The contribution of various training samples in the dictionary to the classification of the test samples is different, and the common neighbor samples have a great effect on the correct classification of the test samples, therefore, considering the selection of the nearest neighbor class and weighting the test sample classification, not only can the computational complexity of the algorithm be reduced, Improve that recognition rate of the algorithm at the same time. In addition, in order to improve the face recognition performance in the case of occlusion, the shielding part is separated by a resolution decomposition algorithm, and the main component analysis is carried out on the common part and the low-rank condition part obtained by the decomposition, and the projection matrix is calculated, And finally, structural sparse representation and classification are carried out on the projection space. (2) in order to solve the sensitivity of the global algorithm to the occlusion, and further reduce the influence of the occlusion on the recognition performance, the image is segmented and partially processed, and the high-weight value is given by the clean module to reduce the influence of the blocking module on the performance of the algorithm by giving a low weight to the shielding module. To this end, several different module weighting schemes are proposed: first, the image is divided into a plurality of modules with overlapping, and the resolution of each module is calculated by using the Fisher rate, the method comprises the following steps of: dividing an image into four parts, weighting the module by using a sparse residual error to estimate the shielding part, and finally carrying out classification judgment on the non-shielding part; and finally, combining the two weighting schemes, and putting forward a module weighting algorithm based on the Fisher discrimination and the sparse residual, The algorithm combines the advantages of Fisher's weight and residual weight to further improve the shielding performance. And (3) in order to accurately detect the occlusion region and realize the occlusion face recognition on the non-occlusion training set, the occlusion detection algorithm on the two pixel levels is proposed, namely, the pixel-level occlusion detection face recognition based on the sparse representation and the double-layer sparse representation classification algorithm based on the block recursive residual analysis. The pixel-level occlusion detection algorithm based on the sparse representation analyzes the various occlusion estimation results according to the class residual, and then counts the results to obtain the final pixel occlusion estimation, and finally, the pixel-level occlusion detection algorithm is only identified on the non-occlusion pixel set. Based on the algorithm of the block recursive residual analysis, the occlusion sample is divided into upper and lower modules, the whole image is reconstructed by using a module with higher sparsity, and the occlusion pixel is estimated to be weighted and classified according to the residual estimated occlusion pixel so as to improve the identification performance of the occlusion face. The pixel-level occlusion detection can avoid the problem of low recognition rate caused by the blocking and non-blocking part in the module in the block-blocking detection algorithm. (4) The kernel space is used for non-linear expansion of the block sparse representation algorithm, and a kernel block sparse representation algorithm (KBSRC: Kernel Block Sparse Representation based Classification) is proposed, and the sample is projected into the reduced-dimension nuclear space, so that the original non-linear space of the sample can be linearized, And the classification performance can be improved by using the structure information classification of the samples in the space.
【學(xué)位授予單位】：燕山大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李映;張艷寧;許星;;基于信號(hào)稀疏表示的形態(tài)成分分析:進(jìn)展和展望[J];電子學(xué)報(bào);2009年01期

2 趙瑞珍;王飛;羅阿理;張彥霞;;基于稀疏表示的譜線自動(dòng)提取方法[J];光譜學(xué)與光譜分析;2009年07期

3 楊蜀秦;寧紀(jì)鋒;何東健;;基于稀疏表示的大米品種識(shí)別[J];農(nóng)業(yè)工程學(xué)報(bào);2011年03期

4 史加榮;楊威;魏宗田;;基于非負(fù)稀疏表示的人臉識(shí)別[J];計(jì)算機(jī)工程與設(shè)計(jì);2012年05期

5 高志榮;熊承義;笪邦友;;改進(jìn)的基于殘差加權(quán)的稀疏表示人臉識(shí)別[J];中南民族大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年03期

6 朱杰;楊萬扣;唐振民;;基于字典學(xué)習(xí)的核稀疏表示人臉識(shí)別方法[J];模式識(shí)別與人工智能;2012年05期

7 耿耀君;張軍英;袁細(xì)國(guó);;一種基于稀疏表示系數(shù)的特征相關(guān)性測(cè)度[J];模式識(shí)別與人工智能;2013年01期

8 張疆勤;廖海斌;李原;;基于因子分析與稀疏表示的多姿態(tài)人臉識(shí)別[J];計(jì)算機(jī)工程與應(yīng)用;2013年05期

9 李正周;王會(huì)改;劉梅;丁浩;金鋼;;基于形態(tài)成分稀疏表示的紅外小弱目標(biāo)檢測(cè)[J];彈箭與制導(dǎo)學(xué)報(bào);2013年04期

10 胡正平;趙淑歡;李靜;;基于塊稀疏遞推殘差分析的稀疏表示遮擋魯棒識(shí)別算法研究[J];模式識(shí)別與人工智能;2014年01期

相關(guān)會(huì)議論文 前3條

1 何愛香;劉玉春;魏廣芬;;基于稀疏表示的煤矸界面識(shí)別研究[A];虛擬運(yùn)營(yíng)與云計(jì)算——第十八屆全國(guó)青年通信學(xué)術(shù)年會(huì)論文集（上冊(cè)）[C];2013年

2 樊亞翔;孫浩;周石琳;鄒煥新;;基于元樣本稀疏表示的多視角目標(biāo)識(shí)別[A];2013年中國(guó)智能自動(dòng)化學(xué)術(shù)會(huì)議論文集（第五分冊(cè)）[C];2013年

3 葛鳳翔;任歲玲;郭鑫;郭良浩;孫波;;微弱信號(hào)處理及其研究進(jìn)展[A];中國(guó)聲學(xué)學(xué)會(huì)水聲學(xué)分會(huì)2013年全國(guó)水聲學(xué)學(xué)術(shù)會(huì)議論文集[C];2013年

相關(guān)博士學(xué)位論文 前10條

1 李進(jìn)明;基于稀疏表示的圖像超分辨率重建方法研究[D];重慶大學(xué);2015年

2 王亞寧;基于信號(hào)稀疏表示的電機(jī)故障診斷研究[D];河北工業(yè)大學(xué);2014年

3 姚明海;視頻異常事件檢測(cè)與認(rèn)證方法研究[D];東北師范大學(xué);2015年

4 黃國(guó)華;蛋白質(zhì)翻譯后修飾位點(diǎn)與藥物適應(yīng)癥預(yù)測(cè)方法研究[D];上海大學(xué);2015年

5 王瑾;基于稀疏表示的數(shù)據(jù)收集、復(fù)原與壓縮研究[D];北京工業(yè)大學(xué);2015年

6 王文卿;基于融合框架與稀疏表示的遙感影像銳化[D];西安電子科技大學(xué);2015年

7 解虎;高維小樣本陣列自適應(yīng)信號(hào)處理方法研究[D];西安電子科技大學(xué);2015年

8 秦振濤;基于稀疏表示及字典學(xué)習(xí)遙感圖像處理關(guān)鍵技術(shù)研究[D];成都理工大學(xué);2015年

9 薛明;基于稀疏表示的在線目標(biāo)跟蹤研究[D];上海交通大學(xué);2014年

10 孫樂;空譜聯(lián)合先驗(yàn)的高光譜圖像解混與分類方法[D];南京理工大學(xué);2014年

相關(guān)碩士學(xué)位論文 前10條

1 吳麗璇;基于稀疏表示的微聚焦X射線圖像去噪方法[D];華南理工大學(xué);2015年

2 趙孝磊;基于圖像分塊稀疏表示的人臉識(shí)別算法研究[D];南京信息工程大學(xué);2015年

3 黃志明;基于辨別式稀疏字典學(xué)習(xí)的視覺追蹤算法研究[D];華南理工大學(xué);2015年

4 張鈴華;非約束環(huán)境下的稀疏表示人臉識(shí)別算法研究[D];南京信息工程大學(xué);2015年

5 賀妍斐;基于稀疏表示與自適應(yīng)倒易晶胞的遙感圖像復(fù)原方法研究[D];南京信息工程大學(xué);2015年

6 楊爍;電能質(zhì)量擾動(dòng)信號(hào)的稀疏表示/壓縮采樣研究[D];西南交通大學(xué);2015年

7 應(yīng)艷麗;基于低秩稀疏表示的目標(biāo)跟蹤算法研究[D];西南交通大學(xué);2015年

8 梁曉捷;基于網(wǎng)絡(luò)攝像頭與稀疏表示分類法的實(shí)時(shí)人臉識(shí)別系統(tǒng)應(yīng)用研究[D];五邑大學(xué);2015年

9 張宏樂;語(yǔ)音信號(hào)稀疏表示方法研究[D];太原理工大學(xué);2016年

10 郭欣;基于K-SVD稀疏表示的語(yǔ)音增強(qiáng)算法研究[D];太原理工大學(xué);2016年

，

本文編號(hào)：2504931