本文選題：深度學(xué)習(xí) + BP神經(jīng)網(wǎng)絡(luò)��； 參考：《蘭州交通大學(xué)》2017年碩士論文

【摘要】：人臉識別是生物識別的一個重要研究方向,隨著眾多學(xué)者的不斷努力和長期探索,人臉識別取得了諸多成就,但對于問題的徹底解決還存在一定的距離。近年來諸多學(xué)者在神經(jīng)網(wǎng)絡(luò)的基礎(chǔ)上提出了深度學(xué)習(xí),由于其超強的學(xué)習(xí)能力,現(xiàn)階段已成為神經(jīng)網(wǎng)絡(luò)的主要研究方向。深度學(xué)習(xí)的提出,為人臉識別問題的徹底解決提供了新的思路。本文利用深度學(xué)習(xí)算法進行人臉識別,進一步改善人臉識別效果。本文的主要研究內(nèi)容為:(1)本文首先利用現(xiàn)階段性價比最高的BP神經(jīng)網(wǎng)絡(luò)(Back Propagation,BP)來進行人臉識別研究。針對BP網(wǎng)絡(luò)在人臉識別過程中,人臉圖像數(shù)據(jù)量大和易陷入局部最優(yōu)的問題,提出利用主成分分析法(Principal Component Analysis,PCA)和遺傳算法(Genetic Algorithm,GA)來優(yōu)化BP網(wǎng)絡(luò),構(gòu)成PCA-GA-BP網(wǎng)絡(luò)。該網(wǎng)絡(luò)首先利用PCA算法來處理人臉圖像,減少人臉圖像數(shù)據(jù)量,再通過GA算法對BP網(wǎng)絡(luò)進行優(yōu)化,提高網(wǎng)絡(luò)性能,最后利用AR數(shù)據(jù)庫和ORL數(shù)據(jù)庫進行實驗。實驗結(jié)果表明,該算法不僅可以克服BP網(wǎng)絡(luò)的缺陷,還能進一步提高人臉識別精度。(2)針對PCA-GA-BP網(wǎng)絡(luò)在訓(xùn)練樣本逐漸增大的情況下,其學(xué)習(xí)能力小范圍下降的問題,提出利用具有超強學(xué)習(xí)能力的深度信念網(wǎng)絡(luò)(Deep Belief Networks,DBNs)來替換BP網(wǎng)絡(luò),構(gòu)成PCA-GA-DBNs網(wǎng)絡(luò)。該網(wǎng)絡(luò)首先利用GA算法和Gibbs采樣來實現(xiàn)網(wǎng)絡(luò)的逐層訓(xùn)練,訓(xùn)練完成后再利用BP網(wǎng)絡(luò)進行微調(diào),使其成為最優(yōu)網(wǎng)絡(luò)。然后通過AR數(shù)據(jù)庫和ORL數(shù)據(jù)庫進行實驗,實驗結(jié)果表明,PCA-GA-DBNs網(wǎng)絡(luò)能夠很好的提高人臉識別精度,實驗最后還分析了不同分類器對人臉識別結(jié)果的影響。(3)針對在較大訓(xùn)練樣本情況下,GA算法爬山能力不足,容易出現(xiàn)早熟收斂的問題,提出利用全局搜索能力更強且不會陷入局部最優(yōu)的模擬退火遺傳算法(Simulated Annealing Genetic Algorithm,SAGA)來代替GA算法,構(gòu)成PCA-SAGA-DBNs網(wǎng)絡(luò)。該網(wǎng)絡(luò)利用SAGA算法結(jié)合Gibbs采樣來逐層訓(xùn)練網(wǎng)絡(luò),訓(xùn)練完成后再利用BP網(wǎng)絡(luò)對其進行微調(diào)并構(gòu)造分類器。然后以AR數(shù)據(jù)庫和ORL數(shù)據(jù)庫為實驗對象,實驗結(jié)果表明該網(wǎng)絡(luò)可以很好的克服GA算法爬山能力不足早熟收斂的缺陷,提高人臉識別精度。最后將本文改進的三種算法進行實驗,通過實驗結(jié)果比較得出PCA-SAGA-DBNs網(wǎng)絡(luò)不僅具有良好的識別效果,還具有較好的穩(wěn)定性,是一種較優(yōu)的人臉識別方法。
[Abstract]:Face recognition is an important research direction of biometrics. With the continuous efforts and long-term exploration of many scholars, face recognition has made many achievements, but there is still a certain distance to solve the problem thoroughly. In recent years, many scholars have put forward deep learning on the basis of neural network. The stage has become the main research direction of neural network. Advanced learning provides a new idea for the complete solution of face recognition. This paper uses depth learning algorithm to carry out face recognition to further improve the effect of face recognition. The main contents of this paper are as follows: (1) this paper first uses the BP nerve with the highest cost performance at the present stage. Back Propagation (BP) for face recognition research. Aiming at the problem of large amount of face image data and easy to fall into local optimal in the process of face recognition in BP network, the principle component analysis (Principal Component Analysis, PCA) and genetic algorithm (Genetic Algorithm, GA) are used to optimize the BP network. Collaterals first use PCA algorithm to deal with face images, reduce the amount of face image data, and then optimize the BP network by GA algorithm to improve the network performance. Finally, the experiment is carried out using the AR database and ORL database. The experimental results show that the algorithm can not only overcome the lack of BP network, but can further improve the accuracy of face recognition. (2) P In the case of the gradual increase of training samples, CA-GA-BP network has the problem of decreasing learning ability, and proposes to replace BP network by using the Deep Belief Networks (DBNs) with super learning ability to form a PCA-GA-DBNs network. The network first uses GA algorithm and Gibbs sampling to train the network by layer by layer training and training. After the completion of the practice, the BP network is used to make the fine-tuning to make it the best network. Then the experiment is carried out through the AR database and the ORL database. The experimental results show that the PCA-GA-DBNs network can improve the face recognition accuracy well. Finally, the experiment also analyzes the influence of different classifiers on the face recognition results. (3) for the larger training sample situation, the experiment results are also analyzed. Under the condition of the GA algorithm, the ability to climb mountains is insufficient and the problem of premature convergence is easy to appear. It is proposed that the simulated annealing genetic algorithm (Simulated Annealing Genetic Algorithm, SAGA), which makes use of the global search ability and will not fall into the local optimal, to replace the GA algorithm to form the PCA-SAGA-DBNs network. The network uses SAGA algorithm to combine Gibbs sampling to layer by layer. Training network, after the completion of training, the BP network is used to fine tune it and construct the classifier. Then the AR database and the ORL database are used as the experimental objects. The experimental results show that the network can overcome the defects of the premature convergence of the climbing ability of the GA algorithm and improve the accuracy of face recognition. Finally, the three improved algorithms in this paper are implemented. It is proved that the PCA-SAGA-DBNs network not only has good recognition effect, but also has good stability. It is a better face recognition method.
【學(xué)位授予單位】：蘭州交通大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41;TP181

【參考文獻】

相關(guān)期刊論文 前10條

1 隋煜舜;齊蘇敏;;基于人臉檢測的模板匹配人臉跟蹤算法研究[J];電子技術(shù);2016年12期

2 馬壯飛;梁栗炎;羅延豐;徐曼;;基于心電信號的身份識別系統(tǒng)研究[J];中國醫(yī)療設(shè)備;2016年06期

3 耿志強;張怡康;;一種基于膠質(zhì)細(xì)胞鏈的改進深度信念網(wǎng)絡(luò)模型[J];自動化學(xué)報;2016年06期

4 王培良;夏春江;;基于PCA-PDBNs的故障檢測與自學(xué)習(xí)辨識[J];儀器儀表學(xué)報;2015年05期

5 葉純青;;淘寶智能開戶將引入“人臉識別”技術(shù)[J];金融科技時代;2015年05期

6 張春霞;姬楠楠;王冠偉;;受限波爾茲曼機[J];工程數(shù)學(xué)學(xué)報;2015年02期

7 劉奕君;趙強;郝文利;;基于遺傳算法優(yōu)化BP神經(jīng)網(wǎng)絡(luò)的瓦斯?jié)舛阮A(yù)測研究[J];礦業(yè)安全與環(huán)保;2015年02期

8 周鑫;馬躍;胡毅;;求解車間作業(yè)調(diào)度問題的混合遺傳模擬退火算法[J];小型微型計算機系統(tǒng);2015年02期

9 尹寶才;王文通;王立春;;深度學(xué)習(xí)研究綜述[J];北京工業(yè)大學(xué)學(xué)報;2015年01期

10 許滬敏;楊森;朱濤;;基于Delaunay網(wǎng)格和DLP的三維指紋識別系統(tǒng)設(shè)計[J];計算機測量與控制;2014年10期

相關(guān)博士學(xué)位論文 前1條

1 段錦;人臉自動識別中若干問題的研究[D];吉林大學(xué);2004年

相關(guān)碩士學(xué)位論文 前7條

1 陳慶一;基于RBM的文本分類算法研究[D];吉林大學(xué);2015年

2 王t熞，

本文編號：1952962