Evaluation of segmentation parameters in OBIA for classification of land covers from UAV images

[EN] Unmanned Aerial Vehicles (UAVs) have given a new boost to remote sensing and image classification techniques due to the high level of detail among other factors. Object-based image analysis (OBIA) could improve classification accuracy unlike to pixel-based, especially in high-resolution images. OBIA application for image classification consists of three stages i.e., segmentation, class definition and training polygons, and classification. However, defining the parameters: spatial radius (SR), range radius (RR) and minimum region size (MR) is necessary during the segmentation stage. Despite their relevance, they are usually visually adjusted, which leads to a subjective interpretation. Therefore, it is of utmost importance to generate knowledge focused on evaluating combinations of these parameters. This study describes the use of the mean-shift segmentation algorithm followed by Random Forest classifier using Orfeo Toolbox software. It was considered a multispectral orthomosaic derived from UAV to generate a suburban map land cover in town of El Pueblito, Durango, Mexico. The main aim was to evaluate efficiency and segmentation quality of nine parameter combinations previously reported in scientific studies.This in terms of number generated polygons, processing time, discrepancy measures for segmentation and classification accuracy metrics. Results evidenced the importance of calibrating the input parameters in the segmentation algorithms. Best combination was RE=5, RR=7 and TMR=250, with a Kappa index of 0.90 and shortest processing time. On the other hand, RR showed a strong and inversely proportional degree of association regarding the classification accuracy metrics.[ES] Los Vehículos Aéreos No Tripulados (VANT) han otorgado un nuevo auge a la teledetección y a las técnicas d clasificación de imágenes debido al alto nivel de detalle entre otros factores. El análisis de imágenes basado en objetos (OBIA) puede mejorar la precisión en la clasificación a diferencia de la basada en píxeles, especialmente en imágenes de alta resolución. La aplicación de OBIA para la clasificación de imágenes consta de tres etapas i.e., segmentación, definición de clases y polígonos de entrenamiento y clasificación. No obstante, en la etapa de segmentación es necesario definir los parámetros: radio espacial (RE), radio de rango (RR) y tamaño mínimo de la región (TMR). Los cuales, pese a su relevancia, suelen ser ajustados de manera visual, lo que conlleva a una interpretación subjetiva. Por lo anterior, es de suma importancia generar conocimiento enfocado a evaluar las combinaciones de estos parámetros. Este estudio describe el uso del algoritmo de segmentación de desplazamiento medio, seguido del clasificador Random Forest mediante el software Orfeo Toolbox. Se consideró un ortomosaico multiespectral derivado de VANT para generar un mapa de cobertura de suelo sub-urbano en la localidad El Pueblito, Durango, México. El objetivo principal fue evaluar la eficiencia y calidad de segmentación de nueve combinaciones de parámetros anteriormente reportadas en estudios científicos. Ello en términos de número de polígonos generados, tiempo de procesamiento, medidas de discrepancia de segmentación y métricas de precisión de la clasificación. Los resultados obtenidos lograron evidenciar la importancia de ajustar los parámetros de entrada en los algoritmos de segmentación. La mejor combinación fue RE=5, RR=7 y TMR=250, con un índice de Kappa de 0,90 y el menor tiempo de procesamiento. Por otro lado, el RR presentó un grado de asociación fuerte e inversamente proporcional con las métricas de precisión de clasificación.Se agradece al Consejo Nacional de Ciencia y Tecnología (Conacyt) por el financiamiento otorgado a la primera autora para la realización de sus estudios de maestría, así como al programa de Maestría en Geomática Aplicada a Recursos Forestales y Ambientales de la Facultad de Ciencias Forestales de la UJED.Hinojosa-Espinoza, SI.; Gallardo-Salazar, JL.; Hinojosa-Espinoza, FJC.; Meléndez-Soto, A. (2021). Evaluación de parámetros de segmentación en OBIA para la clasificación de coberturas del suelo a partir de imágenes VANT. Revista de Teledetección. 0(58):89-103. https://doi.org/10.4995/raet.2021.1478289103058Abburu, S., Golla, S.B. 2015. Satellite image classification methods and techniques: A review. International Journal of Computer Applications, 119(8), 20-25. https://doi.org/10.5120/21088-3779Adelabu, S., Mutanga, O., Adam, E. 2015. Testing the reliability and stability of the internal accuracy assessment of random forest for classifying tree defoliation levels using different validation methods. Geocarto International, 30(7), 810-821. https://doi.org/10.1080/10106049.2014.997303Al-Najjar, H.A.H., Kalantar, B., Pradhan, B., Saeidi, V., Halin, A.A., Ueda, N., Mansor, S. 2019. Land Cover Classification from fused DSM and UAV Images Using Convolutional Neural Networks. Remote Sensing, 11(12), 1461. https://doi.org/10.3390/rs11121461Apriyanto, D., Jaya, I.N., Puspaningsih, N. 2019. Examining the object-based and pixel-based image analyses for developing stand volume estimator model. Indonesian Journal of Electrical Engineering and Computer Science, 15(3), 1586-1596. https://doi.org/10.11591/ijeecs.v15.i3.pp1586-1596Ballari, D., Orellana, D., Acosta, E., Espinoza, A., Morocho, V. 2016. UAV monitoring for environmental management in Galapagos Islands. International Archives of the Photogrammetry, Remote Sensing & Spatial Information Sciences, 41, 1105-1111. https://doi.org/10.5194/isprsarchives- XLI-B1-1105-2016Belgiu, M., Drǎguţ, L., Strobl, J. 2014. Quantitative evaluation of variations in rule-based classifications of land cover in urban neighbourhoods using WorldView-2 imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 87, 205-215. https://doi.org/10.1016/j.isprsjprs.2013.11.007Benarchid, O., Raissouni, N. 2014. Mean-shift Segmentation Parameters Estimator (MSPE): A new tool for Very High Spatial Resolution satellite images. International Conference on Multimedia Computing and Systems -Proceedings, 357-361. https://doi.org/10.1109/ICMCS.2014.6911184Blaschke, T. 2010. Object based image analysis for remote sensing. ISPRS Journal of Photogrammetry and Remote Sensing, 65(1), 2-16. https://doi.org/10.1016/j.isprsjprs.2009.06.004Brooke, C., Clutterbuck, B. 2020. Mapping Heterogeneous Buried Archaeological Features Using Multisensor Data from Unmanned Aerial Vehicles. Remote Sensing, 12(1), 41. https://doi.org/10.3390/rs12010041Burdziakowski, P. 2017. Evaluation of Open Drone Map Toolkit for Geodetic Grade Aerial Drone Mapping- Case Study. En: Proceedings International Multidisciplinary Scientific GeoConference-SGEM 2017, Gdańska, Polonia. 29 Junio-5 Julio. pp. 101-110. https://doi.org/10.5593/sgem2017/23/S10.013Carvajal-Ramírez, F., Marques da Silva, J.R., Agüera- Vega, F., Martínez-Carricondo, P., Serrano, J., Moral, F.J. 2019. Evaluation of Fire Severity Indices Based on Pre- and Post-Fire Multispectral Imagery Sensed from UAV. Remote Sensing, 11(9), 993. https://doi.org/10.3390/rs11090993Chuvieco, E. 2020. Revisión histórica y perspectivas de futuro de la Teledetección: desde el ERTS hasta los Sentinels. Mapping, 29(200), 30-32.Colomina, I., Molina, P. 2014. Unmanned aerial systems for photogrammetry and remote sensing: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 92(2014), 79-97. https://doi.org/10.1016/j.isprsjprs.2014.02.013Comert, R., Avdan, U., Gorum, T., Nefeslioglu, H.A. 2019. Mapping of shallow landslides with object- based image analysis from unmanned aerial vehicle data. Engineering Geology, 260(2019), 105264. https://doi.org/10.1016/j.enggeo.2019.105264Congalton, R.G., Mead, R.A. 1983. A quantitative method to test for consistency and correctness in photointerpretation. Photogrammetric Engineering and Remote Sensing, 49(1), 69-74.Congalton, R.G., Green, K. 2009. Assessing the Accuracy of Remotely Sensed Data: Principles and Practices; Second CRC Press Taylor & Francis Group: Boca Raton, FL, USA, Volume 48. https://doi.org/10.1201/9781420055139Cutler, D.R., Edwards Jr, T.C., Beard, K.H., Cutler, A., Hess, K.T., Gibson, J., Lawler, J.J. 2007. Random forests for classification in ecology. Ecology, 88(11), 2783-2792. https://doi.org/10.1890/07-0539.1Dash, J.P., Pearse, G.D., Watt, M.S. 2018. UAV Multispectral Imagery Can Complement Satellite Data for Monitoring Forest Health. Remote Sensing, 10(8), 1216. https://doi.org/10.3390/rs10081216Dash, J.P., Watt, M.S., Pearse, G.D., Heaphy, M., Dungey, H.S. 2017. Assessing very high resolution UAV imagery for monitoring forest health during a simulated disease outbreak. ISPRS Journal of Photogrammetry and Remote Sensing, 131(2017), 1-14. https://doi.org/10.1016/j.isprsjprs.2017.07.007De Castro, A.I., Torres-Sánchez, J., Peña, J.M., Jiménez- Brenes, F.M., Csillik, O., López-Granados, F. 2018. An Automatic Random Forest-OBIA Algorithm for Early Weed Mapping between and within Crop Rows Using UAV Imagery. Remote Sensing, 10(2), 285. https://doi.org/10.3390/rs10020285De Luca, G., N. Silva, J.M., Cerasoli, S., Araújo, J., Campos, J., Di Fazio, S., Modica, G. 2019. Object-Based Land Cover Classification of Cork Oak Woodlands using UAV Imagery and Orfeo ToolBox. Remote Sensing, 11(10), 1238. https://doi.org/10.3390/rs11101238Dhingra, S., Kumar, D. 2019. A review of remotely sensed satellite image classification. International Journal of Electrical & Computer Engineering, 9(3), 1720-1731. https://doi.org/10.11591/ijece.v9i3.pp1720-1731Dongping M., Tianyu C., Hongyue., Longxiang L., Cheng Q., Jinyang D., 2012. Semivariogram-Based Spatial Bandwidth Selection for Remote Sensing Image Segmentation With Mean-Shift Algorithm. IEEE Geoscience and Remote Sensing Letters, 9(5), 813-817. https://doi.org/10.1109/lgrs.2011.2182604Enderle, D.I.M., Weih Jr, R.C. 2005. Integrating supervised and unsupervised classification methods to develop a more accurate land cover classification. Journal of the Arkansas Academy of Science, 59(1), 65-73.Farfaglia, S., Lollino, G., Iaquinta, M., Sale, I., Catella, P., Martino, M., Chiesa, S. 2015. The use of UAV to monitor and manage the territory: perspectives from the SMAT project. En Engineering Geology for Society and Territory- Volume 5 (pp. 691-695). Springer, Cham. https://doi.org/10.1007/978-3-319-09048-1_134Feng, Q., Liu, J., Gong, J. 2015. UAV Remote Sensing for Urban Vegetation Mapping Using Random Forest and Texture Analysis. Remote Sensing, 7(1), 1074-1094. https://doi.org/10.3390/rs70101074Gallardo-Salazar, J., Pompa-García, M., Aguirre- Salado, C., López-Serrano, P., Meléndez-Soto, A. 2020. Drones: tecnología con futuro promisorio en la gestión forestal. Revista Mexicana de Ciencias Forestales, 11(61), 28-50. https://doi.org/10.29298/ rmcf.v11i61.794Gao, J., Liao, W., Nuyttens, D., Lootens, P., Vangeyte, J., Pižurica, A., He, Y., Pieters, J.G. 2018. Fusion of pixel and object-based features for weed mapping using unmanned aerial vehicle imagery. International Journal of Applied Earth Observation and Geoinformation, 67(2018), 43-53. https://doi. org/10.1016/j.jag.2017.12.012Geneletti, D., Gorte, B.G.H. 2003. A method for object- oriented land cover classification combining Landsat TM data and aerial photographs. International Journal of Remote Sensing, 24(6), 1273-1286. https://doi.org/10.1080/01431160210144499Hasmadi, M., Pakhriazad, H.Z., Shahrin, M.F. 2009. Evaluating supervised and unsupervised techniques for land cover mapping using remote sensing data. Geografia: Malaysian Journal of Society and Space, 5(1), 1-10.Hossain, M.D., Chen, D. 2019. Segmentation for Object-Based Image Analysis (OBIA): A review of algorithms and challenges from remote sensing perspective. ISPRS Journal of Photogrammetry and Remote Sensing, 150(2019), 115-134. https://doi. org/10.1016/j.isprsjprs.2019.02.009Immitzer, M., Vuolo, F., Atzberger, C. 2016. First Experience with Sentinel-2 Data for Crop and Tree Species Classifications in Central Europe. Remote Sensing, 8(3), 166. https://doi.org/10.3390/ rs8030166Inzunza-López, J.O., López-Ariza, B., Valdez-Cepeda, R.D., Mendoza, B., Sánchez-Cohen, I., García- Herrera, G. 2011. La variación de las temperaturas extremas en la 'Comarca Lagunera' y cercanías. Revista Chapingo Serie Ciencias Forestales y del Ambiente, 17(2011), 45-61. https://doi.org/10.5154/r. rchscfa.2010.09.071Jain, M., Tomar, P.S. 2013. Review of image classification methods and techniques. International Journal of Engineering Research and Technology, 2(8), 852-858.Jara, C., Delegido, J., Ayala, J., Lozano, P., Armas, A., Flores, V. 2019. Estudio de bofedales en los Andes ecuatorianos a través de la comparación de imágenes Landsat-8 y Sentinel-2. Revista de Teledetección, 53(2019), 45-57. https://doi.org/10.4995/ raet.2019.11715Kakooei, M., Baleghi, Y. 2017. Fusion of satellite, aircraft, and UAV data for automatic disaster damage assessment. International Journal of Remote Sensing, 38(8-10), 2511-2534. https://doi.org/10.1080/01431161.2017.1294780Khatami, R., Mountrakis, G., Stehman, S.V. 2016. A meta-analysis of remote sensing research on supervised pixel-based land-cover image classification processes: General guidelines for practitioners and future research. Remote Sensing of Environment, 177(2016), 89-100. https://doi.org/10.1016/j.rse.2016.02.028Langhammer, J., Vacková, T. 2018. Detection and Mapping of the Geomorphic Effects of Flooding Using UAV Photogrammetry. Pure and Applied Geophysics, 175(9), 3223-3245. https://doi.org/10.1007/s00024-018-1874-1Li, M., Ma, L., Blaschke, T., Cheng, L., Tiede, D. 2016. A systematic comparison of different object- based classification techniques using high spatial resolution imagery in agricultural environments. International Journal of Applied Earth Observation and Geoinformation, 49(2016), 87-98. https://doi. org/10.1016/j.jag.2016.01.011Li, M., Zang, S., Zhang, B., Li, S., Wu, C. 2014. AReview of Remote Sensing Image Classification Techniques: the Role of Spatio-contextual Information. European Journal of Remote Sensing, 47(1), 389-411. https:// doi.org/10.5721/EuJRS20144723Li, S., Tang, H., Huang, X., Mao, T., Niu, X. 2017. Automated Detection of Buildings from Heterogeneous VHR Satellite Images for Rapid Response to Natural Disasters. Remote Sensing, 9(11), 1177. https://doi.org/10.3390/rs9111177Liu, Y., Biana, L., Menga, Y., Wanga, H., Zhanga, S., Yanga, Y., Shaoa, X., Wang, B., 2012. Discrepancy measures for selecting optimal combination of parameter values in object-based image analysis. ISPRS Journal of Photogrammetry and Remote Sensing, 68(2012), 144-156. https://doi.org/10.1016/j.isprsjprs.2012.01.007Lu, D., Weng, Q. 2007. A survey of image classification methods and techniques for improving classification performance. International Journal of Remote Sensing, 28(5), 823-870. https://doi.org/10.1080/01431160600746456Lyons, M.B., Keith, D.A., Phinn, S.R., Mason, T.J., Elith, J. 2018. A comparison of resampling methods for remote sensing classification and accuracy assessment. Remote Sensing of Environment, 208(2018), 145-153. https://doi.org/10.1016/j. rse.2018.02.026Ma, L., Cheng, L., Li, M., Liu, Y., Ma, X. 2015. Training set size, scale, and features in Geographic Object-Based Image Analysis of very high resolution unmanned aerial vehicle imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 102(2015), 14-27. https://doi.org/10.1016/j. isprsjprs.2014.12.026Ma, L., Li, M., Ma, X., Cheng, L., Du, P., Liu, Y. 2017. A review of supervised object-based land-cover image classification. ISPRS Journal of Photogrammetry and Remote Sensing, 130(2017), 277-293. https://doi.org/10.1016/j.isprsjprs.2017.06.001Mafanya, M., Tsele, P., Botai, J., Manyama, P., Swart, B., Monate, T. 2017. Evaluating pixel and object based image classification techniques for mapping plant invasions from UAV derived aerial imagery: Harrisia pomanensis as a case study. ISPRS Journal of Photogrammetry and Remote Sensing, 129(2017), 1-11. https://doi.org/10.1016/j.isprsjprs.2017.04.009Mangiameli, M., Mussumeci, G., Candiano, A. 2018. A low cost methodology for multispectral image classification. En Computational Science and Its Applications-ICCSA 2018 (pp. 263-280). Springer, Cham. https://doi.org/10.1007/978-3-319-95174- 4_22Matese, A., Toscano, P., Di Gennaro, S.F., Genesio, L., Vaccari, F.P., Primicerio, J., Belli, C., Zaldei, A., Bianconi, R., Gioli, B. 2015. Intercomparison of UAV, Aircraft and Satellite Remote Sensing Platforms for Precision Viticulture. Remote Sensing, 7(3), 2971-2990. https://doi.org/10.3390/ rs70302971Michel, J., Youssefi, D., Grizonnet, M. 2015. Stable Mean-Shift Algorithm and Its Application to the Segmentation of Arbitrarily Large Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing, 53(2), 952-964. https://doi. org/10.1109/TGRS.2014.2330857Monserud, R.A., Leemans, R. 1992. Comparing global vegetation maps with the Kappa statistic. Ecological Modelling, 62(4), 275-293. https://doi.org/10.1016/0304-3800(92)90003-WNenmaoui, A., Torres, M.Á.A., Novelli, A., Marín, M.C.V., Torres, F.J.A., Betlej, M., Cichón, P. 2017. Mapeado de invernaderos mediante teledetección orientada a objetos: relación entre la calidad de la segmentación y precisión de la clasificación. Revista Mapping, 26(181), 4-13. ISSN: 1131-9100Raissouni, N., Benarchid, O., Sobrino, J., Ayyan, A. 2015. Aplicación del Estimador de Parámetros de Segmentación por Media-desplazada (EPSM) a las imágenes de satélite de muy alta resolución espacial: Tetuán (Marruecos). Revista de Teledetección, 43(2015), 91-96. https://doi.org/10.4995/ raet.2015.3511Ramadhan Kete, S.C., Suprihatin, Tarigan, S.D., Effendi, H. 2019. Land use classification based on object and pixel using Landsat 8 OLI in Kendari City, Southeast Sulawesi Province, Indonesia. IOP Conference Series: Earth and Environmental Science, 284(2019), 012019. https://doi.org/10.1088/1755-1315/284/1/012019Rosenfield, G.H., Fitzpatrick-Lins, K. 1986. A coefficient of agreement as a measure of thematic classification accuracy. Photogrammetric Engineering and Remote Sensing, 52(2), 223-227.Sideris, K., Colson, D., Lightfoot, P., Heeley, L., Robinson, P. 2020. Review of image segmentation algorithms for analysing Sentinel-2 data over large geographical areas. JNCC Report No. 655. Peterborough, ISSN 0963-8091.Silalahi, R., Jaya, I.N., Tiryana, T., Mulia, F. 2018. Assessing the Crown Closure of Nypa on UAV Images using Mean-Shift Segmentation Algorithm. Indonesian Journal of Electrical Engineering and Computer Science, 9(3), 722-730. https://doi.org/10.11591/ijeecs.v9.i3.pp722-730Smits, P.C., Dellepiane, S.G., Schowengerdt, R.A. 1999. Quality assessment of image classification algorithms for land-cover mapping: A review and a proposal for a cost-based approach. International Journal of Remote Sensing, 20(8), 1461-1486. https://doi.org/10.1080/014311699212560Teodoro, A.C., Araujo, R. 2014. Exploration of the OBIA methods available in SPRING non- commercial software to UAV data processing. En: Proceedings SPIE 9245, Earth Resources and Environmental Remote Sensing/GIS Applications. https://doi.org/10.1117/12.2066468V. Amsterdam, Netherlands, 10 de Octubre. https://doi.org/10.1117/12.2066468Teodoro, A.C., Araujo, R. 2016. Comparison of performance of object-based image analysis techniques available in open source software (Spring and Orfeo Toolbox/Monteverdi) considering very high spatial resolution data. Journal of Applied Remote Sensing, 10(1), 1-22. https://doi.org/10.1117/1.JRS.10.016011Torres-Sánchez, J., López-Granados, F., Peña, J.M. 2015. An automatic object-based method for optimal thresholding in UAV images: Application for vegetation detection in herbaceous crops. Computers and Electronics in Agriculture, 114(2015), 43-52. https://doi.org/10.1016/j.compag.2015.03.019Trisasongko, B.H., Panuju, D.R., Paull, D.J., Jia, X., Griffin, A.L. 2017. Comparing six pixel-wise classifiers for tropical rural land cover mapping using four forms of fully polarimetric SAR data. International Journal of Remote Sensing, 38(11), 3274-3293. https://doi.org/10.1080/01431161.2017.1292072Villanueva Díaz, J., Stahle, D.W., Cerano Paredes, J., Estrada Ávalos, J., Constante García, V. 2013. Respuesta hidrológica del sabino en bosques de galería del Río San Pedro Mezquital, Durango. Revista Mexicana de Ciencias Forestales, 4(20), 9-19. https://doi.org/10.29298/rmcf.v4i20.366Vu, T.T. 2012. Object-based remote sensing image analysis with OSGeo tools. En: Proceedings FOSS4G Southeast Asia 2012, Johor Bahru, Malaysia. 18-19 Julio. pp. 79-84.Ye, S., Pontius, R.G., Rakshit, R. 2018. A review of accuracy assessment for object-based image analysis: From per-pixel to per-polygon approaches. ISPRS Journal of Photogrammetry and Remote Sensing, 141(2018), 137-147. https://doi.org/10.1016/j. isprsjprs.2018.04.002Zaraza-Aguilera, M.A., Manrique-Chacón, L.M. 2019. Generación de datos de cambio de coberturas vegetales en la sabana de Bogotá mediante el uso de series temporales con imágenes Landsat e imágenes sintéticas MODIS-Landsat entre los años 2007 y 2013. Revista de Teledetección, 54(2019), 41-58.https://doi.org/10.4995/raet.2019.1228