臺灣博碩士論文加值系統

狗和貓經常出現在日常活動中，尤其是在校園地區。 可以通過圖像分類器解決識別被發現的狗或貓的品種的問題。 本文開發了一種用於貓狗品種分類器的應用程序。 基於VGG19的CDC模型是分類器模型，該模型是Keras應用程序中經過微調的模型。 選擇Raspberry Pi作為移動平台的主要計算系統，Raspberry Pi攝像機可以獲取實時視頻圖像作為分類器的輸入。 為了滿足深度學習過程的計算需求，英特爾Movidius神經計算棒已連接到Raspberry Pi。 級聯分類器用作視頻預處理，以減少處理時間並找到對象。 最後，Raspberry Pi控制電機跟隨框架中的狗或貓。

Dogs and cats are commonly seen in daily activity, especially in the campus area. The problem to identify the breed of the dog or cat can be solved with an image classifier. In this thesis, an application for dogs and cats breed classifier is developed. The VGG19-based CDC model, which is a fine-tuned model from the Keras application, was used as the classifier model. The Raspberry Pi was selected as the main computing system of the mobile platform and the Raspberry Pi camera could obtain the real-time video images as the input of the classifier. In order to meet the computing needs of deep learning process, Intel Movidius Neural Compute Stick was connected to the Raspberry Pi. The cascade classifier was used as the video pre-processing, to reduce the processing time and to locate the object. Finally, the Raspberry Pi control the motor to follow the dog or cat in the frame.

Chinese Abstract ....................................................................................... i
English Abstract ........................................................................................ ii
Acknowledgements ................................................................................... iii
Table of Contents ...................................................................................... iv
List of Tables ......................................................................................... vi
List of Figures ......................................................................................... vii
Chapter 1 Introduction ............................................................................. 1
1.1 Motivation ............................................................................... 1
1.2 Background ............................................................................. 2
1.3 Objective ................................................................................. 3
1.4 Thesis Organization ................................................................ 3
Chapter 2 Training of the CDC Models .................................................. 4
2.1 VGG-based CDC Model ......................................................... 4
2.2 ResNet-based CDC Model ...................................................... 8
2.3 Inception-based CDC Model .................................................. 10
2.4 Xception-based CDC Model ................................................... 12
Chapter 3 Testing of the CDC Models .................................................... 15
3.1 VGG-based CDC Model ......................................................... 15
3.2 ResNet-based CDC Model ...................................................... 18
3.3 Inception-based CDC Model .................................................. 20
3.4 Xception-based CDC Model ................................................... 21
Chapter 4 Hardware of the Dogs and Cats Breed Classifier ................... 23
4.1 Building Block of the Dogs and Cats Breed Classifier .......... 23
4.1.1 Cascade Classifier ................................................................... 23
4.1.2 Configuration of the Dogs and Cats Breed Classifier ............ 25
4.2 Dogs and Cats Breed Classifier Flow Chart ........................... 29
Chapter 5 Dogs and Cats Breed Classifier Testing ................................. 31
Chapter 6 Conclusion .............................................................................. 33
Reference ................................................................................................. 34
Extended Abstract ..................................................................................... 38

N. Tajbakhsh, J.Y. Shin, S.R. Gurudu, R.T. Hurst, C.B. Kendall et al., “Convolutional neural networks for medical image analysis: full training or fine tuning,” IEEE Transactions on Medical Imaging, vol. 35, no. 5, pp. 1299-1312, May 2016.
Mahardi, I.H. Wang, K.C. Lee and S.L. Chang, “Images classification of dogs and cats using fine-tuned VGG models,” in 2020 IEEE Eurasia Conference on IOT, Communication and Engineering (ECICE), pp. 230-233, Yunlin, Taiwan, 2020.
O.M. Parkhi, A. Vedaldi, A. Zisserman and C.V. Jawahar, “Cats and dogs,” in IEEE Conference on Computer Vision and Pattern Recognition, Providence, pp. 3498-3505, RI, 2012.
S. Panigrahi, A. Nanda and T. Swarnkar, “Deep learning approach for image classification,” in 2018 2nd International Conference on Data Science and Business Analytics (ICDSBA), pp. 511-516, Changsha, 2018.
T. Jajodia and P. Garg, “Image classification - cat and dog images,” International Research Journal of Engineering and Technology, vol. 6, issue 12, pp. 570-572, 2019.
X. Liu, M. Chi, Y. Zhang and Y. Qin, “Classifying High Resolution Remote Sensing Images by Fine-Tuned VGG Deep Networks,” IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, pp. 7137-7140, Valencia, Spain, 2018.
T. Kaur and T.K. Gandhi, “Automated Brain Image Classification Based on VGG-16 and Transfer Learning,” 2019 International Conference on Information Technology (ICIT), pp. 94-98, Bhubaneswar, India, 2019.
A.S.B. Reddy and D.S. Juliet, “Transfer learning with ResNet-50 for malaria cell-image classification,” 2019 International Conference on Communication and Signal Processing (ICCSP), pp. 0945-0949, Chennai, India, 2019.
W.W. Lo, X. Yang and Y. Wang, “An Xception convolutional neural network for malware classification with transfer learning,” 2019 10th IFIP International Conference on New Technologies, Mobility and Security (NTMS), pp. 1-5, Canary Islands, Spain, 2019.
G. Deng, Y. Zhao, L. Zhang, Z. Li, Y. Liu et al., “Image classification and detection of cigarette combustion cone based on Inception Resnet v2,” 2020 5th International Conference on Computer and Communication Systems (ICCCS), pp. 395-399, Shanghai, China, 2020.
V.G. Buddhavarapu and A.A. Jothi J, “An experimental study on classification of thyroid histopathology images using transfer learning,” Pattern Recognition Letters, vol. 140, pp. 1-9, 2020.
A. Karine, T. Napoléon, J.Y. Mulot and Y. Auffret, “Video Seals Recognition using Transfer Learning of Convolutional Neural Network,” 2020 Tenth International Conference on Image Processing Theory, Tools and Applications (IPTA), pp. 1-4, Paris, France, 2020
Y. Zhao, J. Gu, C. Liu, S. Han, Y. Gao and Q. Hu, “License Plate Location Based on Haar-Like Cascade Classifiers and Edges,” 2010 Second WRI Global Congress on Intelligent Systems, pp. 102-105, Wuhan, China, 2010.
X. Zhang, H. Song and H. Cui, “Road pedestrian detection based on a cascade of feature classifiers,” 2014 International Conference on Audio, Language and Image Processing, pp. 948-951, Shanghai, China, 2014.
L. Cuimei, Q. Zhiliang, J. Nan and W. Jianhua, “Human face detection algorithm via Haar cascade classifier combined with three additional classifiers,” 2017 13th IEEE International Conference on Electronic Measurement & Instruments (ICEMI), pp. 483-487, Yangzhou, China, 2017.
K. Janard and W. Marurngsith, “Accelerating real-time face detection on a raspberry pi telepresence robot,” Fifth International Conference on the Innovative Computing Technology (INTECH 2015), pp. 136-141, Galcia, Spain, 2015.
J. Jommuangbut and K. Sritrakulchai, “Development of the Human Following Robot Control System Using HD Webcam,” 2018 International Electrical Engineering Congress (iEECON), pp. 1-4, Krabi, Thailand, 2018.
F. Jafar, S.F. Fatima, H.R. Mushtaq, S. Khan, A. Rasheed and M. Sadaf, “Eye Controlled Wheelchair Using Transfer Learning,” 2019 International Symposium on Recent Advances in Electrical Engineering (RAEE), pp. 1-5, Islamabad, Pakistan, 2019.
K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” 3rd International Conference on Learning Representations, (ICLR), San Diego, USA, May 2015.
"Dreamtime Stock Photos,” n.d. [Online] Available: https://www. dreamstime.com/photos-images/dreamtime.html
K. He, X. Zhang, S. Ren and J. Sun, “Identity mappings in deep residual networks,” in European conference on computer vision, Springer, Cham, pp. 630-645, 2016.
I.H. Wang, Mahardi, K.C. Lee and S.L. Chang, “Predicting the Breed of Dogs and Cats with Fine-tuned Keras Applications,” submitted to Intelligent Automation & Soft Computing, November 2020.
C. Szegedy, S. Ioffe, V. Vanhoucke and A. Alemi, “Inception-v4, inception-resnet and the impact of residual connections on learning,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 31, no. 1, 2017.
F. Chollet, “Xception: deep learning with depthwise separable convolutions,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, pp. 1800-1807, 2017.
P. Viola and M.J. Jones, “Robust real-time face detection,” International journal of computer vision, vol. 57, no. 2, pp. 137-154, 2004.
S. Soo, “Object detection using Haar-cascade Classifier”. Institute of Computer Science, University of Tartu, vol. 2, no. 3, pp. 1-12, 2014.