進階搜尋


   電子論文尚未授權公開,紙本請查館藏目錄
(※如查詢不到或館藏狀況顯示「閉架不公開」,表示該本論文不在書庫,無法取用。)
系統識別號 U0026-1308201617352700
論文名稱(中文) 應用機器系學習機制改善無人機地面物件偵測
論文名稱(英文) Improving UAV Ground Object Detection with Application of Machine Learning Mechanism
校院名稱 成功大學
系所名稱(中) 民航研究所
系所名稱(英) Institute of Civil Aviation
學年度 104
學期 2
出版年 105
研究生(中文) 李駿宗
研究生(英文) Chun-Tsung Lee
學號 Q46031035
學位類別 碩士
語文別 英文
論文頁數 84頁
口試委員 指導教授-林清一
口試委員-葉泳蘭
口試委員-彭兆仲
中文關鍵字 車輛辨識  機器學習法  灰階化特徵  模糊影像  消除雜訊及噪聲  局部二值特徵  自適應增強演算法 
英文關鍵字 Vehicle detection  Gray level  Blurring  Median filter  Machine learning  Haar-like feature  LBP feature  AdaBoost Algorithm  Cascading Classifier 
學科別分類
中文摘要 現今主要偵測方法分為兩大類,分別為地面監視和空中偵察。地面監視有偵測範圍過小、環境限制及花費時間較多等缺點。在環境條件的限制下,使用空中偵測較適合用於大範圍的影像辨識及追蹤目標之偵測,並且對於複雜環境的機動性較高。本系統是建立在利用機器學習法來偵測及識別物件,整個系統流程包括訓練樣本及測試系統之可行性兩部分。對於物件識別來說,樣本收集為前置作業中極為重要的一部分,在此系統中主要是利用汽車後方通用且明顯的部分來當作訓練特徵,包括擋風玻璃、雨刷及車燈。本文採用兩種不同的機器學系法做測試及比較,分別為Haar-like學習法的圖像灰度變化特徵及LBP學習法的局部二值特徵定理。由最後結果可知,透過影像處理可以有效改善整個系統的偵測率、誤判率及運行速度。本文一共實行了三個實驗,分為比較及分析此系統在不同的時間、高度及姿態下的實驗數據變化。從實驗結果顯示,系統可以在具有恆定姿態及高度的動態平台上有效的運行一整天來檢測各種車輛。此系統可以放置在高度及方向變化不大的UAV上或其他飛行裝置上,也可以應用發展成即時偵測系統。
關鍵字: 車輛辨識,機器學習法,灰階化特徵,模糊影像,消除雜訊及噪聲,局部二值特徵,自適應增強演算法
英文摘要 The present main image surveillance methods can be divided into two categories, aerial surveillance and ground surveillance. The disadvantages of ground surveillance are the smaller detection range, costing more time and environmental restriction. Aerial surveillance is more suitable for a much larger spatial area, and it has higher flexibility for complex environments. This aerial surveillance system is mainly based on the machine learning algorithm to detect the objects which on the ground. The process of proposed system are divided into two parts, training process and testing process. In system preparation, the most important part of object detection is sample collection. The system uses the common and obvious parts which in the vehicle behind as the the main characteristics, such as wind screen, wiper and lamps. The collected samples are trained by using diverse train cascaded Haar-like feature classifier and LBP feature classifier in this thesis. According the final experiment result, it can improve effectively the detection rate, false alarm rate and process speed after the image processing. In this thesis, three experiments are conducted, including the data changes at different time, different altitude and different attitude.
The final experiment results show the proposed detection system can run effectively on a dynamic platform all day with constant altitude to detect different vehicles. The proposed detection system also can be applied on the UAVs or other flying devices in constant altitude or constant direction, in order to be developed into further application, such as real-time surveillance.
Keywords: Vehicle detection, Gray level, Blurring, Median filter, Machine learning, Haar-like feature, LBP feature, AdaBoost Algorithm, Cascading Classifier.
論文目次 ABSTRACT.............I
摘要.............III
誌謝.............V
List of Figures.............VIII
List of Tables.............X
Chapter 1 Introduction.............1
1.1 Motivation.............1
1.2 Literature Survey.............2
1.3 Main Idea.............6
1.4 Goal.............7
1.5 Thesis Outline.............7
Chapter 2 Images and Features in Processing.............9
2.1 Feature sample collection.............9
2.2 Image processing.............14
2.2.1 Gray Level.............14
2.2.2 Blurring and Median Filter.............17
2.2.3 Histogram Equalization.............19
2.3 Remark.............21
Chapter 3 Vehicle Detection System Frame.............22
3.1.1 Haar-like feature.............23
3.1.2 Local Binary Patterns feature.............24
3.2 Boosting and Weak Classifier.............28
3.3 AdaBoost Algorithm.............29
3.4 Cascading Classifier.............35
3.5 System Structure.............37
Chapter 4 Experiment Result.............39
4.1 Experiment and Environment Set up.............39
4.2 Samples Training.............40
4.3 Video in vehicle detection experiments.............42
4.3.1 Fixed point at different time.............43
4.3.2 Fixed point at different height.............61
4.3.3 Dynamic experiment.............66
4.4 Merging two Machine Learning Algorithms.............76
Chapter 5 Conclusion and Future work.............79
References.............81
參考文獻 [1] J. Y. Choi, Y. K. Yang, “Vehicle detection from aerial images using local shape information”, Advances in Image and Video Technology, Vol. 5414, January 2009, pp. 227-236.
[2] L. Eikvil, L. Aurdal and H. Koren, “Classification-Based Vehicle Detection in High-Resolution Satellite Images”, ISPRS J. Photogrammetry and Remote Sensing, Vol. 64, No. 1, pp. 65-72, 2009.
[3] H. Grabner, T. Nguyen, B. Gruber and H. Bischof, “On-Line Boosting-Based Car Detection from Aerial Images”, ISPRS J. Photogrammetry and Remote Sensing, Vol. 63, No. 3, pp. 382-396, 2008.
[4] S. Hinz, A. Baumgartner, “Vehicle detection in aerial images using generic features, grouping, and context”, Proceedings of 23rd DAGM Symposium Munich, Vol. 2191, September 2001, pp. 45-52.
[5] S. Hinz, “Detection and Counting of Cars in Aerial Images”, Proc. International Conference on Image Processing, 2003.
[6] X. Shi, H. Ling, E. Blasch, W. Hu, “Context-driven moving vehicle detection in wide area motion imagery”, 21st IEEE International Conference on Pattern Recognition (ICPR), November 2012, pp. 2512-2515.
[7] J. Leitloff, S. Hinz and U. Stilla, "Vehicle detection in very high resolution satellite images of city areas", IEEE Transactions on Geoscience and Remote Sensing, Vol. 48, No. 7, pp. 2795-2806, 2010.
[8] S. Sahli, Y. Ouyang, D. A. Lavigne, “Robust vehicle detection in low-resolution aerial imagery”, International Society for Optics and Photonics, Vol. 7668, April 2010, pp. 76680G-76680G.
[9] X. Cao, C. Wu, P. Yan and X. Li, "Linear SVM classification using boosting hog features for vehicle detection in low-altitude airborne videos", Image Processing (ICIP), 2011 18th IEEE International Conference, pp. 2421-2424.
[10] R. E. Schapire, "The boosting approach to machine learning: An overview." Nonlinear estimation and classification. Springer New York, 2003. pp. 149-171.
[11] R. Lienhart, J. Maydt, “An extended set of haar-like features for rapid object detection”, 2002 International Conference on Image Processing, Vol. 1, 2002, pp. I-900 - I-903.
[12] T. P. Breckon, S. E. Barnes, M. L. Eichner, K. Wahren, “Autonomous real-time vehicle detection from a medium-level UAV”, 24th International Conference on Unmanned Air Vehicle Systems, 2009, pp. 29.1-29.9.
[13] T. Ojala, M. Pietikäinen, T. Mäenpää, “Multiresolution gray-scale and rotation invariant texture classification with local binary patterns”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, July 2002, pp. 971-987.
[14] Y. Freund, R. E. Schapire, “A decision-theoretic generalization of on-line learning and an application to boosting”, Journal of computer and system sciences, vol. 55, December 1997, pp. 119-139.
[15] T. Ojala, M. Pietikäinen, H. David, "A comparative study of texture measures with classification based on featured distributions." Pattern recognition January 1996, pp. 51-59.
[16] A. Kembhavi, D. Harwood, L. S. Davis, "Vehicle detection using partial least squares." IEEE Transactions on Pattern Analysis and Machine Intelligence June 2011, pp. 1250-1265.
[17] C. H. Yung, C. C. Weng, Y. Y. Chen, "Vehicle detection in aerial surveillance using dynamic Bayesian networks." IEEE transactions on image processing April 2012, pp. 2152-2159.
[18] R. Chellappa, G. Qian, Q. Zheng, "Vehicle detection and tracking using acoustic and video sensors." Acoustics, Speech, and Signal Processing, 2004. Proceedings.(ICASSP'04). IEEE International Conference on. Vol. 3. IEEE, 2004.
[19] C. C. Wang, S. S. Huang, L. C. Fu, "Driver assistance system for lane detection and vehicle recognition with night vision." 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2005.
[20] Z. Sun, G. Bebis, and R. Miller. "On-road vehicle detection: A review." IEEE transactions on pattern analysis and machine intelligence May 2006, pp. 694-711.
[21] C. C. Wang, C. Thorpe, S. Thrun, "Online simultaneous localization and mapping with detection and tracking of moving objects: Theory and results from a ground vehicle in crowded urban areas." Robotics and Automation, 2003. Proceedings. ICRA'03. IEEE International Conference on. Vol. 1. IEEE, 2003.
[22] S. Gupte, O. Masoud, R. F. K. Martin, N. P. Papanikolopoulos, "Detection and classification of vehicles." IEEE Transactions on intelligent transportation systems January 2002, pp. 37-47.
[23] D. M. Gavrila, "Pedestrian detection from a moving vehicle." European conference on computer vision. Springer Berlin Heidelberg, 2000.
[24] W. W. Wierwille, S. S. Wreggit, C. L Kirn, L. A. Ellsworth, “Research on vehicle-based driver status/performance monitoring; development, validation, and refinement of algorithms for detection of driver drowsiness.” final report. No. HS-808 247. 1994.
[25] R. Labayrade, D. Aubert, J. P. Tarel, "Real time obstacle detection in stereovision on non flat road geometry through" v-disparity" representation." Intelligent Vehicle Symposium, 2002. IEEE. Vol. 2. IEEE, 2002.
[26] R. Kasturi, D. Goldgof, P. Soundararajan, V. Manohar, J. Garofolo, R. Bowers, M. Boonstra, V. Korzhova, J. Zhang, "Framework for performance evaluation of face, text, and vehicle detection and tracking in video: Data, metrics, and protocol." IEEE Transactions on Pattern Analysis and Machine Intelligence February 2009. pp. 319-336.
[27] D. M. Gavrila, S. Munder, "Multi-cue pedestrian detection and tracking from a moving vehicle." International journal of computer vision January 2007. pp. 41-59.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2021-07-01起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw