# Lost Person Search Area Prediction Based on Regression and Transfer Learning Models

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

#### 1.1. Related Work

#### 1.2. Proposed Method

## 2. Methodology

#### 2.1. Description and Sources of Data

#### 2.2. Data Preprocessing

- id—unique identifier of the data sample
- LC id—land cover type identifier as described in Table 1
- DEM—value read from the digital elevation model file associated with the start point denoting elevation above sea level in meters,
- abs slope—absolute value of the slope tangent, calculated as a fraction of vertical elevation difference (in meters) and horizontal distance (in meters).
- dist wgs—distance length of the segment between two geographical points in World Geodetic System in meters,
- d from start—distance, i.e., position of the segment in the collection from the start of the GPX trail
- speed 2d kmh—average speed of walking on the segment by the particular user expressed in km/h

#### 2.3. Linear Regression Model

- land cover id value, determined as described in the previous section,
- terrain slope,
- distance length a segment,
- difference in elevation of end and start point,
- elevation above the sea level.

#### 2.4. Model Calibration with Transfer Learning

#### 2.5. Search Area Prediction

## 3. Results and Discussion

${q}_{0}$ | 4.786872732767515 |

${q}_{1}$ | 0.013315859975442301 |

${q}_{2}$ | 0.0019411657191748125 |

${q}_{3}$ | −16.319148163916193 |

${q}_{4}$ | −0.026739066247719285 |

${q}_{5}$ | 0.5717657455052271 |

and | |

$L{C}_{id}$ | land cover id value |

dem | elevation above sea level in m |

absslope | absolute value of segment slope tangent |

elev d | difference in elevation between and and start point |

dist | distance walked in m |

${q}_{0}$ | 1.1967181831918787 |

${q}_{1}$ | 0.0033289649938605752 |

${q}_{2}$ | 0.0004852914297937031 |

${q}_{3}$ | −4.079787040979048 |

${q}_{4}$ | −0.006684766561929821 |

${q}_{5}$ | 0.14294143637630677 |

## 4. Conclusions and Future Work

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Acknowledgments

## Conflicts of Interest

## Abbreviations

SAR | Search and Rescue |

IPP | initial planning point |

PLS | Point Last Seen |

ICT | Information and Communication Technologies |

DSS | Decision Support System |

CA | Cellular Automata |

ISRID | International Search and Rescue Incident Database |

GIS | Geographical Information System |

GPS | Global Positioning System |

GPX | GPS Exchange Format |

XML | eXchangable Markup Language |

CLC | Corine Land Cover |

DEM | Digital Elevation Model |

## References

- Croatian Mountain Rescue Service. CMRS—Croatian Mountain Rescue Service. Available online: http://www.hgss.hr/ (accessed on 15 December 2020).
- Lu, Y. Industry 4.0: A survey on technologies, applications and open research issues. J. Ind. Inf. Integr.
**2017**, 6, 1–10. [Google Scholar] [CrossRef] - Jansen, A. The Understanding of ICTs in Public Sector and Its Impact on Governance. In Electronic Government; Scholl, H.J., Janssen, M., Wimmer, M.A., Moe, C.E., Flak, L.S., Eds.; Springer: Berlin/Heidelberg, Germany, 2012; pp. 174–186. [Google Scholar]
- Zhou, X.; Cheng, L.; Zhang, F.; Yan, Z.; Ruan, X.; Min, K.; Li, M. Integrating island spatial information and integer optimization for locating maritime search and rescue bases: A case study in the south China sea. ISPRS Int. J. Geo-Inf.
**2019**, 8, 88. [Google Scholar] [CrossRef][Green Version] - Phillips, K.; Longden, M.J.; Vandergraff, B.; Smith, W.R.; Weber, D.C.; McIntosh, S.E.; Wheeler, A.R., III. Wilderness search strategy and tactics. Wilderness Environ. Med.
**2014**, 25, 166–176. [Google Scholar] [CrossRef] [PubMed] - Profound Journey. 13 Fascinating Facts about Lost Person Behaviour. Available online: https://profoundjourney.com/13-facts-lost-person-behaviour/ (accessed on 15 December 2020).
- Kelley, D.E. Mountain Search for the Lost Victim; David E Kelley: Montrose, CO, USA, 1973. [Google Scholar]
- Koester, R.J. Lost Person Behavior: A Search and Rescue; dbS Productions LLC: Charlottesville, VA, USA, 2008. [Google Scholar]
- Doke, J. Analysis of Search Incidents and Lost Person Behavior in Yosemite National Park. Ph.D. Thesis, University of Kansas, Lawrence, KS, USA, 2012. [Google Scholar]
- Sava, E.; Twardy, C.; Koester, R.; Sonwalkar, M. Evaluating Lost Person Behavior Models. Trans. GIS
**2016**, 20, 38–53. [Google Scholar] [CrossRef] - Wysokinski, M.; Marcjan, R.; Dajda, J. Decision support software for search & rescue operations. In Proceedings of the 18th Annual International Conference on Knowledge-Based and Intelligent Information and Engineering Systems (KES), Pomeranian Sci & Technol, Gdynia, Poland, 15–17 September 2014; Volume 35, pp. 776–785. [Google Scholar] [CrossRef][Green Version]
- Lin, L.; Goodrich, M.A. A Bayesian approach to modeling lost person behaviors based on terrain features in wilderness search and rescue. Comput. Math. Organ. Theory
**2010**, 16, 300–323. [Google Scholar] [CrossRef] - Keijsers, N.; Stolwijk, N.; Renzenbrink, G.; Duysens, J. Prediction of walking speed using single stance force or pressure measurements in healthy subjects. Gait Posture
**2016**, 43, 93–95. [Google Scholar] [CrossRef] [PubMed] - Feng, A.; Gordon, A.S. Latent terrain representations for trajectory prediction. In Proceedings of the 1st ACM SIGSPATIAL International Workshop on Computing with Multifaceted Movement Data, Chicago, IL, USA, 5 November 2019; pp. 1–4. [Google Scholar]
- Wang, L.; Geng, X.; Ma, X.; Liu, F.; Yang, Q. Cross-city transfer learning for deep spatio-temporal prediction. arXiv
**2018**, arXiv:1802.00386. [Google Scholar] - Torrey, L.; Shavlik, J. Transfer learning. In Handbook of Research on Machine Learning Applications and Trends: Algorithms, Methods, and Techniques; IGI Global: Hershey, PA, USA, 2010; pp. 242–264. [Google Scholar]
- Xinshun, L.; Xin, H.; Hui, M.; Jing, L.; Weizhong, L.; Qingwen, Y. Automatic Cross-Domain Transfer Learning for Linear Regression. arXiv
**2020**, arXiv:2005.04088. [Google Scholar] - Chopard, B.; Droz, M. Cellular Automata; Springer: Berlin/Heidelberg, Germany, 1998; Volume 1. [Google Scholar]
- Vasic, J.; Ruskin, H.J. Cellular automata simulation of traffic including cars and bicycles. Phys. A Stat. Mech. Its Appl.
**2012**, 391, 2720–2729. [Google Scholar] [CrossRef] - Yue, H.; Guan, H.; Zhang, J.; Shao, C. Study on bi-direction pedestrian flow using cellular automata simulation. Phys. A Stat. Mech. Its Appl.
**2010**, 389, 527–539. [Google Scholar] [CrossRef] - Hu, X.; Li, X.; Lu, L. Modeling the Land Use Change in an Arid Oasis Constrained by Water Resources and Environmental Policy Change Using Cellular Automata Models. Sustainability
**2018**, 10, 2878. [Google Scholar] [CrossRef][Green Version] - Bodrožić, L.; Stipaničev, D.; Šerić, M. Forest fires spread modeling using cellular automata approach. CEEPUS Summer Sch. Mod. Trends Control
**2006**, 23–33. [Google Scholar] - Han, T.; Zhao, J.; Li, W. Smart-Guided Pedestrian Emergency Evacuation in Slender-Shape Infrastructure with Digital Twin Simulations. Sustainability
**2020**, 12, 9701. [Google Scholar] [CrossRef] - Hashimoto, A.; Abaid, N. An Agent-Based Model of Lost Person Dynamics for Enabling Wilderness Search and Rescue. In Dynamic Systems and Control Conference; American Society of Mechanical Engineers: New York, NY, USA, 2019; Volume 59155. [Google Scholar]
- GPX the GPS Exchange Format. 2002. Available online: http://www.topografix.com/gpx.asp (accessed on 15 December 2020).
- Corine Land Cover (CLC) 2018 Version 2020 20u1 European Environment Agency. Available online: https://land.copernicus.eu/pan-european/corine-land-cover/clc2018 (accessed on 15 December 2020).
- European Digital Elevation Model (EU-DEM), Version 1.1. Available online: https://land.copernicus.eu/imagery-in-situ/eu-dem/eu-dem-v1.1/view (accessed on 15 December 2020).
- Ritter, N.; Ruth, M.; Grissom, B.B.; Galang, G.; Haller, J.; Stephenson, G.; Covington, S.; Nagy, T.; Moyers, J.; Stickley, J.; et al. Geotiff format specification geotiff revision 1.0. SPOT Image Corp
**2000**, 1. Available online: http://geotiff.maptools.org/spec/geotiffhome.html (accessed on 15 December 2020). - Van Brummelen, G. Heavenly Mathematics: The Forgotten Art of Spherical Trigonometry; Princeton University Press: Princeton, NJ, USA, 2012. [Google Scholar]
- Tobler, W. Three Presentations on Geographical Analysis and Modeling: Non-Isotropic Geographic Modeling Speculations on the Geometry of Geography Global Spatial Analysis; Technical Report; National Center for Geographic Information and Analysis: Santa Barbara, CA, USA, 1993; Volume 93. [Google Scholar]
- Freedman, D.A. Statistical Models: Theory and Practice; Cambridge University Press: Cambridge, UK, 2009. [Google Scholar]
- GDAL Development Team. GDAL—Geospatial Data Abstraction Library, Version 2.2.3; Open Source Geospatial Foundation: San Michele all’Adige, Italy, 2011. [Google Scholar]
- QGIS Development Team. QGIS Geographic Information System; Open Source Geospatial Foundation: San Michele all’Adige, Italy, 2009. [Google Scholar]
- Van Rossum, G.; Drake, F.L., Jr. Python Tutorial; Centrum voor Wiskunde en Informatica Amsterdam: Amsterdam, The Netherlands, 1995; Volume 620. [Google Scholar]
- Gillies, S.; Ward, B.; Petersen, A.S. Rasterio: Geospatial Raster I/O for Python Programmers. 2013. Available online: https://github.com/mapbox/rasterio (accessed on 15 December 2020).

**Figure 2.**Spatial distribution of the GPX files data. The GPX trails are collected from three Mountain rescue service departments—Split, Karlovac and Dubrovnik. The trails are collected in a wide area of the three cities.

**Figure 3.**An example of simulation and comparison of lost person end point and distance walked predicted with pre trained model.

**Figure 7.**Result of simulation with isochrones showing maximal reached area in every half an hour. IPP is taken from archive of SAR activities, and location where subject is found is denoted with red cross.

LC id | CLC Code | Label |
---|---|---|

22 | 244 | Agro-forestry areas |

23 | 311 | Broad-leaved forest |

24 | 312 | Coniferous forest |

25 | 313 | Mixed forest |

26 | 321 | Natural grasslands |

27 | 322 | Moors and heathland |

28 | 323 | Sclerophyllous vegetation |

29 | 324 | Transitional woodland-shrub |

id | LC id | DEM | abs Slope | Elev d | Dist Wgs | d from Start | Sped 2d kmh |
---|---|---|---|---|---|---|---|

0 | 29 | 68.39 | 0.000137 | 0.96 | 7.0102 | 0 | 2.80408 |

3 | 29 | 68.39 | 0.000203 | 0.96 | 4.721129 | 3 | 1.545097 |

7 | 29 | 67.91 | 0.000424 | 0.96 | 2.26302 | 7 | 0.581919 |

11 | 29 | 68.87 | −0.000168 | −0.96 | 5.731096 | 11 | 1.375463 |

Model | Test Score |
---|---|

Linear Regression | 0.412 |

Polynomial Regression | 0.454 |

Decision Tree Regressor | 0.647 |

Bayesian Ridge Regression | 0.412 |

Elastic Net Regression | 0.410 |

Ridge Regression | 0.401 |

Number of Isochrones within the Person Is Found | Number of Cases |
---|---|

1 | 9 |

2 | 6 |

3 | 1 |

outside the predicted area | 4 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 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

**MDPI and ACS Style**

Šerić, L.; Pinjušić, T.; Topić, K.; Blažević, T.
Lost Person Search Area Prediction Based on Regression and Transfer Learning Models. *ISPRS Int. J. Geo-Inf.* **2021**, *10*, 80.
https://doi.org/10.3390/ijgi10020080

**AMA Style**

Šerić L, Pinjušić T, Topić K, Blažević T.
Lost Person Search Area Prediction Based on Regression and Transfer Learning Models. *ISPRS International Journal of Geo-Information*. 2021; 10(2):80.
https://doi.org/10.3390/ijgi10020080

**Chicago/Turabian Style**

Šerić, Ljiljana, Tomas Pinjušić, Karlo Topić, and Tomislav Blažević.
2021. "Lost Person Search Area Prediction Based on Regression and Transfer Learning Models" *ISPRS International Journal of Geo-Information* 10, no. 2: 80.
https://doi.org/10.3390/ijgi10020080