指紋作為最常用的生物特徵之一，被廣泛應用在各個領域上。指紋識別也已經被研究了幾十年，然而即使是最先進的識別方法的表現，也會因為指紋質量受到很大的影響。如果能找出這些低質量有缺陷的區域，我們可以針對這些區域做處理，來提高識別率。
在本篇論文中，我們提出了一個基於全卷積網路的指紋缺陷區塊偵測系統。該系統能夠在指紋圖像中偵測出背景以及過乾和過濕的地方，并給出像素級別的分割。這個系統主要是利用深度神經網路在分類問題上有效的特徵學習能力和極高的準確率，結合全卷積網路能做像素級別分割的特點，來實現對指紋缺陷區域的偵測。為了解決缺少用於給全卷積網路訓練的分割資料問題，我們提出了一種指紋合成的方法，能生成帶有缺陷區域的指紋圖像，并能同時得到這些缺陷區域在哪些像素上。此外我們還將我們的網路在指紋區塊上作為一個分類器來做預訓練，學習不同類別指紋的特征，來在分割問題上獲得更好的表現。在我們的實驗中，結果證明該系統能與現有的指紋識別方法結合，提升識別率。

Fingerprint, one of the most popular and acceptable biometric features, is widely used in all kinds of fields. Fingerprint recognition technology has been researched for decades. However, even the state-of-the-art recognition approach, its performance depends greatly on the fingerprint quality. If we can detect the defective or low-quality regions in fingerprint images, we can repair them or downgrade these regions to improve the recognition performance.
In this thesis, we propose a system for defective region detection from fingerprint images based on fully convolutional network (FCN). This system can detect defective regions which are too dry, too wet, or belongs to the background, in the pixel-wise level. The proposed region detection algorithm takes advantage of the high classification accuracy and great learning power of the deep neural network. Combined with the feature of FCN for pixel-wise prediction, we can accomplish the task of defective region detection. With pre-training on fingerprint patches and the defective fingerprint synthesis approach, we develop a FCN model with satisfactory performance on the fingerprint segmentation task. At last, our experiments show that the proposed FCN fingerprint segmentation model can be used to improve the fingerprint recognition accuracy by integrating it with the existing fingerprint recognition system.

Chapter 1. Introduction 1
1.1 Motivation 1
1.2 Problem Description 3
1.3 Main Contribution 4
1.4 Thesis Organization 6
Chapter 2. Related Works 7
2.1 Fingerprint Quality Assessment 7
2.2 Deep Network and Object Detection 8
2.3 Inspiration from Previous Works 9
Chapter 3. Proposed Method 10
3.1 Data Preparation 10
3.1.1 Generate Fingerprint Data for CNN Classifier 11
3.1.2 Generate Fingerprint Data for FCN 14
3.2 Training CNN Classifier 17
3.2.1 FPNet Architecture 17
3.2.2 Training Details 19
3.3 Training FCN for Segmentation 20
3.3.1 From FPNet to Dense FCN 20
3.3.2 Training Details 21
Chapter 4. Experimental Results 23
4.1 Data Preparation 23
4.2 Defective Region Detection 24
4.2.1 Prediction on Synthetic Data 24
4.2.2 Prediction on Real Data 26
4.3 FCN8s vs. FCN32s 26
4.4 Experiments with Fingerprint Recognition 28
4.4.1 Review 28
4.4.2 Experiment Setting 29
4.4.3 Minutiae Based Recognition Algorithm 30
4.4.4 Improvement Strategy 31
4.4.5 Results and Discussion 37
Chapter 5. Conclusions 39
Reference 40

[1] Cappelli, Raffaele, Matteo Ferrara, and Davide Maltoni. "Minutia cylinder-code: A new representation and matching technique for fingerprint recognition." IEEE Transactions on Pattern Analysis and Machine Intelligence 32.12: 2128-2141, 2010.
[2] Jea, Tsai-Yang, and Venu Govindaraju. "A minutia-based partial fingerprint recognition system." Pattern Recognition 38.10: 1672-1684, 2005.
[3] Jiang, Xudong, and Wei-Yun Yau. "Fingerprint minutiae matching based on the local and global structures." Pattern Recognition, 2000.
[4] Fronthaler, Hartwig, et al. "Fingerprint image-quality estimation and its application to multialgorithm verification." IEEE Transactions on Information Forensics and Security 3.2: 331-338, 2008.
[5] Eskicioglu, Ahmet M., and Paul S. Fisher. "Image quality measures and their performance." IEEE Transactions on Communications 43.12: 2959-2965, 1995.
[6] Shen, LinLin, Alex Kot, and Wai Mun Koo. "Quality measures of fingerprint images." AVBPA. 2001.
[7] Lee, Bongku, Jihyun Moon, and Hakil Kim. "A novel measure of fingerprint image quality using the Fourier spectrum." Proc. SPIE. Vol. 5779, 2005.
[8] Tabassi, Elham, and Charles L. Wilson. "A novel approach to fingerprint image quality." IEEE International Conference on Image Processing, 2005.
[9] Bazen, Asker M., and Sabih H. Gerez. "Segmentation of fingerprint images." ProRISC 2001 Workshop on Circuits, Systems and Signal Processing, 2001.
[10] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification with deep convolutional neural networks." Advances in Neural Information Processing Systems, 2012.
[11] Girshick, Ross, et al. "Region-based convolutional networks for accurate object detection and segmentation." IEEE Transactions on Pattern Analysis and Machine Intelligence 38.1: 142-158, 2016.
[12] Girshick, Ross. "Fast r-cnn." Proceedings of the IEEE International Conference on Computer Vision, 2015.
[13] Ren, Shaoqing, et al. "Faster R-CNN: Towards real-time object detection with region proposal networks." Advances in Neural Information Processing Systems, 2015.
[14] Redmon, Joseph, et al. "You only look once: Unified, real-time object detection." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016.
[15] Long, Jonathan, Evan Shelhamer, and Trevor Darrell. "Fully convolutional networks for semantic segmentation." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015.
[16] Maio, Dario, et al. "FVC2000: Fingerprint verification competition." IEEE Transactions on Pattern Analysis and Machine Intelligence 24.3: 402-412, 2002.
[17] Maio, Dario, et al. "FVC2002: Second fingerprint verification competition." International Conference on Pattern recognition, 2002.
[18] Maio, Dario, et al. "FVC2004: Third fingerprint verification competition." Biometric Authentication: 31-35, 2004.
[19] Cappelli, Raffaele, et al. "Fingerprint verification competition 2006." Biometric Technology Today 15.7: 7-9, 2007.
[20] Magalhaes, F., H. P. Oliveira, and A. Campilho. "Spd 2010-fingerprint singular points detection competition database." 2010.
[21] Cappelli, Raffaele, D. Maio, and D. Maltoni. "SFinGe: an approach to synthetic fingerprint generation." International Workshop on Biometric Technologies, 2004.
[22] Perlin, Ken. "An image synthesizer." ACM Siggraph Computer Graphics 19.3: 287-296, 1985.
[23] Cappelli, Raffaele, Dario Maio, and Davide Maltoni. "An improved noise model for the generation of synthetic fingerprints." Control, Automation, Robotics and Vision Conference, 2004.
[24] Simonyan, Karen, and Andrew Zisserman. "Very deep convolutional networks for large-scale image recognition." arXiv preprint arXiv:1409.1556, 2014.
[25] Yao, Zhigang, et al. "Literature review of fingerprint quality assessment and its evaluation." IET Biometrics 5.3: 243-251, 2016.
[26] Garris, Michael D., et al. "User's guide to NIST fingerprint image software (NFIS)." NIST Interagency/Internal Report (NISTIR)-6813, 2001.
[27] Wong, Wei Jing, et al. "Enhanced multi-line code for minutiae-based fingerprint template protection." Pattern Recognition Letters 34.11: 1221-1229, 2013.
[28] Everingham, Mark, et al. "The pascal visual object classes challenge: A retrospective." International Journal of Computer Vision 111.1: 98-136, 2015.