發(fā)布時(shí)間：2018-02-01 14:53

  本文關(guān)鍵詞： 深度神經(jīng)網(wǎng)絡(luò) 人臉識(shí)別 深度多模型融合 卷積神經(jīng)網(wǎng)絡(luò) 組合特征　出處：《西南交通大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著人工智能的快速發(fā)展,如何準(zhǔn)確、有效的識(shí)別用戶身份,提升信息安全成為一項(xiàng)重要的研究課題。相較于傳統(tǒng)的卡片識(shí)別、指紋識(shí)別和虹膜識(shí)別,人臉識(shí)別具有許多優(yōu)點(diǎn)。它的非接觸性、非強(qiáng)制性和并發(fā)性,易被用戶所接受,已廣泛應(yīng)用于教育、電子商務(wù)等多個(gè)領(lǐng)域。深度學(xué)習(xí)是機(jī)器學(xué)習(xí)領(lǐng)域中的新興分支之一。與傳統(tǒng)淺層網(wǎng)絡(luò)不同,深度學(xué)習(xí)受人腦工作機(jī)制啟發(fā),構(gòu)建了深層網(wǎng)絡(luò)結(jié)構(gòu)和相應(yīng)的訓(xùn)練方法。深度卷積神經(jīng)網(wǎng)絡(luò)(deep convolution neural networks,DCNN)源于多層前向網(wǎng)絡(luò),經(jīng)過(guò)不斷發(fā)展,已成為當(dāng)前圖像識(shí)別領(lǐng)域的研究熱點(diǎn)。它依靠深層非線性網(wǎng)絡(luò)結(jié)構(gòu)和大規(guī)模的訓(xùn)練數(shù)據(jù),實(shí)現(xiàn)復(fù)雜函數(shù)逼近,從而獲得更本質(zhì)和魯棒的圖像特征,有效提升了后續(xù)分類與識(shí)別的效果。近年來(lái)隨著深度卷積神經(jīng)網(wǎng)絡(luò)的引入,人臉識(shí)別的準(zhǔn)確率得以跨越式提升。然而,不同模型的訓(xùn)練集和網(wǎng)絡(luò)結(jié)構(gòu)差異較大,使得每個(gè)模型都有各自的特點(diǎn)。對(duì)此,本文研究了一種基于深度多模型融合的人臉識(shí)別方法,通過(guò)融合多個(gè)人臉識(shí)別模型提取的特征構(gòu)成組合特征,再利用深度神經(jīng)網(wǎng)絡(luò)訓(xùn)練組合特征構(gòu)建人臉識(shí)別分類器,可以得到融合多個(gè)模型優(yōu)點(diǎn)的改進(jìn)模型。主要的工作如下:(1)分析和對(duì)比基于卷積神經(jīng)網(wǎng)絡(luò)且開源的人臉識(shí)別算法,通過(guò)實(shí)驗(yàn)篩選了 2種基礎(chǔ)模型。對(duì)基礎(chǔ)模型提取的基礎(chǔ)特征進(jìn)行降維、歸一化、融合,得到組合特征,作為后續(xù)深度神經(jīng)網(wǎng)絡(luò)的輸入。(2)構(gòu)建基于深度多模型融合的深度神經(jīng)網(wǎng)絡(luò),訓(xùn)練組合特征,獲得融合不同基礎(chǔ)模型優(yōu)點(diǎn)的改進(jìn)模型。(3)進(jìn)一步分析改進(jìn)模型并設(shè)計(jì)了多組實(shí)驗(yàn),包括不同的訓(xùn)練集、DNN參數(shù)和基礎(chǔ)特征權(quán)重。統(tǒng)計(jì)了基礎(chǔ)模型和改進(jìn)模型在LFW數(shù)據(jù)集上的詳細(xì)測(cè)試數(shù)據(jù),探索改進(jìn)模型提升的原因。在采用較小數(shù)據(jù)集的情況下,本文方法在人臉識(shí)別權(quán)威測(cè)試集LFW和YTF上獲得了 99.1%和93.32%的精度,相對(duì)于基礎(chǔ)模型分別提高0.57%和0.52%。而且通過(guò)對(duì)LFW測(cè)試數(shù)據(jù)的進(jìn)一步分析,探討了改進(jìn)模型在融合不同基礎(chǔ)模型優(yōu)點(diǎn)方面的有效性。
[Abstract]:With the rapid development of artificial intelligence, how to accurately and effectively identify users and improve information security has become an important research topic. Compared with traditional card recognition, fingerprint recognition and iris recognition. Face recognition has many advantages. Its non-contact, non-mandatory and concurrent, easy to be accepted by users, has been widely used in education. E-commerce and other fields. Deep learning is one of the new branches in the field of machine learning. Unlike the traditional shallow network, deep learning is inspired by the working mechanism of human brain. The deep convolution neural networks and deep convolution neural network were constructed. DCNN, which originates from multilayer forward network, has become a research hotspot in the field of image recognition through continuous development. It relies on deep nonlinear network structure and large-scale training data to achieve complex function approximation. In recent years, with the introduction of deep convolution neural network, the accuracy of face recognition can be improved by leaps and bounds. Different models have different training sets and network structure, which makes each model have their own characteristics. In this paper, a face recognition method based on depth multi-model fusion is proposed. The features extracted from multiple face recognition models are fused to form the combined features, and then the combined features are trained by the depth neural network to construct the face recognition classifier. The main work is as follows: 1) analyze and compare the face recognition algorithm based on convolution neural network and open source. Through experiments, two basic models are selected. The basic features extracted from the basic model are reduced, normalized, fused, and combined features are obtained. As the input of the subsequent depth neural network, we construct the depth neural network based on depth multi-model fusion, and train the combination features. An improved model combining the advantages of different basic models is obtained. (3) further analysis of the improved model and design of a number of experiments, including different training sets. DNN parameters and basic feature weights. The detailed test data of the basic model and the improved model on the LFW data set are analyzed to explore the reasons for the improvement of the improved model. In the case of using smaller data sets. In this paper, the accuracy of 99.1% and 93.32% is obtained on the face recognition authoritative test set LFW and YTF. Compared with the basic model, the improvement is 0.57% and 0.52 respectively. The effectiveness of the improved model in combining the advantages of different basic models is discussed through further analysis of the LFW test data.
【學(xué)位授予單位】：西南交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41;TP181

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 蘇楠;吳冰;徐偉;蘇光大;;人臉識(shí)別綜合技術(shù)的發(fā)展[J];信息安全研究;2016年01期

2 郭麗麗;丁世飛;;深度學(xué)習(xí)研究進(jìn)展[J];計(jì)算機(jī)科學(xué);2015年05期

3 余凱;賈磊;陳雨強(qiáng);徐偉;;深度學(xué)習(xí)的昨天、今天和明天[J];計(jì)算機(jī)研究與發(fā)展;2013年09期

，

本文編號(hào)：1482129