本文關(guān)鍵詞：基于FPGA的人臉識別算法的設(shè)計與實(shí)現(xiàn)　出處：《中國科學(xué)技術(shù)大學(xué)》2016年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： FPGA ARM 人臉識別 膚色檢測 離散余弦

【摘要】：在信息爆炸式增長的當(dāng)今,信息安全已經(jīng)成為一個非常重要的課題,而人臉識別是生物特征識別的重要識別方法之一,相對于其他識別方法,比如指紋識別、基因識別、語音識別以及虹膜識別等,人臉識別具有不需要與人接觸,操作簡單、可跟蹤以及識別高等優(yōu)點(diǎn),因而被廣泛應(yīng)用于國家安全、安保、出入口檢測等領(lǐng)域。傳統(tǒng)PC端的人臉識別方法已經(jīng)比較成熟,但隨著電子消費(fèi)等概念的誕生,這些方法都需要移植到嵌入式的設(shè)備中。PC機(jī)一般不能滿足嵌入式的要求,而DSP受到資源的限制,不能滿足高速視頻流的數(shù)據(jù)處理。而近二十年來,FGPA得到了高速的發(fā)展,性能和集成度不斷提高,使得FPGA處理圖像和視頻成為可能。FPGA具有邏輯資源多,運(yùn)算能力強(qiáng),速度快,靈活性和可移植性等特點(diǎn),而且目前很多家廠商都提供了FPGA+ARM的嵌入式設(shè)計,用戶可以很容易的實(shí)現(xiàn)軟硬件一體的協(xié)同設(shè)計。使用FPGA對圖像和視頻進(jìn)行處理已成為一種趨勢。比較通用的人臉識別的過程主要有人臉檢測與定位、圖像預(yù)處理、人臉特征提取以及人臉匹配等幾部分。在人臉的檢測與定位部分,本文主要根據(jù)膚色這一顯著特征來實(shí)現(xiàn)人臉區(qū)域的分割,并通過形態(tài)學(xué)濾波等手段實(shí)現(xiàn)對圖像中的人臉進(jìn)行定位。圖像預(yù)處理部分則實(shí)現(xiàn)了高斯濾波、中值濾波、圖像的直方圖均衡化算法以及canny邊緣檢測算子。在人臉特征提取部分,首先用MATLAB軟件實(shí)現(xiàn)了主成分分析、獨(dú)立成分分析、非負(fù)矩陣分解以及離散余弦變換四種特征提取方法,并將這些算法應(yīng)用于ORL人臉數(shù)據(jù)庫,其次分析了這些算法提取的特征向量維度以及訓(xùn)練集的大小對最終人臉識別率的影響,對比四種算法,離散余弦算法對ORL的數(shù)據(jù)庫識別效果最好,識別率最高能達(dá)到97.5%,最后對DCT算法進(jìn)行了FPGA的移植。將FPGA中提取的人臉特征向量通過AXI總線傳到ARM中,并與ARM中的人臉數(shù)據(jù)庫進(jìn)行人臉匹配。本文最后在ZYNQ平臺上實(shí)現(xiàn)了一個簡要的人臉識別系統(tǒng),該系統(tǒng)對實(shí)驗(yàn)室的人臉進(jìn)行識別,系統(tǒng)最終能夠達(dá)到91.6%的識別率。
[Abstract]:In the information explosion today, information security has become a very important topic, and face recognition is one of the important recognition method of biometric recognition, compared with other identification methods, such as fingerprint identification, gene recognition, speech recognition and iris recognition, face recognition has no need to contact with people, simple operation, can be tracking and recognition of the advantages, so it has been widely used in national security, security, access detection etc.. The traditional PC face recognition method is more mature, but with the birth of the concept of electronic consumption, these methods need to be transplanted into embedded devices. PC can not meet the requirements of embedded system, but DSP is limited by resources and can not meet the data processing of high speed video stream. In the past twenty years, FGPA has been developing rapidly, and its performance and integration are increasing, making it possible for FPGA to deal with images and video. FPGA has many advantages, such as many logical resources, strong computing power, fast speed, flexibility and portability, and many manufacturers have provided FPGA+ARM's embedded design. Users can easily achieve collaborative design of hardware and software. The use of FPGA to process images and video has become a trend. The general process of face recognition mainly includes face detection and location, image preprocessing, face feature extraction and face matching. In the part of face detection and location, this paper mainly realizes the segmentation of face regions according to the salient feature of skin color, and realizes the location of faces in images by morphological filtering. The image preprocessing part realizes the Gauss filter, median filter, image histogram equalization algorithm and Canny edge detection operator. In the part of extracting face features, firstly, using MATLAB software, principal component analysis, independent component analysis, non negative matrix factorization and discrete cosine transform four feature extraction methods, and these algorithms are applied to ORL face database, then analyzes these algorithms to extract the feature vector dimension and the size of the training set of end face the recognition rate, the comparison of four algorithms, the best database recognition algorithm for ORL discrete cosine, the highest recognition rate can reach 97.5%, and finally the DCT algorithm is the transplantation of FPGA. The face feature vectors extracted from FPGA are passed through the AXI bus to ARM, and the face database is matched with the face database in ARM. Finally, a brief face recognition system is implemented on the ZYNQ platform. The system can recognize the face of the laboratory, and the system can reach 91.6% recognition rate.
【學(xué)位授予單位】：中國科學(xué)技術(shù)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9條

1 宋士軍;劉東明;姜爾赤;石剛;;基于FPGA的視頻圖像邊緣提取算法實(shí)現(xiàn)[J];電子世界;2013年09期

2 李昌盛;王亞娟;黃啟俊;常勝;;基于FPGA的人臉檢測系統(tǒng)設(shè)計[J];電子技術(shù)應(yīng)用;2011年10期

3 張素文;陳志星;蘇義鑫;;Canny邊緣檢測算法的改進(jìn)及FPGA實(shí)現(xiàn)[J];紅外技術(shù);2010年02期

4 齊永鋒;張家樹;火元蓮;;一種基于膚色與改進(jìn)的LBP的人臉檢測方法[J];光電子.激光;2009年06期

5 祁蒙;劉鑫;姜學(xué)東;;灰度梯度自適應(yīng)閾值二值化算法的FPGA實(shí)現(xiàn)[J];激光與紅外;2008年09期

6 王洪;郭娟;賴宗聲;李小進(jìn);趙建龍;李光顯;;基于FPGA的指紋識別算法硬件實(shí)現(xiàn)[J];微電子學(xué)與計算機(jī);2007年04期

7 劉直芳,游志勝,王運(yùn)瓊;基于PCA和ICA的人臉識別[J];激光技術(shù);2004年01期

8 劉青山,盧漢清,馬頌德;綜述人臉識別中的子空間方法[J];自動化學(xué)報;2003年06期

9 周杰,盧春雨,張長水,李衍達(dá);人臉自動識別方法綜述[J];電子學(xué)報;2000年04期

相關(guān)博士學(xué)位論文 前1條

1 周昌軍;基于圖像重構(gòu)和特征融合的人臉識別方法研究[D];大連理工大學(xué);2008年

相關(guān)碩士學(xué)位論文 前3條

1 李鋼;基于FPGA的人臉識別門禁系統(tǒng)的設(shè)計[D];成都理工大學(xué);2012年

2 靳艷紅;基于改進(jìn)canny算法的圖像邊緣檢測的研究[D];重慶師范大學(xué);2011年

3 韓建強(qiáng);智能人臉識別算法及其FPGA的實(shí)現(xiàn)[D];上海交通大學(xué);2009年

，

本文編號：1342307