基於孿生神經網路之乳牛身分識別

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：51

、訪客IP：3.21.93.108

姓名

戴宏林(Hong Ling Tai) 查詢紙本館藏

畢業系所

資訊工程學系在職專班

論文名稱

基於孿生神經網路之乳牛身分識別
(Siamese Neural Network Based Biometric Identification in Dairy Cows)

相關論文

★ 整合GRAFCET虛擬機器的智慧型控制器開發平台	★ 分散式工業電子看板網路系統設計與實作
★ 設計與實作一個基於雙攝影機視覺系統的雙點觸控螢幕	★ 智慧型機器人的嵌入式計算平台
★ 一個即時移動物偵測與追蹤的嵌入式系統	★ 一個固態硬碟的多處理器架構與分散式控制演算法
★ 基於立體視覺手勢辨識的人機互動系統	★ 整合仿生智慧行為控制的機器人系統晶片設計
★ 嵌入式無線影像感測網路的設計與實作	★ 以雙核心處理器為基礎之車牌辨識系統
★ 基於立體視覺的連續三維手勢辨識	★ 微型、超低功耗無線感測網路控制器設計與硬體實作
★ 串流影像之即時人臉偵測、追蹤與辨識─嵌入式系統設計	★ 一個快速立體視覺系統的嵌入式硬體設計
★ 即時連續影像接合系統設計與實作	★ 基於雙核心平台的嵌入式步態辨識系統

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2026-10-1以後開放)

摘要(中)

乳牛是種難以控制其行為的一種動物，不容易對其進行拍攝取樣，特別是牛臉這種局部特徵。因此本研究設計一套基於孿生神經網路的乳牛身分識別系統，能夠利用少量牛臉影像資料集，訓練出深度學習模型，達到有效的身分識別效果。利用Yolov4能快速偵測並分割影像中的牛臉，再將影像測試樣本與已訓練過的影像樣本輸入孿生神經網路(Siamese Neural Network)中的神經網路模型各別學習特徵，利用特徵向量計算出歐式距離(Euclidean Distance)，再經由sigmoid將歐式距離正規化後，得出兩個輸入影像之間的相似度，最後比較相似度以判定乳牛身分。實驗結果顯示，我們的系統中基於Resnet50的孿生神經網路在少樣訓練樣本數(每隻牛20張影像)得到95.5%辨識率，優於單獨使用Resnet50的TOP1辨識率70%。我們系統中的網路模型除了使用Resnet50也嘗試替換不同神經網路模型，用以比較辨識效果，實驗結果顯示替換為Resnet18的辨識率可以達到99.5%；替換為我們自建的CNN網路模型辨識率為75%；替換為VGG16辨識率為90.5%。這顯示我們所設計的系統在基於Resnet18的孿生神經網路能達到最佳的辨識效果，在少量訓練樣本中能夠佔有優勢。

摘要(英)

Dairy cows are an animal whose behavior is difficult to control. It is difficult to capture the ideal angle when shooting images on it, especially local features such as cow faces. In this research, a set of dairy cow biometric identity recognition system based on siamese neural network is designed, which can use a small amount of cow face image data set to train deep learning model to achieve effective identity recognition effect. In this research, Yolov4 can quickly detect and segment the dairy cow face in the image, and then input the image test sample and the trained image sample into the neural network model in the Siamese Neural Network to learn features, and use the feature vector to calculates the Euclidean Distance, and then normalize the Euclidean Distance by sigmoid to obtain the similarity between the two input images, and finally compares the similarity to determine the identity of the dairy cow. The experimental results show that Resnet50-based Siamese Neural Network in our system has a 95.5% recognition rate with a small number of training samples (20 images per cow), which is better than the 70% recognition rate of TOP1 using Resnet50 only. In addition to using Resnet50, we also tries to replace different neural network models in our system experiment to compare the recognition effect. The experimental results show that the recognition rate of replacing with Resnet18 can reach 99.5%, replacing with CNN network model has a recognition rate of 75%, replacing with VGG16 has a recognition rate of 90.5%. This shows that the dairy cow biometric identity recognition system we designed can achieve the best recognition effect in the Siamese Neural Network based on Resnet18 can achieve the best recognition effect, and it can have an advantage in a small number of training samples.

關鍵字(中)

★ 乳牛
★ 辨識

關鍵字(英)

★ Dairy Cows
★ Biometric Identification

論文目次

摘　要 I
Abstract II
謝誌 III
目錄 IV
圖目錄 VI
表目錄 VIII
第一章、緒論 1
1.1研究背景 1
1.2研究目的 3
1.3論文架構 4
第二章、技術回顧 5
2.1 YOLOv4 5
2.2 Siamese Neural Network 11
2.2.1 Siamese網路結構 12
2.2.2 Loss Function 14
2.3 One-Shot Learning 16
2.4 Resnet 17
第三章、孿生網路乳牛身分識別系統設計 20
3.1 MIAT系統設計方法論 20
3.2 乳牛身分識別系統架構 23
3.3 牛臉影像分割模組架構 26
3.4 孿生神經網路訓練模組架構 28
3.5 乳牛身分識別模組 30
第四章、實驗 32
4.1 實驗環境 32
4.2 牛臉影像分割實驗 35
4.2.1 LabelImg建立Label 35
4.2.2 調整YOLOv4參數 37
4.2.3 YOLOv4模型訓練實驗 39
4.2.4 乳牛臉部分割實驗 42
4.3 乳牛身分識別實驗 42
4.3.1 Resnet50乳牛身分識別實驗 44
4.3.2 孿生神經網路乳牛身分識別實驗 49
4.3.3 基於各神經網路模型的孿生神經網路統計結果 60
第五章、結論與未來發展 61
5.1 結論 61
5.2 未來展望 62
參考文獻 63
附錄一 66

參考文獻

[1] 李盼, 余祁暐, "全球智慧生乳產業趨勢研究及案例分析," 臺灣經濟研究月刊, vol. 42, no. 3, pp. 41-51, 2019.
[2] A. I. Awad, "From classical methods to animal biometrics: a review on cattle identification and tracking," Computers and Electronics in Agriculture, vol. 123, pp. 423-435, 2016.
[3] K. S. Schwartzkopf-Genswein, J. M. Stookey, and R. Welford, "Behavior of cattle during hot-iron and freeze branding and the effects on subsequent handling ease, " Journal of animal science, vol. 75, no. 8, pp. 2064-2072, 1997.
[4] J.-J. Hilpert, "Animal ear tag," ed: Google Patents, 2003.
[5] S. Stankovski, G. Ostojic, I. Senk, M. Rakic-Skokovic, S. Trivunovic, and D. Kucevic, "Dairy cow monitoring by RFID," Scientia Agricola, vol. 69, no. 1, pp. 75-80, 2012.
[6] Y. Lu, X. He, Y. Wen, and P. S. Wang, "A new cow identification system based on iris analysis and recognition," International journal of biometrics, vol. 6, no. 1, pp. 18-32, 2014.
[7] T. Gaber, A. Tharwat, A. E. Hassanien, and V. Snasel, "Biometric cattle identification approach based on Weber’s local descriptor and adaboost classifier," Computers and Electronics in Agriculture, vol. 122, pp. 55-66, 2016.
[8] S. Kumar, S. Tiwari, and S. K. Singh, "Face recognition for cattle," in 2015 Third International Conference on Image Information Processing (ICIIP), 2015, pp. 65-72: IEEE.
[9] 羅文圻, "牧場乳牛偵測與身分識別", 國立中央大學資訊工程所碩士論文, 108.
[10] Y. Qiao, D. Su, H. Kong, S. Sukkarieh, S. Lomax, and C. Clark, "BILSTM-based individual cattle identification for automated precision livestock farming," in 2020 IEEE 16th International Conference on Automation Science and Engineering (CASE), 2020, pp. 967-972.
[11] A. A. R. Putra and S. Setumin, "The performance of siamese neural network for face recognition using different activation functions," 2021 International Conference of Technology, Science and Administration (ICTSA), 2021, pp. 1-5.
[12] X. Dong, J. Shen, D. Wu, K. Guo, X. Jin and F. Porikli, "Quadruplet network with one-shot learning for fast visual object tracking," in IEEE Transactions on Image Processing, vol. 28, no. 7, pp. 3516-3527, July 2019.
[13] R. W. Arief, I. Nurtanio and F. A. Samman, "Traffic signs detection and recognition system using the YOLOv4 algorithm," 2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS), 2021.
[14] S. F. Dodge and L. J. Karam, "Understanding how image quality affects deep neural networks," Eighth International Conference on Quality of Multimedia Experience (QoMEX), 2016.
[15] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: unified, real-time object detection," in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 779–788, 2016.
[16] J. Redmon and A. Farhadi, "YOLO9000: better, faster, stronger," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 6517-6525
[17] J. Zhang, M. Huang, X. Jin, and X. Li, "A real-time Chinese traffic sign detection algorithm based on modified YOLOv2," Algorithms, vol. 10, no. 4, p. 127, Nov. 2017
[18] J. Redmon and A. Farhadi, "YOLOv3: an incremental improvement," arXiv preprint arXiv:1804.02767, 2018.
[19] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, "YOLOv4: optimal speed and accuracy of object detection," arXiv preprint arXiv:2004.10934, 2020.
[20] G. Yang et al., "Face mask recognition system with YOLOv5 based on image recognition," 2020 IEEE 6th International Conference on Computer and Communications (ICCC), 2020, pp. 1398-1404
[21] A. Bochkovskiy, C. Wang, H. Mark Liao, "YOLOv4: optimal speed and accuracy of object detection," arXiv preprint arXiv:.10934, 2004
[22] J. Deng, W. Dong, R. Socher, L. Li, Kai Li and Li Fei-Fei, "Imagenet: a large-scale hierarchical image database," 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 248-255
[23] C.-Y. Wang, H.-Y. M. Liao, I.-H. Yeh, Y.-H. Wu, P.-Y.Chen, and J.-W. Hsieh, "Cspnet: a new backbone that can enhance learning capability of CNN," arXiv preprint arXiv:.11929, 2019
[24] A. Mikołajczyk and M. Grochowski, "Data augmentation for improving deep learning in image classification problem," 2018 International Interdisciplinary PhD Workshop (IIPhDW), 2018, pp. 117-122.
[25] P. Rama Devi, R. Yashashvini, G. Navyadhara and M. Ruchitha, "Similar face detection for Indian faces using siamese neural networks," 2021 2nd International Conference for Emerging Technology (INCET), 2021, pp. 1-5
[26] M. Muzammul, Xi Li, "A survey on deep domain adaptation and tiny object detection challenges, techniques and datasets," arXiv:.07927, 2021.
[27] S. Kousha, M. A. Brubaker, "Zero-shot learning with class description regularization," arXiv preprint arXiv:.16108, 2021.
[28] H. Mukhtar, M. Z. Khan, M. U. G. Khan and H. Younis, "Wheat disease recognition through one-shot learning using fields images," 2021 International Conference on Artificial Intelligence (ICAI), 2021, pp. 229-233
[29] J. Yan, Y. Zhu, Z. Liang, Y. Zhu, K. Wu and Z. Lin, "Tire pattern classification based on few-shot learning," in 2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), 2021, pp. 1-7
[30] K. He, X. Zhang, S. Ren, J. Sun, "Deep residual learning for image recognition," arXiv:.03385, 2015
[31] C.-H. Chen, M.-Y. Lin, and X.-C. Guo, "High-level modeling and synthesis of smart sensor networks for industrial internet of things," Computers and Electrical Engineering, vol. 61, pp. 48-66, 2017.

指導教授

陳慶瀚(Pierre Wang)

審核日期

2021-10-15

推文