(3.238.96.184) 您好!臺灣時間:2021/05/08 04:32
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:::

我願授權國圖
: 
twitterline
研究生:朱牧仁
研究生(外文):Mu-JenChu
論文名稱:利用時空特徵資料預測城市中區域性的違停案件數量
論文名稱(外文):Dynamic Sequential Prediction of Urban Vehicle Illegal-Parking Events Using Regional Spatial-Temporal Features
指導教授:解巽評
指導教授(外文):Hsun-Ping Hsieh
學位類別:碩士
校院名稱:國立成功大學
系所名稱:電腦與通信工程研究所
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:2020
畢業學年度:108
語文別:英文
論文頁數:36
中文關鍵詞:違停預測機器學習時空特徵長短期記憶模型時間序列預測
外文關鍵詞:Illegal-parking event predictionMachine learningSpatial-Temporal featureLong Short-Term Memory modelTime series forecasting
相關次數:
  • 被引用被引用:0
  • 點閱點閱:34
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
1980年代之後,隨著世界各國車輛日漸普及,我們的日常生活早已大大發生了改變。然而,除了各種移動上的便利性提升,卻也同時衍生了相當多負面的交通問題。其中最嚴重、混亂,並同時是我們幾乎每日出門一抬頭就必須面對的情況: 交通雍塞,已造成了不計其數的國家無法估計的經濟及生命財產損失。探究其根本原因,「違規停車」便是其罪魁禍首。違規停車必然會對我們的交通系統造成重大負擔,不僅僅是塞車,甚至是為數眾多的交通事故起因之一。而行人或騎士往往必須冒著生命危險閃避違規停車的現象更是令人難以容忍。為此,我們這篇論文主要致力於違規停車的數量預測。目標利用過往時間的地區性時空特徵,將各種特徵視為隨時間變化的圖片序列組,餵入模型進行特徵卷積、建模,進而預測出特定區域在連續時段的違停件數變化及消長趨勢。為此目標,我們提出了兩個以卷積長短期記憶模型(Convolutional Long Short-Term Memory model, ConvLSTM)為基底而改造的架構,來克服我們遭遇的困難,並且以最終實驗結果來看,我們提出的兩個架構都比其它的基本模型表現得更好。我們提出的題目及架構具有相當高的實用性及對於其它類似的預測問題或規劃問題也有相當不錯的泛用性與可應用性。
The fast growth in number of private vehicles has caused huge impact to our daily lives. There are too many countries to enumerate which faces the traffic congestion almost every day. Illegal parking is one of the culprits that causes heavy burden for our transportation system, which not only results in traffic jam but also leads to traffic accidents and even put the lives of pedestrians under risk. In this work, we focused on predicting the numbers of illegal parking events locally using past few hours’ regional features and proposed two novel frameworks based on convolutional long short-term memory model (ConvLSTM) to overcome the challenges we addressed. All the metrics we used to measure models’ performances in the experiment results part showed that our framework handled the task well enough to beat other baselines. Our work has a high reproducibility as well as flexibility for various kinds of applications. We expect it will play an important role in part of urban traffic system in the near future.
Abstract ------------------------------------------------------------------------------------ II
List of Figures ---------------------------------------------------------------------------- VI
List of Tables ------------------------------------------------------------------------------ VII
Chapter 1 INTRODUCTION --------------------------------------------------------- 1
Chapter 2 RELATED WORK ------------------------------------------------------- 4
Chapter 3 NOTATIONS AND DATASETS ---------------------------------------- 6
3.1 Problem Definition & Objectives ---------------------------------------- 6
3.2 Dataset & Features --------------------------------------------------------- 7
3.3 Data Analysis & Preprocessing -------------------------------------------- 9
Chapter 4 METHODOLOGY ------------------------------------------------------- 16
Chapter 5 EXPERIMENT ------------------------------------------------------------ 23
5.1 Data Splitting & Experimental Settings -------------------------------- 23
5.2 Evaluation Metrics --------------------------------------------------------- 25
5.3 Parameter Optimization --------------------------------------------------- 26
5.4 Experiment Results in Four Metrics ------------------------------------ 28
Chapter 6 CONCLUSION ------------------------------------------------------------ 32
Chapter 7 FUTURE WORK --------------------------------------------------------- 33
REFERENCES -------------------------------------------------------------------------- 34
[1] Frederic Bastien, Pascal Lamblin, Razvan Pascanu, James Bergstra, Ian Goodfellow, Arnaud Bergeron, Nicolas Bouchard, David Warde-Farley, and Y. Bengio. 2012. Theano: New features and speed improvements. NIPS 2012 Workshop.
[2] Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. 2014. Neural machine translation by jointly learning to align and translate. arXiv: 1409.0473.
[3] Zhiyong Cui, Ruimin Ke, Ziyuan Pu, and Yinhai Wang. 2017. Deep stacked bidirectional and unidirectional lstm recurrent neural network for network wide traffic speed prediction. KDD 2017 Workshops.
[4] Xiangyang Chen, and Ruqing Chen. 2019. A Review on Traffic Prediction Methods for Intelligent Transportation System in Smart Cities. 2019 12th International Congress on Image and Signal Processing, Bio-Medical Engineering and Informatics (CISP-BMEI)
[5] Dingxiong Deng, Cyrus Shahabi, Ugur Demiryurek, Linhong Zhu, Rose Yu, and Yan Liu. 2016. Latent space model for road networks to predict time-varying traffic. KDD 2016.
[6] Alex Graves. 2013. Generating sequences with recurrent neural networks. arXiv:1308.0850.
[7] Rui Jia, Pengcheng Jiang, Lei Liu, Lizhen Cui, and Yuliang Shi. 2018. Data driven congestion trends prediction of urban transportation. 2018 IEEE Internet of Things Journal.
[8] Yaguang Li, Rose Yu, Cyrus Shahabi, and Yan Liu. 2018. Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. ICLR 2018.
[9] Ioannis Loumiotis, Konstantinos Demestichas, Evgenia Adamopoulou, Pavlos Kosmides, Vasilis Asthenopoulos, and Efstathios Sykas. 2018. Road Traffic Prediction Using Artificial Neural Networks. IEEE 2018 South-Eastern European Design Automation, Computer Engineering, Computer Networks and Society Media Conference
[10]Rohan More, Abhishek Mugal, Sheetal Rajgure, Rahul Baliram Adhao, and Vinod Pachghare. 2016. Road traffic prediction and congestion control using artificial neural networks. 2016 International Conference on Computing, Analytics and Security Trends.
[11] Xiaolei Ma, Zhuang Dai, Zhengbing He, Jihui Na, Yong Wang, and Yunpeng Wang. 2017. Learning traffic as images: a deep convolutional neural network for large-scale transportation network speed prediction. arXiv:1701.04245.
[12] Xingjian Shi, Zhourong Chen, Hao Wang, Dit-Yan Yeung, Wai-kinWong, and Wang-chun Woo. 2015. Convolutional LSTM Network: A Machine Learning Approach for Precipitation Now casting. NIPS 2015.
[13] Changhee Song, Heeyun Lee, Changbeom Kang, Wonyoung Lee, Young B. Kim, and Suk W. Cha. 2017. Traffic speed prediction under weekday using convolutional neural networks concepts. 2017 IEEE Intelligent Vehicles Symposium (IV).
[14] Fei Wu, Hongjian Wang, and Zhenhui Li. 2016. Interpreting traffic dynamics using ubiquitous urban data. SIGSPATIAL 2016.
[15] Kelvin Xu, Jimmy Ba, Ryan Kiros, Kyunghyun Cho, Aaron Courville, Ruslan Salakhutdinov, Richard Zemel, and Yoshua Bengio. 2015. Show, attend and tell: Neural image caption generation with visual attention. ICML 2015.
[16] Huaxiu Yao, Fei Wu, Jintao Ke, Xianfeng Tang, Yitian Jia, Siyu Lu, Pinghua Gong, Jieping Ye, and Zhenhui Li. 2018. Deep multi-view spatial-temporal network for taxi demand prediction. AAAI 2018
[17] Jiangchuan Zheng, and Lionel M. Ni. 2013. Time-dependent trajectory regression on road networks via multi-task learning. AAAI 2013.
[18] Dongxu Zhang, and Dong Wang. 2015. Relation classification via recurrent neural network. arXiv:1508.01006.
[19] Shu Zhang, Dequan Zheng, Xinchen Hu, and Ming Yang. 2015. Bidirectional long short-term memory networks for relation classification. PACLIC 2015
[20] Peng Zhou, Wei Shi, Jun Tian, Zhenyu Qi, Bingchen Li, Hongwei Hao, and Bo Xu. 2016. Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification. ACL 2016.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔