Integrasi Lidar dan Citra Satelit Quickbird untuk Ekstraksi Bangunan Menggunakan Metode Klasifikasi Berbasis Objek
Keywords:
Segmentation, OBIA, Quickbird Imagery, NDSMAbstract
Pemanfaatan citra satelit resolusi tinggi dalam pemetaan kawasan perkotaan telah sangat banyak digunakan dalam berbagai aplikasi, salah satunya untuk ekstraksi bangunan. Terdapat dua pendekatan yang dapat digunakan untuk ekstraksi bangunan dengan menggunakan metode penginderaan jauh yaitu metode klasifikasi berbasis piksel dan metode klasifikasi berbasis objek. Akan tetapi metode klasifikasi berbasis piksel memiliki kelemahan yaitu mengabaikan aspek spasial sehingga dapat mengurangi akurasi dari hasil klasifikasi dan rentan terjadi gangguan salt and pepper yang berdampak pada hasil akurasi. Hal tersebut menyebabkan digunakan klasifikasi berbasis objek dengan metode segmentasi citra menggunakan algoritma multiresolusi pada penelitian ini. Metode segmentasi dengan algoritma multiresolusi memiliki keunggulan yaitu dapat menggabungkan informasi spektral dan spasial dimana objek geografis dipisahkan tidak hanya berdasarkan aspek spektral namun berdasarkan aspek spasial seperti ukuran, pola, dan tekstur. Variasi ketinggian bangunan dan kerapatan yang tinggi di wilayah perkotaan menjadi permasalahan yang ditemukan dalam melakukan ekstraksi bangunan. Untuk mengatasi permasalahan tersebut, dilakukan penambahan data ketinggian berupa data NDSM yang diturunkan dari data LiDAR. Lokasi penelitian terbagi menjadi 3 Area of Interest (AOI) dengan kriteria untuk setiap AOI ditentukan berdasarkan kerapatan bangunan. Hasil dari klasifikasi berbasis objek sebelum penambahan data NDSM memberikan hasil akurasi sebesar 58.16 % untuk AOI 1, 58.92 % untuk AOI 2, dan 71.43 % untuk AOI 3, sedangkan setelah penambahan data NDSM nilai akurasi yang diperoleh sebesar 69.35% untuk AOI 1, 81.90% untuk AOI 2, dan 97.37% untuk AOI 3. Hasil dari penelitian ini menunjukkan bahwa terdapat perbedaan hasil klasifikasi sebelum dan sesudah penambahan data NDSM. Penambahan data NDSM dalam proses klasifikasi berbasis objek untuk ekstraksi bangunan dapat membantu dalam memisahkan objek bangunan dan non-bangunan serta dapat meningkatkan akurasi dari hasil klasifikasi.
Utilization of high resolution satellite imagery for mapping in urban areas has been very widely used in various application, one of them for building extraction. Two common approaches for building extraction using remote sensing method are pixel-based classification method and object-based classification method. However, pixel-based classification method has the disadvantage of ignoring the spatial elements so as to reduce the accuracy of classification results also the susceptible of salt and pepper noise affects the accuracy of classification results. It is led to use object-based classification techniques with segmentation methods and multiresolution algorithm in this research. Segmentation methods with multiresolution algorithms has the advantage of being able to combine spectral and spatial information where geographic objects are distinguished not only on the spectral aspect but also spatial aspects such as shapes, patterns, and textures. Various heights of buildings and high density in the urban area become the next challenge in building extraction. In order to overcome these problems, the addition of height data NDSM derived from LiDAR data. The research location is divided into 3 Areas of Interest (AOI) with the criteria for each AOI being determined based on building density. The result of object-based classification before addition of NDSM data gives an accuracy result of 58.16% for AOI 1, 58.92 % for AOI 2, and 71.43 % for AOI 3, while after addition NDSM data the accuracy value obtained from the classification were 69.35 % in AOI 1, 81.90 % in AOI 2, and 97.37% in AOI 3. The result of classification shown the differences in classification results before and after the addition of NDSM data. Ancillary data NDSM in object based image classification for building extraction can distinguish between building and non-building class, also improve the accuracy of the classification results.
References
Baatz, M., & Schäpe, A. (2000). Multiresolution Segmentation: an optimization approach for high quality multi-scale image segmentation. Beutrage zum AGIT-Symposium. Vol. XII. hal.12–23.
Blaschke, T., Hay, G. J., Kelly, M., Lang, S., Hofmann, P., Addink, E., ... & Tiede, D. (2014). Geographic object-based image analysis–towards a new paradigm. ISPRS journal of photogrammetry and remote sensing. Vol.87. hal. 180-191: https://doi.org/10.1016/j.isprsjprs.2013.09.014
Benz, U. C., Hofmann, P., Willhauck, G., Lingenfelder, I., & Heynen, M. (2004). Multi-resolution , object-oriented fuzzy analysis of remote sensing data for GIS-ready information. ISPRS Journal of Photogrammetry and Remote Sensing. Vol.58. hal. 239–258: https://doi.org/10.1016/j.isprsjprs.2003.10.002
Burnett, C., & Blaschke, T. (2003). A multi-scale segmentation/object relationship modelling methodology for landscape analysis. Ecological Modelling. No.3. Vol.168. hal.233–249: https://doi.org/10.1016/S0304-3800(03)00139-X
Hay, G. J., Blaschke, T., Marceau, D. J., & Bouchard, A. (2003). A comparison of three image-object methods for the multiscale analysis of landscape structure. ISPRS Journal of Photogrammetry and Remote sensing. No.5-6. Vo.57. hal.327-345: https://doi.org/10.1016/S0924-2716(02)00162-4
Hay G.J., Castilla G. (2008) Geographic Object-Based Image Analysis (GEOBIA): A new name for a new discipline. In: Blaschke T., Lang S., Hay G.J. (eds) Object-Based Image Analysis. Lecture Notes in Geoinformation and Cartography. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-77058-9_4
Marpu, P. R., Neubert, M., Herold, H., & Niemeyer, I. (2010). Enhanced evaluation of image segmentation results. Journal of spatial science, 55(1), 55-68. https://doi.org/10.1080/14498596.2010.487850
Meng, X., Currit, N., Wang, L., & Yang, X. (2010). Object-Oriented Residential Building Land-Use Mapping Using LiDAR and Aerial Photographs. ASPRS Annual Conference, 1–17.
Niemeyer, I., & Canty, M. J. (2002). Object-Oriented Post-Classification of Change Images. Remote Sensing of Environment Monitoring. Vol. 4545. hal.100–108. International Society for Optics and Photonics. https://doi.org/10.1117/12.453661
Schirokauer, D., Yu, Q., Gong, P., Clinton, N., Biging, G., & Kelly, M. (2006). Object-based Detailed Vegetation Classification with Airborne High Spatial Resolution Remote Sensing Imagery. Photogrammetric Engineering & Remote Sensing. No.7. Vol.72. hal.799–811: https://doi.org/10.14358/PERS.72.7.799
Singh, M., Evans, D., Tan, B. S., & Nin, C. S. (2015). Mapping and characterizing selected canopy tree species at the Angkor world heritage site in Cambodia using aerial data. PLoS ONE. No.4. Vo.10. hal. 1–26: https://doi.org/10.1371/journal.pone.0121558
Sun, X., Shen, S., & Hu, Z. (2016). Automatic Building Extraction from Oblique Aerial Images. In 23rd International Conference on Pattern Recognition (ICPR) (pp. 2–7). Cancun. Dec 4-6.
Walker, J. S., & Briggs, J. M. (2007). An Object-oriented Approach to Urban Forest Mapping in Phoenix. Photogrammetric Engineering & Remote Sensing. No.5. Vol.73. hal. 577–583: https://doi.org/10.14358/PERS.73.5.577
Yongmin, K., Youkyung, H., Junho, Y., Dongyeob, H., & Yongil, K. (2012). Object-Based Classification and Building Extraction by Integrating Airborne LiDAR Data and Aerial Image. IEEE International Geoscience and Remote Sensing Symposium, 6200–6202. Munich, Germany. July 22-27.
Zhang, J., & Jia, L. (2014, June). A comparison of pixel-based and object-based land cover classification methods in an arid/semi-arid environment of northwestern China. In 2014 Third International Workshop on Earth Observation and Remote Sensing Applications (EORSA) (pp. 403-407). IEEE. Changsha, China. June 11-14.
Zou, X., Zhao, G., Li, J., Yang, Y., & Fang, Y. (2016). Object Based Image Analysis Combining High Spatial Resolution Imagery and Laser Point Clouds for Urban Land Cover. ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol.XLI-B3(July). hal.733–739: https://doi.org/10.5194/isprsarchives-XLI-B3-733-2016
