跳到主要內容

臺灣博碩士論文加值系統

(44.200.86.95) 您好!臺灣時間:2024/05/28 08:43
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:周克霖
研究生(外文):CHOU, KE-LIN
論文名稱:基於深度學習之即時火災偵測
論文名稱(外文):Real-Time Fire Detection Based On Deep Learning
指導教授:康立威
指導教授(外文):KANG, LI-WEI
口試委員:林智揚許朝詠
口試委員(外文):LIN, CHIH-YANGHSU,CHAO-YUNG
口試日期:2019-07-10
學位類別:碩士
校院名稱:國立雲林科技大學
系所名稱:資訊工程系
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:2019
畢業學年度:107
語文別:中文
論文頁數:40
中文關鍵詞:深度學習卷積神經網路YOLO火災偵測
外文關鍵詞:Deep learningConvolutional Neural NetworkYOLOFire detection
相關次數:
  • 被引用被引用:1
  • 點閱點閱:574
  • 評分評分:
  • 下載下載:77
  • 收藏至我的研究室書目清單書目收藏:0
近年來,影像處理、物體辨識變成非常熱門的議題,主要是因為深度學習及卷積神經網路的興起,由於此進步,不僅開啟了人類歷史性的一頁、帶給人們便利性,現在也應用在許多層面上,例如:生活、娛樂或是安全等等。而本論文將把此技術應用在火災偵測上。
長久以來,由於火災的原因,導致有不少人受傷甚至死亡,所以為了預防火災,人們做很多研究或是透過許多安全措施來預防,但是儘管優秀的研究或完善的安全措施都無法完全預防火災,因此本論文著重在發生火災時的情況,當火災發生時,透過其方法能夠有效偵測到火災,然而採取措施,例如:聯繫消防局,救護車,警察等方式來減少火災帶來的傷害。
本篇論文的方法是基於YOLOv3-tiny 的架構來偵測火災,此方法不僅可以即時偵測火災的位置,還可以偵測不同場景所發生的火災,相信透過此方法一定能夠大幅度的減少火災傷亡人數。
關鍵字:深度學習、卷積神經網路、YOLO、火災偵測
In recent years, due to the rise of deep learning and convolutional neural network,image processing and object recognition have become very popular topics, which is not only opened up a historical page of human beings, but also brought convenience to people, and now used on many levels, such as life, entertainment, safety. This paper will use the method to do fire detection.
For a long time, many people have been injured or even killed due to fires, so in order to prevent fires, people have done a lot of research and adopted many safety measures, but they can’t be completely prevented despite excellent research or perfect safety measures, therefore, our paper focuses on the situation of fire happening, when fire happening, our method will detect the fire, and then do some ways to reduce the damage caused by fire, such as, contact to fire station, ambulance, police.
The method of this paper is based on YOLOv3-tiny to detect fires. This method can not only detect the location of fires in real time, but also detect the fires in different scenes. We believe that this method will definitely reduce the number of people who get hurt or die in the fire.
Keyword:Deep learning, Convolutional Neural Network, YOLO, Fire detection
摘要 i
ABSTRACT ii
誌謝 iii
目錄 iv
表目錄 vi
圖目錄 vii
第壹章、簡介 1
1.1 研究動機 1
1.2 研究目的 2
1.3 文獻回顧 2
1.3.1 火災偵測 2
1.4 論文架構 4
第貳章、相關技術討論 5
2.1 CNN(Convolutional Neural Network) 5
2.1.1 卷積層(Convolution Layer) 6
2.1.2 池化層(Pooling Layer) 7
2.1.3 全連接層(Fully Connected Layer) 8
2.2 R-CNN(Regions with CNN features) 9
2.2.1 Fast R-CNN 9
2.2.2 Faster R-CNN 10
2.3 YOLO(You Only Look Once) 11
2.3.1 YOLOv1 11
2.3.2 YOLOv2 12
2.3.3 YOLOv3 14
第參章、研究方法 16
3.1 資料集介紹 16
3.2 實驗流程 19
3.3 實驗架構與方法 20
第肆章、研究結果 22
4.1 環境架設 22
4.2 參數調整 23
4.3 實驗結果 26
4.4 相關論文方法比較 27
第伍章、結論與未來展望 28
參考文獻 29
[1] 102年至107年全台火災次數統計表:https://www.nfa.gov.tw/cht/index.php
[2] V. Hüttner, C. R. Steffens, S. S. da C. Botelho, “First response fire combat: Deep leaning based visible fire detection” in 2017 Latin American Robotics Symposium (LARS) and 2017 Brazilian Symposium on Robotics (SBR). pp. 1–6, 2017.
[3] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna,“Rethinking the inception architecture for computer vision,” The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2818-2826, 2016.
[4] C. Hu, P. Tang, W. Jin, Z. He, W. Li, “Real-Time Fire Detection Based on Deep Convolutional Long-Recurrent Networks and Optical Flow Method” in Proceedings of the 37th Chinese Control Conference (CCC), pp. 9061–9066, July 2018.
[5] A. Krizhevsky, I. Sutskever, G E.Hinton “ImageNet Classification with Deep Convolutional Neural Networks” in Proceeding of the 25th International Conference on Neural Information Processing Systems, vol. 1, pp.1097-1105, December 2012.
[6] S. Ren, K. He, R. Girshick, J. Sun. “Faster r-cnn: Towards real-time object detection with region proposal networks”, arXiv preprint arXiv:1506.01497, 2015.
[7] P. Barmpoutis, K. Dimitropoulos, K. Kaza, N. Grammalidis “Fire Detection from Images Using Faster R-CNN and Multidimensional Texture Analysis”, in 2019 IEEE International Conference on Acoustics. ICASSP 2019.
[8] Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, “Gradient-Based Learning Applied to Document Recognition,” in Proceedings of the IEEE, vol. 18, pp.2278-2324, November 1998.
[9] K. Simonyan, A. Zisserman “Very deep convolutional networks for large-scale image recognition”, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.770-778, 2016.
[10] C. Szegedy, W. Liu, Y. Jia, et al. “Going deeper with convolutions” in Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA, June 2015.
[11] K. He, X. Zhang, S. Ren, et al. “Deep residual learning for image recognition” in Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, June 2016.
[12] D. Lowe, "Distinctive Image Features from Scale-Invariant Keypoints",
International Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, Nov. 2004.
[13] R. Girshick, J. Donahue, T. Darrell, J. Malik, “Rich feature hierarchies for accurate object detection and semantic segmentation” in CVPR, 2014.
[14] K. He, X. Zhang, S. Ren, J. Sun. “Spatial pyramid pooling in deep convolutional networks for visual recognition” in ECCV, 2014.
[15] R. Girshick, “Fast R-CNN” in IEEE International Conference on Computer Vision (ICCV), 2015. [16] J. Redmon, S. Divvala, R. Girshick, 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.
[17] J. Redmon and A. Farhadi. “Yolo9000: Better, faster, stronger” in Computer Vision and Pattern Recognition (CVPR), 2017 IEEE Conference on, pp.6517–6525. IEEE, 2017.
[18] J. Redmon, Ali Farhadi, “YOLOv3: An Incremental Improvement”, in Computer Vision and Pattern Recognition 2018.
[19] T. Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, S. Belongie, “Feature pyramid networks for object detection”, arXiv preprint arXiv:1612.03144, 2016.
[20] YOLOv3架構圖:https://mropengate.blogspot.com/2018/06/yolo-yolov3.html
[21] Furg-fire-dataset:https://github.com/steffensbola/furg-fire-dataset#furg-fire-dataset
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊