A Novel Segment-Based Approach for Improving Classification Performance of Transport Mode Detection
<p>Multi-tiered architecture for transport mode detection.</p> "> Figure 2
<p>ROC of (<b>a</b>) walking activity detection by varying the values of <math display="inline"> <semantics> <mrow> <mi>T</mi> <msub> <mi>h</mi> <mi>W</mi> </msub> </mrow> </semantics> </math> and (<b>b</b>) stationary activity detection by varying the values of <math display="inline"> <semantics> <mrow> <mi>T</mi> <msub> <mi>h</mi> <mrow> <mi>S</mi> <mi>t</mi> </mrow> </msub> </mrow> </semantics> </math></p> "> Figure 3
<p>Results of initial classification (first row) and the Healing algorithm (second row) for the first special case.</p> "> Figure 4
<p>Results of initial classification (first row) and the Healing algorithm (second row) for the second special case.</p> "> Figure 5
<p>Results of initial classification (first row) and the Healing algorithm (second row) for the third special case.</p> "> Figure 6
<p>The effect of sampling frequency on the classification results.</p> "> Figure 7
<p>The effect of window size on the classification results.</p> "> Figure 8
<p>The effect of the overlapping ratio on the classification results.</p> "> Figure 9
<p>The effect of the number of features on the classification results.</p> "> Figure 10
<p>User interface of mobile Transport Mode Detection application. (<b>a</b>) Transport Mode Detection Screen; (<b>b</b>) results of Initial Transport Mode Detection and Healing algorithm for a given date; (<b>c</b>) Statistics of user actions.</p> "> Figure 11
<p>Contribution of the Healing algorithm to the recall of initial transport mode detection.</p> ">
Abstract
:1. Introduction
- A novel post-processing algorithm, namely Healing, for improving the results after transport mode classification
- Developing a new set of features for determining the type of movement.
- Determining the optimum window size and sampling frequency for feature extraction
2. Related Works
3. System Architecture
3.1. Data Acquisition
3.2. Initial Transport Mode Detection
3.2.1. Vehicular Activity Detection
3.2.2. Vehicular Activity Classification
3.3. Improvement of Classification Results by the Proposed Healing Algorithm
Algorithm 1 Healing algorithm. |
|
4. Experimental Results
4.1. Effects of Data Acquisition and Feature Extraction Parameters on Transport Mode Detection Performance
4.2. Performance Evaluation of the Initial Transport Mode Detection
4.2.1. Evaluation of Vehicular Activity Detection
4.2.2. Evaluation of Vehicular Activity Classification
4.3. Performance Evaluation of the Proposed Healing Algorithm
4.4. Performance Comparison to the State-Of-The-Art
5. Discussion
6. Conclusions
Author Contributions
Conflicts of Interest
References
- Ranasinghe, S.; Al Machot, F.; Mayr, H.C. A review on applications of activity recognition systems with regard to performance and evaluation. Int. J. Distrib. Sens. Netw. 2016, 12. 1550147716665520. [Google Scholar] [CrossRef]
- Garg, S.; Singh, P.; Ramanathan, P.; Sen, R. VividhaVahana: smartphone based vehicle classification and its applications in developing region. In Proceedings of the 11th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, London, UK, 2–5 December 2014; ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering): Brussels, Belgium, 2014; pp. 364–373. [Google Scholar]
- Semanjski, I.; Gautama, S. Smart city mobility application—Gradient boosting trees for mobility prediction and analysis based on crowdsourced data. Sensors 2015, 15, 15974–15987. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Singh, G.; Bansal, D.; Sofat, S.; Aggarwal, N. Smart patrolling: An efficient road surface monitoring using smartphone sensors and crowdsourcing. Pervasive Mob. Comput. 2017, 40, 71–88. [Google Scholar] [CrossRef]
- Gonçalves, J.; Gonçalves, J.S.; Rossetti, R.J.; Olaverri-Monreal, C. Smartphone sensor platform to study traffic conditions and assess driving performance. In Proceedings of the 2014 IEEE 17th International Conference on Intelligent Transportation Systems (ITSC), Qingdao, China, 8–11 October 2014; pp. 2596–2601. [Google Scholar]
- Shin, D.; Aliaga, D.; Tunçer, B.; Arisona, S.M.; Kim, S.; Zünd, D.; Schmitt, G. Urban sensing: Using smartphones for transportation mode classification. Comput. Environ. Urban Syst. 2015, 53, 76–86. [Google Scholar] [CrossRef]
- Ohmori, N.; Nakazato, M.; Harata, N.; Sasaki, K.; Nishii, K. Activity diary surveys using GPS mobile phones and PDA. In Proceedings of the 85th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 22–26 January 2006; pp. 22–26. [Google Scholar]
- Hemminki, S.; Nurmi, P.; Tarkoma, S. Accelerometer-based transportation mode detection on smartphones. In Proceedings of the 11th ACM Conference on Embedded Networked Sensor Systems, Roma, Italy, 11–15 November 2013; ACM: New York, NY, USA, 2013; p. 13. [Google Scholar]
- Nikolic, M.; Bierlaire, M. Review of transportation mode detection approaches based on smartphone data. In Proceedings of the 17th Swiss Transport Research Conference, Ascona, Switzerland, 18 May 2017. number EPFL-CONF-229181. [Google Scholar]
- Prelipcean, A.C.; Gidófalvi, G.; Susilo, Y.O. Transportation mode detection—An in-depth review of applicability and reliability. Transp. Rev. 2017, 37, 442–464. [Google Scholar] [CrossRef]
- Feng, T.; Timmermans, H.J. Comparison of advanced imputation algorithms for detection of transportation mode and activity episode using GPS data. Transp. Plan. Technol. 2016, 39, 180–194. [Google Scholar] [CrossRef]
- Ohashi, H.; Akiyama, T.; Yamamoto, M.; Sato, A. Modality classification method based on the model of vibration generation while vehicles are running. In Proceedings of the Sixth ACM SIGSPATIAL International Workshop on Computational Transportation Science, Orlando, FL, USA, 5–8 November 2013; ACM: New York, NY, USA, 2013; p. 37. [Google Scholar]
- Han, M.; Vinh, L.T.; Lee, Y.-K.; Lee, S. Comprehensive context recognizer based on multimodal sensors in a smartphone. Sensors 2012, 12, 12588–12605. [Google Scholar] [CrossRef]
- Widhalm, P.; Nitsche, P.; Brändie, N. Transport mode detection with realistic smartphone sensor data. In Proceedings of the 2012 21st International Conference on Pattern Recognition (ICPR), Tsukuba, Japan, 11–15 November 2012; pp. 573–576. [Google Scholar]
- Stenneth, L.; Wolfson, O.; Yu, P.S.; Xu, B. Transportation mode detection using mobile phones and GIS information. In Proceedings of the 19th ACM SIGSPATIAL International Conference on Advances in Geographic Information System, Chicago, IL, USA, 1–4 November 2011; ACM: New York, NY, USA, 2011; pp. 54–63. [Google Scholar]
- Stenneth, L.; Thompson, K.; Stone, W.; Alowibdi, J. Automated transportation transfer detection using GPS enabled smartphones. In Proceedings of the 2012 15th International IEEE Conference on Intelligent Transportation Systems (ITSC), Anchorage, AK, USA, 16–19 September 2012; pp. 802–807. [Google Scholar]
- Zhang, L.; Dalyot, S.; Eggert, D.; Sester, M. Multi-stage approach to travel-mode segmentation and classification of gps traces. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci. 2011, 38, 87–93. [Google Scholar] [CrossRef]
- Brunauer, R.; Hufnagl, M.; Rehrl, K.; Wagner, A. Motion pattern analysis enabling accurate travel mode detection from GPS data only. In Proceedings of the Intelligent Transportation Systems-(ITSC), The Hague, The Netherlands, 6–9 October 2013; pp. 404–411. [Google Scholar]
- Witayangkurn, A.; Horanont, T.; Ono, N.; Sekimoto, Y.; Shibasaki, R. Trip reconstruction and transportation mode extraction on low data rate gps data from mobile phone. In Proceedings of the International Conference on Computers in Urban Planning and Urban Management (CUPUM 2013), Utrecht, The Netherlands, 2–5 July 2013; pp. 1–19. [Google Scholar]
- Wang, S.; Chen, C.; Ma, J. Accelerometer based transportation mode recognition on mobile phones. In Proceedings of the 2010 Asia-Pacific Conference on Wearable Computing Systems (APWCS), Shenzhen, China, 17–18 April 2010; pp. 44–46. [Google Scholar]
- Bedogni, L.; Di Felice, M.; Bononi, L. By train or by car? Detecting the user’s motion type through smartphone sensors data. In Proceedings of the 2012 IFIP on Wireless Days (WD), Dublin, Ireland, 21–23 November 2012; pp. 1–6. [Google Scholar]
- Bedogni, L.; Di Felice, M.; Bononi, L. Context-aware Android applications through transportation mode detection techniques. Wirel. Commun. Mob. Comput. 2016, 16, 2523–2541. [Google Scholar] [CrossRef]
- Sankaran, K.; Zhu, M.; Guo, X.F.; Ananda, A.L.; Chan, M.C.; Peh, L.S. Using mobile phone barometer for low-power transportation context detection. In Proceedings of the 12th ACM Conference on Embedded Network Sensor Systems, Memphis, TN, USA, 3–6 November 2014; ACM: New York, NY, USA, 2014; pp. 191–205. [Google Scholar]
- Lan, G.; Xu, W.; Khalifa, S.; Hassan, M.; Hu, W. Transportation mode detection using kinetic energy harvesting wearables. In Proceedings of the 2016 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops), Sydney, Australia, 14–18 March 2016; pp. 1–4. [Google Scholar]
- Yu, M.C.; Yu, T.; Wang, S.C.; Lin, C.J.; Chang, E.Y. Big Data Small Footprint: The Design of a Low-power Classifier for Detecting Transportation Modes. Proc. VLDB Endow. 2014, 7, 1429–1440. [Google Scholar] [CrossRef]
- Fang, S.H.; Liao, H.H.; Fei, Y.X.; Chen, K.H.; Huang, J.W.; Lu, Y.D.; Tsao, Y. Transportation modes classification using sensors on smartphones. Sensors 2016, 16, 1324. [Google Scholar] [CrossRef] [PubMed]
- Reddy, S.; Mun, M.; Burke, J.; Estrin, D.; Hansen, M.; Srivastava, M. Using mobile phones to determine transportation modes. ACM Trans. Sens. Netw. (TOSN) 2010, 6, 13. [Google Scholar] [CrossRef]
- Lorintiu, O.; Vassilev, A. Transportation mode recognition based on smartphone embedded sensors for carbon footprint estimation. In Proceedings of the 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC), Rio de Janeiro, Brazil, 1–4 November 2016; pp. 1976–1981. [Google Scholar]
- de MS Quintella, C.A.; Andrade, L.C.; Campos, C.A.V. Detecting the transportation mode for context-aware systems using smartphones. In Proceedings of the 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC), Rio de Janeiro, Brazil, 1–4 November 2016; pp. 2261–2266. [Google Scholar]
- Byon, Y.J.; Liang, S. Real-time transportation mode detection using smartphones and artificial neural networks: Performance comparisons between smartphones and conventional global positioning system sensors. J. Intell. Transp. Syst. 2014, 18, 264–272. [Google Scholar] [CrossRef]
- Nitsche, P.; Widhalm, P.; Breuss, S.; Maurer, P. A strategy on how to utilize smartphones for automatically reconstructing trips in travel surveys. Procedia-Soc. Behav. Sci. 2012, 48, 1033–1046. [Google Scholar] [CrossRef]
- Figo, D.; Diniz, P.C.; Ferreira, D.R.; Cardoso, J.M. Preprocessing techniques for context recognition from accelerometer data. Pers. Ubiquitous Comput. 2010, 14, 645–662. [Google Scholar] [CrossRef]
- Nick, T.; Coersmeier, E.; Geldmacher, J.; Goetze, J. Classifying means of transportation using mobile sensor data. In Proceedings of the 2010 International Joint Conference on Neural Networks (IJCNN), Barcelona, Spain, 18–23 July 2010; pp. 1–6. [Google Scholar]
- Abdulazim, T.; Abdelgawad, H.; Habib, K.; Abdulhai, B. Using smartphones and sensor technologies to automate collection of travel data. Transp. Res. Rec. J. Transp. Res. Board 2013, 2383, 44–52. [Google Scholar] [CrossRef]
- Cardoso, N.; Madureira, J.; Pereira, N. Smartphone-based transport mode detection for elderly care. In Proceedings of the 2016 IEEE 18th International Conference on e-Health Networking, Applications and Services (Healthcom), Munich, Germany, 14–17 September 2016; pp. 1–6. [Google Scholar]
- Zhou, X.; Yu, W.; Sullivan, W.C. Making pervasive sensing possible: Effective travel mode sensing based on smartphones. Comput. Environ. Urban Syst. 2016, 58, 52–59. [Google Scholar] [CrossRef]
- Shafique, M.A.; Hato, E. Travel mode detection with varying smartphone data collection frequencies. Sensors 2016, 16, 716. [Google Scholar] [CrossRef] [PubMed]
- Das, R.D.; Winter, S. Detecting urban transport modes using a hybrid knowledge driven framework from GPS trajectory. ISPRS Int. J. Geo-Inf. 2016, 5, 207. [Google Scholar] [CrossRef]
- Bloch, A.; Erdin, R.; Meyer, S.; Keller, T.; de Spindler, A. Battery-Efficient Transportation Mode Detection on Mobile Devices. In Proceedings of the 2015 16th IEEE International Conference on Mobile Data Management (MDM), Pittsburgh, PA, USA, 15–18 June 2015; Volume 1, pp. 185–190. [Google Scholar]
- Su, X.; Caceres, H.; Tong, H.; He, Q. Online travel mode identification using smartphones with battery saving considerations. IEEE Trans. Intell. Transp. Syst. 2016, 17, 2921–2934. [Google Scholar] [CrossRef]
- Buber, E.; Guvensan, A.M. Discriminative time domain features for activity recognition on a mobile phone. In Proceedings of the 2014 IEEE Ninth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), Singapore, 21–24 April 2014; pp. 1–6. [Google Scholar]
- Hall, M.; Frank, E.; Holmes, G.; Pfahringer, B.; Reutemann, P.; Witten, I.H. The WEKA data mining software: an update. ACM SIGKDD Explor. Newsl. 2009, 11, 10–18. [Google Scholar] [CrossRef]
- Vu, T.H.; Dung, L.; Wang, J.C. Transportation Mode Detection on Mobile Devices Using Recurrent Nets. In Proceedings of the 2016 ACM on Multimedia Conference, Amsterdam, The Netherlands, 15–19 October 2016; ACM: New York, NY, USA, 2016; pp. 392–396. [Google Scholar]
Transport Mode | Number of Trips | Total Time (min) |
---|---|---|
Bus | 53 | 1186 |
Car | 27 | 500 |
Ferry | 15 | 179 |
Metro | 56 | 976 |
Train | 34 | 462 |
Tram | 25 | 677 |
Walking | 82 | 413 |
Stationary | 33 | 315 |
Features | |
---|---|
MinimumReduction | |
MaximumReduction | |
MinimumIncrease | |
MaximumIncrease | |
MinimumValue | |
MaximumValue | |
Range | |
ArithmeticMean | |
HarmonicMean | |
QuadraticMean | |
Mod | |
Median | |
Variance | |
StandardDeviation | |
Arithmetic Mean of Instant Exchange | |
Quadratic Mean of Instant Exchange | |
Covariance | |
Freq _above_median | |
Freq_below_median | |
Freq_between_median | |
Freq_above_mean | |
Freq_below_mean | |
Freq_between_mean | |
MaxConsecutive_above_median | |
MaxConsecutive_below_median | |
MaxConsecutive_between_median | |
MaxConsecutive_above_mean | |
MaxConsecutive_below_mean | |
MaxConsecutive_between_mean |
Classification Algorithm | Recall |
---|---|
Random Forest | 80.62% |
J48 | 72.22% |
k-NN | 70.04% |
Naive Bayes | 71.03% |
Transport Mode | Recall Rates Obtained by Using Only Common Features | Recall Rates Obtained by Whole Feature Set |
---|---|---|
Bus | 94.04 | 95.39 |
Car | 96.87 | 93.22 |
Ferry | 90.8 | 92.3 |
Metro | 63.58 | 68.33 |
Train | 78.94 | 89.47 |
Tram | 41.02 | 55.31 |
Overall | 78.17 | 82.80 |
Actual Class | Predicted Class | Ground Truth | Recall | |
---|---|---|---|---|
Pedestrian Activities | Vehicular Activities | |||
Pedestrian Activities | 269 | 46 | 315 | 85.4% |
Vehicular Activities | 80 | 2875 | 2955 | 97.3% |
Actual Class | Predicted Class | Ground Truth | Recall | |||||||
---|---|---|---|---|---|---|---|---|---|---|
Bus | Car | Ferry | Metro | Train | Tram | Walking | Stationary | |||
Bus | 143 | 12 | 5 | 1 | 6 | 4 | 6 | 6 | 183 | 78.1% |
Car | 6 | 226 | 11 | 1 | 2 | 1 | 4 | 3 | 254 | 90.0% |
Ferry | 20 | 2 | 59 | 2 | 0 | 2 | 0 | 6 | 91 | 64.8% |
Metro | 9 | 4 | 0 | 167 | 1 | 39 | 0 | 7 | 227 | 73.6% |
Train | 22 | 9 | 0 | 8 | 177 | 25 | 0 | 11 | 252 | 70.2% |
Tram | 8 | 6 | 0 | 19 | 7 | 108 | 0 | 5 | 153 | 70.6% |
Walking | 6 | 0 | 0 | 1 | 1 | 0 | 331 | 5 | 344 | 96.2% |
Stationary | 14 | 9 | 0 | 0 | 0 | 0 | 0 | 242 | 265 | 91.3% |
Precision | 62.7% | 84.3% | 78.6% | 83.9% | 91.2% | 60.3% | 97.1% | 84.9% |
Actual Class | Predicted Class | Ground Truth | Recall | |||||||
---|---|---|---|---|---|---|---|---|---|---|
Bus | Car | Ferry | Metro | Train | Tram | Walking | Stationary | |||
Bus | 162 | 15 | 0 | 0 | 0 | 0 | 6 | 0 | 183 | 88.5% |
Car | 0 | 250 | 0 | 0 | 0 | 0 | 3 | 0 | 254 | 98.4% |
Ferry | 9 | 0 | 82 | 0 | 0 | 0 | 0 | 0 | 91 | 90.1% |
Metro | 0 | 0 | 0 | 227 | 0 | 0 | 0 | 0 | 227 | 100% |
Train | 8 | 0 | 0 | 0 | 234 | 10 | 0 | 0 | 252 | 92.9% |
Tram | 0 | 0 | 0 | 8 | 0 | 145 | 0 | 0 | 153 | 94.8% |
Walking | 6 | 5 | 0 | 1 | 1 | 0 | 331 | 0 | 344 | 96.2% |
Stationary | 16 | 8 | 0 | 0 | 0 | 0 | 0 | 241 | 265 | 90.9% |
Precision | 80.6% | 89.9% | 100% | 96.1% | 99.6% | 93.5% | 97.4% | 100% |
Classification Algorithm | Recall |
---|---|
Fang et al. [26] | 83.57% |
Before Applying Healing Algorithm | 84.38% |
After Applying Healing Algorithm | 91.63% |
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Guvensan, M.A.; Dusun, B.; Can, B.; Turkmen, H.I. A Novel Segment-Based Approach for Improving Classification Performance of Transport Mode Detection. Sensors 2018, 18, 87. https://doi.org/10.3390/s18010087
Guvensan MA, Dusun B, Can B, Turkmen HI. A Novel Segment-Based Approach for Improving Classification Performance of Transport Mode Detection. Sensors. 2018; 18(1):87. https://doi.org/10.3390/s18010087
Chicago/Turabian StyleGuvensan, M. Amac, Burak Dusun, Baris Can, and H. Irem Turkmen. 2018. "A Novel Segment-Based Approach for Improving Classification Performance of Transport Mode Detection" Sensors 18, no. 1: 87. https://doi.org/10.3390/s18010087