養殖漁業在全球漁獲供給中扮演著相當重要的角色，而在全球石斑魚產量上，台灣因為掌握了重要的石斑魚養殖技術，是主要供給源之一。然而各式樣的內外因素，導致台灣養殖漁業發展低落。隨著影像辨識的導入，將有機會降低養殖漁業中，人力需求以及災害風險的問題。對於影像辨識在養殖漁業中的應用，物件偵測可被視為其基礎之一。因此，此研究旨在開發一個水下石斑魚偵測的模型，此模型將奠基於目前已公開，且基於深度學習之物件偵測模型。而為了克服影像資料不足的問題，此研究使用從線上資料集所取得之水下石斑魚影像作為訓練資料，並將從養殖漁場所收集到的少量水下石斑魚影像資料，用作模型的訓練與測試資料。此外，此研究使用不同訓練資料集和偵測模型進行實驗，以找出最佳組合參數。最後透過RetinaNet偵測模型，並搭配加躁處理後的線上水下石斑魚影像、少部分養殖漁場的水下石斑魚影像，取得92.40%的Average Precision。

Aquaculture has played an important role in global fish supply, while Taiwan has mastered the important grouper breeding techniques. However, various factors lead to a low industrial level of aquaculture in Taiwan. With the introduction of image recognition, manpower requirement and disaster risk in the fish farming fields are expected to decrease. For object detection can be a foundation of image recognition application in aquaculture, this research aimed at developing an underwater grouper detection model. The model was based on a currently available deep learning-based object detection model. Underwater grouper images from public domain datasets were used for training data to overcome the issue of insufficient image data, while a few images collected from fish farming fields were used for training and testing. Experiments with different training datasets and detection models were conducted to find out the best parameters combination. Ultimately, this research reached 92.40% of Average Precision by using RetinaNet model with a training dataset of noise-added online underwater grouper images and a few underwater grouper images from fish farming fields.

摘要----------I
Abstract----------II
致謝----------III
Contents----------IV
List of Tables----------VI
List of Figures----------VII
Chapter 1: Introduction----------1
1.1 Background----------1
1.2 Motivation and Objectives----------4
1.3 Research Method----------6
1.4 Organization of Thesis----------8
Chapter 2: Literature Review----------9
2.1 Object Detection Method----------9
2.2 Object Detection Application----------17
2.2.1 Terrestrial Animal Detection----------17
2.2.2 Aquatic Animal Detection----------18
2.3 Research Comparison----------22
Chapter 3: Experimental Materials and Methods----------28
3.1 Experimental Materials----------28
3.1.1 Dataset Investigation----------28
3.1.2 Database Construction----------32
3.1.3 Image Labeling----------33
3.2 Experimental Methods----------35
Chapter 4: Experiment----------40
4.1 Response Variable----------40
4.2 Factors----------44
4.2.1 Training Dataset----------45
4.2.2 Image Preprocessing----------45
4.2.3 Training Model----------46
4.2.3.1 SSD----------46
4.2.3.2 RetinaNet----------47
4.2.3.3 YOLO v3----------49
Chapter 5: Results----------51
5.1 System Configuration----------52
5.2 Experimental Result----------52
5.3 Advanced Analysis----------58
Chapter 6: Conclusion----------64
Reference----------66
Appendix----------71
Appendix A: Detection Results of Average Precision (Precision, Recall) on Testing Dataset in the Experiment, Unit: Percentage (%)----------71
Appendix B: Precision-Recall Curve of Each Experiment in SSD----------72
Appendix C: Precision-Recall Curve of Each Experiment in RetinaNet----------73
Appendix D: Precision-Recall Curve of Each Experiment in YOLOv3----------74
Appendix E: Detection Results of Precision, Recall and Average Precision on Testing Dataset in Advanced Analysis----------75
Appendix F: Precision-Recall Curve of Each Experiment in Advanced Analysis----------76

1. Bowley, C., Andes, A., Ellis-Felege, S., & Desell, T. (2016, October). Detecting wildlife in uncontrolled outdoor video using convolutional neural networks. In 2016 IEEE 12th International Conference on e-Science (e-Science) (pp. 251-259). IEEE.
2. Burghardt, T., & Ćalić, J. (2006). Analysing animal behaviour in wildlife videos using face detection and tracking. IEE Proceedings-Vision, Image and Signal Processing, 153(3), 305-312.
3. Cheema, G. S., & Anand, S. (2017, September). Automatic detection and recognition of individuals in patterned species. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (pp. 27-38). Springer, Cham.
4. Chen, G., Han, T. X., He, Z., Kays, R., & Forrester, T. (2014, October). Deep convolutional neural network based species recognition for wild animal monitoring. In 2014 IEEE International Conference on Image Processing (ICIP) (pp. 858-862). IEEE.
5. Choi, S. (2015). Fish Identification in Underwater Video with Deep Convolutional Neural Network: SNUMedinfo at LifeCLEF Fish task 2015. In CLEF (Working Notes).
6. Chu, W., & Cai, D. (2018). Deep feature based contextual model for object detection. Neurocomputing, 275, 1035-1042.
7. Chuang, M. C., Hwang, J. N., & Williams, K. (2016). A feature learning and object recognition framework for underwater fish images. IEEE Transactions on Image Processing, 25(4), 1862-1872.
8. Forson, E. (2017, Nov 18). Understanding SSD MultiBox — Real-Time Object Detection In Deep Learning. Medium. Retrieved August 12, 2019, from https://towardsdatascience.com/understanding-ssd-multibox-real-time-object-detection-in-deep-learning-495ef744fab
9. Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. (2016). Deep learning. vol. 1.
10. Guo, Y., Liu, Y., Oerlemans, A., Lao, S., Wu, S., & Lew, M. S. (2016). Deep learning for visual understanding: A review. Neurocomputing, 187, 27-48.
11. Han, J., Zhang, D., Cheng, G., Liu, N., & Xu, D. (2018). Advanced deep-learning techniques for salient and category-specific object detection: a survey. IEEE Signal Processing Magazine, 35(1), 84-100.
12. Huang, T. (2018, July 10). Machine Learning/Statistics Method: Model Evaluation—Validation Index. Medium. Retrieved July 4, 2019, from https://medium.com/@chih.sheng.huang821
13. Hulstaert, L. (2018, Apr 23). Going deep into object detection. Medium. Retrieved August 12, 2019, from https://towardsdatascience.com/going-deep-into-object-detection-bed442d92b34
14. Jäger, J., Rodner, E., Denzler, J., Wolff, V., & Fricke-Neuderth, K. (2016). SeaCLEF 2016: Object Proposal Classification for Fish Detection in Underwater Videos. In CLEF (Working Notes) (pp. 481-489).
15. Kathuria, A. (2018, April 23). What’s new in YOLO v3? Medium. Retrieved July 4, 2019, from https://towardsdatascience.com/yolo-v3-object-detection-53fb7d3bfe6b
16. Kellenberger, B., Volpi, M., & Tuia, D. (2017, July). Fast animal detection in UAV images using convolutional neural networks. In 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) (pp. 866-869). IEEE.
17. Kortli, Y., Jridi, M., Al Falou, A., & Atri, M. (2018, April). A comparative study of CFs, LBP, HOG, SIFT, SURF, and BRIEF techniques for face recognition. In Pattern Recognition and Tracking XXIX (Vol. 10649, p. 106490M). International Society for Optics and Photonics.
18. Kotsiantis, S. B., Zaharakis, I., & Pintelas, P. (2007). Supervised machine learning: A review of classification techniques. Emerging artificial intelligence applications in computer engineering, 160, 3-24.
19. Kumar, G., & Bhatia, P. K. (2014, February). A detailed review of feature extraction in image processing systems. In 2014 Fourth international conference on advanced computing & communication technologies (pp. 5-12). IEEE.
20. Li, X., Shang, M., Hao, J., & Yang, Z. (2016, April). Accelerating fish detection and recognition by sharing CNNs with objectness learning. In OCEANS 2016-Shanghai (pp. 1-5). IEEE.
21. Li, X., Shang, M., Qin, H., & Chen, L. (2015, October). Fast accurate fish detection and recognition of underwater images with fast r-cnn. In OCEANS 2015-MTS/IEEE Washington (pp. 1-5). IEEE.
22. Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017). Focal loss for dense object detection. In Proceedings of the IEEE international conference on computer vision (pp. 2980-2988).
23. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016, October). Ssd: Single shot multibox detector. In European conference on computer vision (pp. 21-37). Springer, Cham.
24. Lu, D., & Weng, Q. (2007). A survey of image classification methods and techniques for improving classification performance. International journal of Remote sensing, 28(5), 823-870.
25. Mehrnejad, M., Albu, A. B., Capson, D., & Hoeberechts, M. (2013, September). Detection of stationary animals in deep-sea video. In 2013 OCEANS-San Diego (pp. 1-5). IEEE.
26. Nath, S. S., Mishra, G., Kar, J., Chakraborty, S., & Dey, N. (2014, July). A survey of image classification methods and techniques. In 2014 International Conference on Control, Instrumentation, Communication and Computational Technologies (ICCICCT) (pp. 554-557). IEEE.
27. Zeng, N. (2018, December 5). RetinaNet Explained and Demystified. Academic. Retrieved August 12, 2019, from https://blog.zenggyu.com/en/post/2018-12-05/retinanet-explained-and-demystified/
28. Parham, J., & Stewart, C. (2016, March). Detecting plains and Grevy's Zebras in the realworld. In 2016 IEEE Winter Applications of Computer Vision Workshops (WACVW) (pp. 1-9). IEEE.
29. ping Tian, D. (2013). A review on image feature extraction and representation techniques. International Journal of Multimedia and Ubiquitous Engineering, 8(4), 385-396.
30. Schneider, S., Taylor, G. W., & Kremer, S. (2018, May). Deep learning object detection methods for ecological camera trap data. In 2018 15th Conference on Computer and Robot Vision (CRV) (pp. 321-328). IEEE.
31. Sharma, S., Shah, D., Bhavsar, R., Jaiswal, B., & Bamniya, K. (2014, January). Automated detection of animals in context to indian scenario. In 2014 5th International Conference on Intelligent Systems, Modelling and Simulation (pp. 334-338). IEEE.
32. Spampinato, C., Chen-Burger, Y. H., Nadarajan, G., & Fisher, R. B. (2008). Detecting, Tracking and Counting Fish in Low Quality Unconstrained Underwater Videos. VISAPP (2), 2008(514-519), 1.
33. Sung, M., Yu, S. C., & Girdhar, Y. (2017, June). Vision based real-time fish detection using convolutional neural network. In OCEANS 2017-Aberdeen (pp. 1-6). IEEE.
34. Villon, S., Chaumont, M., Subsol, G., Villéger, S., Claverie, T., & Mouillot, D. (2016, October). Coral reef fish detection and recognition in underwater videos by supervised machine learning: Comparison between Deep Learning and HOG+ SVM methods. In International Conference on Advanced Concepts for Intelligent Vision Systems (pp. 160-171). Springer, Cham.
35. Walther, D., Edgington, D. R., & Koch, C. (2004, June). Detection and tracking of objects in underwater video. In Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.(Vol. 1, pp. I-I). IEEE.
36. Zhang, W., Sun, J., & Tang, X. (2010). From tiger to panda: animal head detection. IEEE Transactions on Image Processing, 20(6), 1696-1708.
37. Zhang, Z., He, Z., Cao, G., & Cao, W. (2016). Animal detection from highly cluttered natural scenes using spatiotemporal object region proposals and patch verification. IEEE Transactions on Multimedia, 18(10), 2079-2092.
38. Zhao, Z. Q., Zheng, P., Xu, S. T., & Wu, X. (2019). Object detection with deep learning: A review. IEEE transactions on neural networks and learning systems.
39. Zhuang, P., Xing, L., Liu, Y., Guo, S., & Qiao, Y. (2017). Marine Animal Detection and Recognition with Advanced Deep Learning Models. In CLEF (Working Notes).