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 U0026-3007201922575300
論文名稱(中文) 道路交通之事故辨識與碰撞機率估計: SVM與Random Forests之應用
論文名稱(英文) Accident Identification and Collision Probability Estimation for Roadway Traffic: Applications of SVM and Random Forests
校院名稱 成功大學
系所名稱(中) 交通管理科學系
系所名稱(英) Department of Transportation & Communication Management Science
學年度 107
學期 2
出版年 108
研究生(中文) 黃柏宗
研究生(英文) Po-Tsung Huang
電子信箱 bt2011aa@gmail.com
學號 R56061025
學位類別 碩士
語文別 英文
論文頁數 144頁
口試委員 指導教授-胡大瀛
口試委員-林佐鼎
口試委員-董啟崇
口試委員-盧宗成
口試委員-許聿廷
中文關鍵字 交通事故  自駕車  事故預防 
英文關鍵字 Traffic Accidents  Autonomous Vehicles  Accident Prevention 
學科別分類
中文摘要 近年來,伴隨著資訊與通訊科技的發展與進步,各國政府積極地推行法規和安全措施,並以增進安全、效率與改善交通問題的智慧型運輸系統來預防交通事故之發生,其乃透過分析大數據和距離等實質的數據來判斷道路安全風險,並給予駕駛人應有的提示與警示;除此之外,大眾對於自駕車的期許也隨著時間日益增高,希望提高自駕車風險預測及事故預防等機制來提升行車之安全。
本研究以自駕車人工智慧的角度,藉此模擬自駕車判讀外在環境或移動物件時的運作方式,如果有移動的物件被歸類為會發生交通事故的潛在威脅,此時該如何去定義兩車發生交通事故的碰撞機率。本研究將以SVM(Support Vector Machine)與Random forests作為使用之研究方法,透過從肇事影片當中擷取特定之特徵值,並且一併求得在肇事發生之前,各個特徵值在不同時間格之變化。
爾後,將不同時間格之特徵值匯入這兩種演算法以訓練出高準確度的模型,此舉能用以辨識交通事故在不同情況下的碰撞機率為何。相互比較兩種演算法的訓練結果,除了能得知何種演算法具有較優良的表現之外,本研究將以擁有最佳結果的演算法作為採用的高準確率模型,未來匯入一部影片之特徵值,該模型將能夠預測出該影片當中某一時間格之碰撞機率為何。
英文摘要 Currently, with the development and progress of communication technology, governments all over the world also vigorously promote regulations and security measures. They utilize the intelligent transportation systems (ITS) to not only improve safety, efficiency, and traffic problems, but also prevent traffic accidents. Through analyzing big data, distance, and other substantial data, ITS not only infers the risks of road safety, but also gives drivers the necessary promptings and cautions. Furthermore, the expectation of people for autonomous vehicles are also increasing because they hope that autonomous vehicles can be improved the risk prediction and accident prevention mechanisms to achieve driving safety.
This study simulates the operating method of autonomous vehicles judging the external environment or moving objects from the angle of artificial intelligence. If there has moving objects classified as the potential threat of occurring traffic accidents, how we should define the collision probability between two cars. This study chooses support vector machine (SVM) and Random Forests as our methodology. Through capturing the specific features from the videos of traffic accidents, we acquire the change of every feature in different time periods before happening traffic accidents.
After that, the features in different time periods are imported into two algorithms to train a high-accuracy model to identify the collision probability between two cars in different situations. Comparing the training results of two algorithms, we can know which algorithm has better performance, and adopt the algorithm with the best results as the high-accuracy model. In the future, this model can predict the collision probability in certain time period of video when we import a video’s features.
論文目次 ABSTRACT I
摘要 II
誌謝 III
TABLE OF CONTENTS IV
LIST OF TABLES VIII
LIST OF FIGURES X
CHAPTER 1 INTRODUCTION 1
1.1 Research Background and Motivation 1
1.2 Research Objectives 4
1.3 Research Flow Chart 6
CHAPTER 2 LITERATURE REVIEW 10
2.1 Autonomous Vehicles 10
2.1.1 The Development Status of Autonomous Vehicles 11
2.1.2 Advanced Driver Assistance System (ADAS) 12
2.2 Classification of Collision Types for Traffic Accidents 16
2.3 Definition of the Features 20
2.4 The Advantages of SVM 25
2.4.1 The Applications of SVM in Traffic Accident Prevention 25
2.4.2 The Performance between SVM with Other Algorithms 27
2.4.3 Algorithm for Improving the Efficiency of SVM 28
2.4.4 The Effect of Samples’ Numbers on SVM 29
2.5 Random Forests 30
2.5.1 The Difference and Advantage of Random Forests than SVM 31
2.5.2 The Application in Traffic Accident Prevention 32
2.5.3 The Performance between Random Forests with Other Algorithms 37
2.6 Vehicle Collision Probability 39
2.7 Comparison of Cross Validation in Different Folds 41
2.8 Summary 46
CHAPTER 3 RESEARCH METHODOLOGY 47
3.1 Problem Statement 47
3.2 Research Framework 48
3.3 SVM 52
3.3.1 Support Vector Machine (SVM) 52
3.3.2 Algorithm for Improving the Efficiency of SVM 55
3.3.3 LibSVM 57
3.4 Random Forests 58
3.4.1 Random Forests 58
3.4.2 Classification and Regression Tree (CART) 60
3.4.3 Bootstrap and Bagging 62
3.5 Scikit-learn 63
3.6 Definition of Vehicle Collision Probability 65
3.7 10-fold Cross Validation 67
3.8 Evaluation Criteria of Building Modals 68
CHAPTER 4 ACCIDENT IDENTIFICATION 71
4.1 Selection of the Features 73
4.1.1 Dynamic Features 73
4.1.2 Static Features 75
4.2 Data Collection and Illustration 78
4.2.1 Accident Data 78
4.2.2 No-accident Data 79
4.2.3 Calculation Method 79
4.3 SVM 83
4.3.1 The Parameters of LibSVM 83
4.3.2 Construction of the Kernel Function 85
4.4 Random Forests 87
4.4.1 The Parameters of Random Forests 87
4.4.2 Random Search 89
4.5 Experiment Results and Analysis 91
4.5.1 The Evaluating Result of SVM 91
4.5.2 The Evaluating Result of Random Forests 98
4.6 Comparison of the Performance between SVM and Random Forests 101
4.7 Summary 104
CHAPTER 5 VEHICLE COLLISION PROBABILITY 105
5.1 Estimating of the Vehicle Collision Risk 105
5.1.1 Feature Importance Ranking 106
5.1.2 Visualization of CART in Random Forests 111
5.2 Heuristic Algorithm 115
5.2.1 Vehicle Collision Risk 115
5.2.2 The Evaluating Result and Analysis 126
5.3 Multiple Classification Problem 128
5.4 Comparison of the Performance between SVM and Random Forests 133
5.5 Summary 135
CHAPTER 6 CONCLUSIONS AND SUGGESTIONS 137
6.1 Conclusions 137
6.2 Suggestions 139
REFERENCE 141
參考文獻 1. Ammoun, S., & Nashashibi, F. (2009). ‘‘Real time trajectory prediction for collision risk estimation between vehicles.’’ In 2009 IEEE 5th International Conference on Intelligent Computer Communication and Processing, pp. 417-422.
2. Arbabzadeh, N., & Jafari, M. (2018). ‘‘A data-driven approach for driving safety risk prediction using driver behavior and roadway information data.’’ IEEE Transactions on Intelligent Transportation Systems, Vol. 19, No. 2, pp. 446-460.
3. Basso, F., Basso, L. J., Bravo, F., & Pezoa, R. (2018). ‘‘Real-time crash prediction in an urban expressway using disaggregated data.’’ Transportation Research Part C: Emerging Technologies, Vol. 86, pp. 202-219.
4. Bergstra, J., & Bengio, Y. (2012). ‘‘Random search for hyper-parameter optimization.’’ Journal of Machine Learning Research, Vol. 13, No. Feb, pp. 281-305.
5. Bosch, A., Zisserman, A., & Munoz, X. (2007). ‘‘Image classification using random forests and ferns.’’ 2007 IEEE 11th international conference on computer vision, pp. 1-8.
6. Breiman, L. (2001). ‘‘Random forests.’’ Machine learning, Vol. 45, No. 1, pp. 5-32.
7. Chang, C.-C., & Lin, C.-J. (2011). ‘‘LIBSVM: a library for support vector machines.’’ ACM transactions on intelligent systems and technology (TIST), Vol. 2, No. 3, pp. 27.
8. Chen, G.-C. (1998). ‘‘The Analysis of the Cause of accidents in Different Collision Type Behaviors and Responsibility Identification.’’ Road Traffic Safety and Law Enforcement Seminar.
9. Chen, S., Wang, W., & Van Zuylen, H. (2009). ‘‘Construct support vector machine ensemble to detect traffic incident.’’ Expert systems with applications, Vol. 36, No. 8, pp. 10976-10986.
10. Chou, W.-S., & Chen, W.-W. (2001). ‘‘Decision Support System for Analyzing Causes of Traffic Accidents.’’ Transportation Planning Journal, Vol. 30, No. 3, pp. 635-656.
11. Geisser, S., & Eddy, W. F. (1979). ‘‘A predictive approach to model selection.’’ Journal of the American Statistical Association, Vol. 74, No. 365, pp. 153-160.
12. Guo, X.-E., Wu, W., Liu, C.-G., & Zhang, J.-A. (2010). ‘‘Research of Multi-class Support Vector Machines Classification Algorithm.’’ Journal of Shanxi Datong University, Vol. 26, No. 3, pp. 6-8+14.
13. Harb, R., Yan, X., Radwan, E., & Su, X. (2009). ‘‘Exploring precrash maneuvers using classification trees and random forests.’’ Accident Analysis & Prevention, Vol. 41, No. 1, pp. 98-107.
14. Hastie, T., Tibshirani, R., & Friedman, J. (2001). ‘‘The elements of statistical learning.’’ Springer Series in Statistics. New York, NY, USA: Springer New York Inc.
15. Hsu, C.-W., Chang, C.-C., & Lin, C.-J. (2003). ‘‘A practical guide to support vector classification.’’.
16. Joachims, T. (1998). ‘‘Text categorization with support vector machines: Learning with many relevant features.’’ European conference on machine learning, pp. 137-142.
17. Joerer, S., Segata, M., Bloessl, B., Cigno, R. L., Sommer, C., & Dressler, F. (2014). ‘‘A vehicular networking perspective on estimating vehicle collision probability at intersections.’’ IEEE Transactions on Vehicular Technology, Vol. 63, No. 4, pp. 1802-1812.
18. Kecman, V. (2005). ‘‘Support vector machines–an introduction.’’ Support vector machines: theory and applications, pp. 1-47.
19. Kim, J., & Kum, D. (2017). ‘‘Collision Risk Assessment Algorithm via Lane-Based Probabilistic Motion Prediction of Surrounding Vehicles.’’ IEEE Transactions on Intelligent Transportation Systems.
20. Kohavi, R. (1995). ‘‘A study of cross-validation and bootstrap for accuracy estimation and model selection.’’ IJCAI, Vol. 14, No. 2, pp. 1137-1145.
21. Li, X., Lord, D., Zhang, Y., & Xie, Y. (2008). ‘‘Predicting motor vehicle crashes using support vector machine models.’’ Accident Analysis & Prevention, Vol. 40, No. 4, pp. 1611-1618.
22. Lin, Z.-P. (2017). ‘‘The LiDAR Applied to Autonomous Vehicles.’’ Science Monthly, No. 431, Retrieved from http://scitechreports.blogspot.com/2017/11/lidar.html.
23. Lu, M., Wevers, K., & Van Der Heijden, R. (2005). ‘‘Technical Feasibility of Advanced Driver Assistance Systems (ADAS) for Road Traffic Safety.’’ Transportation Planning Technology, Vol. 28, No. 3, pp. 167-187.
24. National Police Agency MOI. (2015). ‘‘Accident Statistics.’’, Retrieved from https://www.npa.gov.tw/NPAGip/wSite/lp?ctNode=12744&nowPage=1&pagesize=15.
25. Platt, J. (1998). ‘‘Sequential minimal optimization: A fast algorithm for training support vector machines.’’ Technical Report MSR-TR-98-14, Microsoft Research.
26. Savino, G., Giovannini, F., Fitzharris, M., & Pierini, M. (2016). ‘‘Inevitable collision states for motorcycle-to-car collision scenarios.’’ IEEE Transactions on Intelligent Transportation Systems, Vol. 17, No. 9, pp. 2563-2573.
27. Scikit-learn. (2018). ‘‘Choosing the right estimator.’’, Retrieved from https://scikit-learn.org/stable/tutorial/machine_learning_map/index.html.
28. Shan, Z., Zhu, Q., & Zhao, D. (2017). ‘‘Vehicle collision risk estimation based on RGB-D camera for urban road.’’ Multimedia Systems, Vol. 23, No. 1, pp. 119-127.
29. Srinidhi, S. (2018). ‘‘Different types of Validations in Machine Learning (Cross Validation).’’ The Tech Check, Retrieved from https://blog.contactsunny.com/data-science/different-types-of-validations-in-machine-learning-cross-validation.
30. Taiwan Area National Freeway Bureau. (2007). ‘‘Introduction to the Intelligent Transportation Systems (ITS).’’, Retrieved from https://www.freeway.gov.tw/Publish.aspx?cnid=1556.
31. Tang, Y., Krasser, S., He, Y., Yang, W., & Alperovitch, D. (2008). ‘‘Support vector machines and random forests modeling for spam senders behavior analysis.’’ In IEEE GLOBECOM 2008-2008 IEEE Global Telecommunications Conference, pp. 1-5.
32. Valdimir, V., & Vapnik, N. (1995). ‘‘The nature of statistical learning theory.’’ Springer science & business media.
33. Wang, R.-B., You, F., Cui, G.-J., & Yu, T.-H. (2005). ‘‘Analysis on Lane-changing Safety of Vehicle.’’ Journal of Jilin University of Technology (Natural Science Edition), Vol. 35, No. 2, pp. 179-182.
34. Wang, W.-J., Wang, C.-L., Wang, J., & Zhao, X.-Z. (2010). ‘‘Highway Minimum Driving Safe Distance Calculation and Early Warning System Design.’’ Journal of Dalian Chiao Tung University, Vol. 31, No. 5, pp. 6-10+20.
35. Wikipedia. (2008). ‘‘Cross validation.’’, Retrieved from https://zh.wikipedia.org/wiki/%E4%BA%A4%E5%8F%89%E9%A9%97%E8%AD%89.
36. Witten, I. H., Hall, M., & Frank, E. (2005). ‘‘Data Mining: Practical Machine Learning Tools Techniques.’’ Morgan Kaufmann.
37. World Health Organization. (2017). ‘‘Top Ten Causes of Death.’’, Retrieved from http://www.who.int/zh/news-room/fact-sheets/detail/the-top-10-causes-of-death.
38. Xu, Q. (2017). ‘‘The Development and Follow-up of Autonomous Vehicles.’’ Taiwan Economic Research Monthly Journal, Vol. 40, No. 11, pp. 50-61.
39. Yu, H.-B., & Zeng, Z.-Q. (2009). ‘‘Sequential Minimal Optimization Algorithm Based on Parallel Processing.’’ Journal of Zhengzhou University of Light Industry, Vol. 24, No. 3, pp. 28-31.
40. Zaklouta, F., Stanciulescu, B., & Hamdoun, O. (2011). ‘‘Traffic sign classification using kd trees and random forests.’’ The 2011 International Joint Conference on Neural Networks, pp. 2151-2155.
41. Zhang, X.-G. (2000). ‘‘Introduction to Statistical Learning Theory and Support Vector Machines.’’ Acta Automatica Sinica, Vol. 26, No. 1, Retrieved from http://see.xidian.edu.cn/faculty/chzheng/bishe/indexfiles/New_Folder/about.PDF.
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2024-07-31起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2024-07-31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