Basic Search / Detailed Display

Author: 簡韜
Thesis Title: 基於深度學習的Android行動裝置人臉偵測與辨識
Face Detection And Recognition in Android Mobile Device based on Deep Neural Network
Advisor: 洪西進
Shi-Jinn Horng
Committee: 吳金雄
Chin-Shyurng Fahn
Shi-Jinn Horng
Degree: 碩士
Department: 電資學院 - 資訊工程系
Department of Computer Science and Information Engineering
Thesis Publication Year: 2019
Graduation Academic Year: 107
Language: 中文
Pages: 78
Keywords (in Chinese): 人臉辨識人臉偵測深度學習
Keywords (in other languages): Face Reconition, Face detection, deep learning
Reference times: Clicks: 293Downloads: 0
School Collection Retrieve National Library Collection Retrieve Error Report

  智慧型手機的功能需要靠各式各樣的APP來擴充,本論文成功將深度學習的「SSD 物件偵測」、「MTCNN人臉偵測」、「arc face人臉辨識」演算法實作在Android手機環境,做出具有實用功能的人臉解鎖APP。並設計出計算速度較快且準度較高的SSD人臉偵測的演算法,也比較了SSD 及MTCNN演算法的優缺點,分析了人臉辨識流程中的各階段的耗時,以及分析人臉偵測準確度難以提升的原因。本論文之人臉辨識APP針對手機硬體效能量身訂做,在準確度與速度間取得平衡,製作出良好使用者體驗的APP。

A smart phone is a mobile phone that has a mobile operating system and can be expanded by installing application software, games, and the like. Its computing power and functions are superior to traditional functional phones.
The function of the smart phone is expanded by various APPs. This paper successfully implements the deep learning "SSD object detection", "MTCNN face detection" and "arc face face recognition" algorithm in Android mobile phone environment, make a face unlocking app with practical functions. The algorithm of SSD face detection with faster calculation speed and higher accuracy is designed. The advantages and disadvantages of SSD and MTCNN algorithms are also compared. The time consumption of each stage in the face recognition process is analyzed. And found out the reason why face detection accuracy is difficult to improve. The face recognition APP of this thesis is tailor-made for mobile phones, achieving a balance between accuracy and speed, and making an APP with a good user experience.

目錄 摘要------------------------------------------------ I Abstract-------------------------------------------- II 致謝------------------------------------------------ III 國立台灣科技大學學位論文創新聲明-------------------- IV 目錄------------------------------------------------ V 圖目錄---------------------------------------------- VII 表目錄---------------------------------------------- IX 第一章緒論------------------------------------------- 1 1.1研究動機與目的------------------------------------ 1 1.2相關研究-------------------------------------------2 第二章 系統架構與相關硬體規格-------------------------4 2.1系統架構-------------------------------------------4 2.2相關硬體規格---------------------------------------5 2.3使用環境與操作說明---------------------------------5 第三章 深度學習介紹-----------------------------------8 3.1深度學習原理---------------------------------------8 3.2卷積神經網路--------------------------------------10 3.3 池化層-------------------------------------------12 3.4深度學習人臉辨識原理------------------------------13 3.4.1深度學習分類器原理--------------------------13 3.4.2 深度學習人臉辨識原理-----------------------13 3.4.3 神經網路設計-------------------------------17 第四章 深度學習中的物件偵測介紹與相關研究探討--------20 4.1 目標檢測介紹-------------------------------------20 4.2 R-CNN--------------------------------------------21 4.3 fast R-CNN---------------------------------------23 4.4 Faster R-CNN-------------------------------------24 4.5 Yolo---------------------------------------------26 4.6 SSD----------------------------------------------30 4.7 MTCNN--------------------------------------------34 第五章 驗證標準介紹----------------------------------39 5.1二元分類器驗證方式--------------------------------39 5.2物件偵測驗證方式----------------------------------42 5.3驗證資料集介紹------------------------------------45 5.3.1 LFW資料集介紹------------------------------45 5.3.2 wider face資料集介紹-----------------------46 第六章 系統效能與實作---------------------------------46 6.1 SSD 演算法於Android之實現------------------------46 6.2 MTCNN演算法於Android之實現-----------------------53 6.3 人臉辨識模型訓練及Android實現---------------------57 6.4 Android UI 實作------------------------------------59 6.5 系統效能-------------------------------------------61 第七章 結論--------------------------------------------65 7.1研究成果-------------------------------------------66 7.2未來展望-------------------------------------------67

1. Alex Krizhevsky, Ilya Sutskever, Geoffrey E Hinton ,“ImageNet Classification with Deep Convolutional Neural Networks”, Advances in neural information processing systems, 1097-1105
2. Karen Simonyan, Andrew Zisserman,”Very Deep Convolutional Networks for Large-Scale Image Recognition”, ICLR 2015
3. Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich ,”Going Deeper with Convolutions”,CVPR 2015
4. Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun ,”Deep Residual Learning for Image Recognition”,CoRR 2015
5. Gao Huang, Zhuang Liu, Laurens van der Maaten, Kilian Q. Weinberger,” Densely Connected Convolutional Networks”, CVPR 2017
6. Forrest N. Iandola, Song Han, Matthew W. Moskewicz, Khalid Ashraf, William J. Dally, Kurt Keutzer,” SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size”,CVPR 2016
7. Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam,” MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications”,CVPR 2017
8. Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin, Jian Sun,” MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications”,CVPR 2017
9. François Chollet,” Xception: Deep Learning with Depthwise Separable Convolutions”,CVPR 2017
10. Elad Hoffer, Nir Ailon,” Deep metric learning using Triplet network”, International Workshop on Similarity-Based Pattern 2015
11. Yandong Wen, Kaipeng Zhang, Zhifeng Li , and Yu Qiao,” A Discriminative Feature Learning Approach for Deep Face Recognition”,ECCV 2016
12. Weiyang Liu, Yandong Wen, Zhiding Yu, Ming Li, Bhiksha Raj, Le Song,” SphereFace: Deep Hypersphere Embedding for Face Recognition”, CVPR, 2017
13. Hao Wang, Yitong Wang, Zheng Zhou, Xing Ji, Dihong Gong, Jingchao Zhou, Zhifeng Li, Wei Liu,” CosFace: Large Margin Cosine Loss for Deep Face Recognition”, CVPR 2018
14. Jiankang Deng, Jia Guo, Niannan Xue, Stefanos Zafeiriou,” ArcFace: Additive Angular Margin Loss for Deep Face Recognition”, arXiv preprint arXiv:1801.07698, 2018
15. Ross Girshick, Jeff Donahue, Trevor Darrell, Jitendra Malik,” Rich feature hierarchies for accurate object detection and semantic segmentation”,CVPR 2014
16. Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun,” Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition”,CVPR2014
17. Ross Girshick,” Fast R-CNN”, ICCV 2015
18. Shaoqing Ren, Kaiming He, Ross Girshick, Jian Sun,” Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks”,CVPR 2015
19. Joseph Redmon, Santosh Divvala, Ross Girshick, Ali Farhadi,” You Only Look Once: Unified, Real-Time Object Detection”,CVPR 2015
20. Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed, Cheng-Yang Fu, Alexander C. Berg,” SSD: Single Shot MultiBox Detector”, ECCV 2016
21. Kaipeng Zhang, Zhanpeng Zhang, Zhifeng Li, Yu Qiao,” Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks”,CVPR 2016
22. W. Tian, Z. Wang, H. Shen, W. Deng, B. Chen, X. Zhang ,” DFS - W. Tian, Z. Wang, H. Shen, W. Deng, B. Chen, X. Zhang Learning Better Features for Face Detection with Feature Fusion and Segmentation Supervision”, arXiv preprint arXiv:1811.08557, 2018.
23. S. Zhang, R. Zhu, X. Wang, H. Shi, T. Fu, S. Wang, T. Mei, Stan Z. Li,” Improved Selective Refinement Network for Face Detection”, arXiv preprint arXiv:1901.06651, 2019.
24. J. Li, Y. Wang, C. Wang, Y. Tai, J. Qian, J. Yang, C. Wang, J. Li, F. Huang ,”DSFD: Dual Shot Face Detector.”, arXiv preprint arXiv:1810.10220, 2018.

無法下載圖示 Full text public date 2024/08/02 (Intranet public)
Full text public date This full text is not authorized to be published. (Internet public)
Full text public date This full text is not authorized to be published. (National library)