跳到主要內容

臺灣博碩士論文加值系統

(44.211.239.1) 您好!臺灣時間:2023/01/31 06:06
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:陳冠珽
研究生(外文):Kuan-Ting Chen
論文名稱:基於深度學習之鈔票辨識系統
論文名稱(外文):A banknotes recognition system based on Yolact
指導教授:王文俊王文俊引用關係
指導教授(外文):Wen-June Wang
學位類別:碩士
校院名稱:國立中央大學
系所名稱:電機工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:中文
論文頁數:74
中文關鍵詞:深度學習實例分割鈔票辨識影像特徵處理
相關次數:
  • 被引用被引用:2
  • 點閱點閱:342
  • 評分評分:
  • 下載下載:79
  • 收藏至我的研究室書目清單書目收藏:0
本論文目標是建立一個方便視障人士使用的新台幣鈔票辨識系統。使用深度學習的架構結合影像處理的演算法,本系統使用深度學習的架構結合影像處理的演算法達成目標。當使用者以手機拍攝多張面額不限之鈔票且無論重疊與否,都可以準確辨識鈔票的面額、數量和相對位置,並用語音告知使用者。如此視障人士就可以用手機輕易辨識鈔票付錢購物或是檢查店員找錢金額的正確性了。
此技術利用深度學習Yolact (You Only Look At CoefficienTs)的架構,辨識出每種不同幣值鈔票的國字特徵、數字特徵、反光數字特徵,再利用MBR (minimum bounding rectangle) 及 ORB (Oriented FAST and Rotated BRIEF)兩種影像處理的演算法對各個特徵進行處理。MBR用來找到特徵的方向,特徵框的大小及相對位置。ORB用來解決一些特徵破碎的情況,並做特徵定位。利用上兩個方法所得的結果去做分析,求出鈔票的面額及數量。最後把所有的特徵,去做排列組合,並利用線性回歸找出所有組合的回歸線,找到最短距離的最佳解。利用最佳解將鈔票的相對位置計算出來。
本系統的伺服器建立在本地端電腦上,使用者端則是設計在Android手機APP內,只要利用智慧型手機和原有的內建鏡頭,就可以快速的判別鈔票的面額、數量和相對位置。我們以智慧型手機Sony Xperia Z4來測試,在單一鈔票和多張鈔票等各種情況下,平均有85%的準確率。
The goal of this thesis is to establish a recognizing system of NTD (New Taiwan Dollars) banknotes for helping the visually impaired to shop. We combine deep learning neural networks and traditional image processing algorithms to achieve the goal. We designed an APP in which when the mobile phone takes the photos of several banknotes whether they are overlapped or not. The number, denominations and relative positions of the banknotes can be accurately recognized and transmitted by the blind’s smart phone. Therefore, using this APP, the blind can pay the money to shop and check the correctness of the return change from the clerk.
This technology uses the structure of the deep learning neural network Yolact (You Only Look At CoefficienTs) to identify the Chinese characters, digital features, and reflective digital features for each kind of currency banknotes. Then two traditional image processing methods MBR (minimum bounding rectangle) and ORB (Oriented FAST and Rotated BRIEF) algorithms to deal with these features. MBR is used to find the direction of features, the size and relative position of feature boxes. The ORB is used to solve the situation of some broken features, and doing feature positioning. Based on the above two processes, we can find the denomination and quantity of banknotes. Finally, we find the optimal regression lines to fit all features so that the relative positions of the banknotes are obtained. Therefore, the denominations, the relative position and the quantity of banknotes are recognized correctly.
The server of this system is built on the local computer, and the user terminal is designed in the Android mobile phone APP. As long as we use the smartphone and the original built-in camera. The denomination, quantity and relative position of the banknote can be quickly determined. The correct recognition accuracy is about 85% in our experiment with Sony Xperia Z4 smartphone in various situations such as single banknotes and multiple banknotes.
摘要 1
Abstract 2
致謝 4
圖目錄 7
表目錄 10
第一章 緒論 1
1.1. 研究動機與背景 1
1.2. 文獻回顧 2
1.3. 研究目標 5
1.4. 論文架構 5
第二章 系統架構與軟硬體介紹 6
2.1. 系統架構 6
2.2. 硬體介紹 7
2.2.1. 桌上型電腦 7
2.2.2. 顯示卡 7
2.2.3. 行動裝置 8
2.3. 軟體介紹 9
2.3.1. VGG Image Annotator (VIA) 9
2.3.2. Pytorch 9
第三章 系統網路架構及訓練 11
3.1. 深度網路特徵辨識 11
3.1.1. 網路介紹 11
3.1.2. 網路架構 12
3.2. 訓練過程 18
3.3. 資料收集及擴增 22
第四章 鈔票特徵處理與分析 24
4.1. 濾掉偏差特徵 25
4.2. 分析各特徵參數 26
4.2.1. MBR最小矩形框 (minimum bounding rectangle) 28
4.2.2. ORB向量的定義 29
4.2.3. CNN來定義正向或反向 33
4.3. 整合各個鈔票的特徵 36
4.4. 計算鈔票相對位置 41
第五章 手機應用程式設計 46
5.1. Android Studio系統介紹 46
5.2. APP操作過程 49
第六章 實驗結果 51
6.1. 單張鈔票辨識結果 51
6.2. 多張不重疊桌上辨識結果 52
6.3. 多張重疊桌上 53
6.4. 多張重疊手上 54
6.5. 辨識結果總結 55
第七章 結論與未來展望 56
7.1. 結論 56
7.2. 未來展望 57
參考文獻 58
[1] R. Kumar, T. Maktabi, and S. O’Brien, "2018 Findings from the Diary of Consumer Payment Choice," Cash product office federal reserve bank, San Francisco, 2019.
[2] K. Yan, and R. Sukthankar, "PCA-SIFT: A more distinctive representation for local image descriptors," in 2004 IEEE computer society conference on computer vision and pattern recognition, 2004.
[3] H. Bay, A. Ess, T. Tuytelaars, L. V. Gool, "Speeded-up robust features (SURF)," Computer vision and image understanding, vol. 110, no. 3, pp. 346-359, Jun, 2008.
[4] E. Rublee, V. Rabaud, K. Konolige, G. Bradski, "ORB: An efficient alternative to SIFT or SURF," in 2011 International conference on computer vision, 2011, pp. 2564-2571.
[5] S. M. Pizer, E. P. Amburn, J. D. Austin, R. Cromartie, A. Geselowitz, T. Greer, B. H. Romeny, J. B. Zimmerman, K.Zuidervled, "Adaptive histogram equalization and its variations," Computer vision graphics and image processing, vol. 39, no. 3, pp. 355-368, Sep, 1987
[6] A. M. Reza, "Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement," Journal of VLSI signal processing systems for signal image and video technology, vol. 38, pp. 35-44, Nov, 2004.
[7] M. Elad, "On the origin of the bilateral filter and ways to improve it," IEEE transactions on image processing, vol. 11, no. 10, pp. 1141-1151, Oct, 2002.
[8] R. A. Hummel, B. Kimia, and S. W. Zucker, "Deblurring gaussian blur," Computer vision graphics and image processing, vol. 38, no. 1, pp. 66-80, Apr, 1987.
[9] L. O. Chua, and T. Roska, "The CNN paradigm," IEEE transactions on circuits and systems I: fundamental theory and applications, vol. 40, no. 3, pp. 147-156, Mar, 1993.
[10] R. P. Bernardino, and P. H. S. Torr, "Recurrent instance segmentation," in European conference on computer vision, 2016, pp. 312-329.
[11] C. M. Costa, G. Veiga, and A. Sousa, "Recognition of banknotes in multiple perspectives using selective feature matching and shape analysis," in 2016 International conference on autonomous robot systems and competitions, 2016.
[12] F. M. Hasanuzzaman, Y. X. Dong, and T. Y. Li, "Robust and effective component-based banknote recognition for the blind," IEEE transactions on systems, man, and cybernetics, part C (Applications and Reviews), vol. 42, no. 6, pp. 1021-1030, Jan, 2012.
[13] S. Baiqing, and L. Jilu, "Recognition for the banknotes grade based on CPN, " in 2008 international conference on computer science and software engineering, vol. 1, 2008.
[14] S. Baiqing, and L. Jilu, "The recognition of new and old banknotes based on SVM," in 2008 second international symposium on intelligent information technology application, vol. 2, 2008.
[15] G. Felipe, J. C. Rodríguez, "Smartphone recognition of the US banknotes' denomination, for visually impaired people," in 2010 IEEE Andsecond, Sep, 2010.
[16] 曹丹華, 劉斌昺, and 吳裕斌, "投影特徵匹配的快速鈔幣面值識別算法," 光電工程, vol. 31, no. 1, pp. 59-61, 2004.
[17] 李文宏, "基于支持向量机的人民币纸币序列号识别方法," 信息與控制, vol. 39, no. 4, pp. 462-465, 2010.
[18] A. Dutta, A. Gupta and A. Zisserman, "VGG Image Annotator (VIA)," Available: http://www.robots.ox.ac.uk/~vgg/software/via/, 2019.
[19] G. S. Hu, Introduction to digital signal processing, Tsinghua University Press, Beijing, 2005.
[20] 江雅雯, "植基於色彩及紋理分析之鈔票偵測系統," 2006, pp. 1-65.
[21] AI研習社, "2019 年機器學習框架之爭:PyTorch 和 TensorFlow 誰更有勝算?" Available: www.chainnews.com/zh-hant/articles/819230104830.htm, 2019.
[22] D. Bolya, C. Zhou, F. Xiao, Y. J. Lee. "YOLACT: real-time instance segmentation," in IEEE international conference on computer vision. 2019, pp. 9157-9166.
[23] K. He, G. Gkioxari, P. Dollár, "Mask r-cnn," in IEEE international conference on computer vision, 2017, pp. 2961-2969.
[24] T. Y. Lin, P. Dollar, R. Girshick, K. He, B. Hariharan, S. Belongie, "Feature pyramid networks for object detection," in IEEE conference on computer vision and pattern recognition, 2017, pp. 2117-2125.
[25] H. Kaiming, Z. Xiangyu, R. Shaoqing, S. Jian, "Deep residual learning for image recognition," in IEEE conference on computer vision and pattern recognition, 2016, pp. 770-778 pp. 770-778.
[26] X. Sean, "YOLACT簡介 — Real-time Instance Segmentation," Available :https://medium.com/@xiaosean5408/yolact%E7%B0%A1%E4%BB%8B-real-time-instance-segmentation-293d4ea2aac7, 2019
[27] R. Ethan, E. Rublee, V. Rabaud, K. Konolige, G. Bradski, "ORB: An efficient alternative to SIFT or SURF," in 2011 IEEE conference on computer vision, IEEE, 2011.
[28] 天晴, "python openCV掩膜的通俗理解," Available: https://zhuanlan.zhihu.com/p/36656952, 2019.
[29] 碗豆代理, "python: openCV Arithmetic Operations on Images," Available: https://www.wandouip.com/t5i86548/, 2019.
[30] "Finding minimum-area-rectangle for given points?," Available: https://gis.stackexchange.com/questions/22895/finding-minimum-area-rectangle-for-given-points, 2011.
[31] 程鹏飞, 闫浩文, and 韩振辉, "一个求解多边形最小面积外接矩形的算法," 工程圖學學報, vol. 29, no. 1, pp. 122-126, 2008.
[32] K. Ebrahim, P. Siva, and S. Mohamed, "Image matching using SIFT, SURF, BRIEF and ORB: performance comparison for distorted images," arXiv preprint arXiv:1710.02726, 2017.
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊