Mapping shallow waters habitats using OBIA by applying several approaches of depth invariant index in North Kepulauan Seribu

4th International Symposium on LAPAN-IPB Satellite for Food Security and Environmental Monitoring

农业科学;生态环境科学

Siregar, V.P.^1 ; Agus, S.B.^1 ; Subarno, T.^1 ; Prabowo, N.W.^1

Department of Marine Science and Technology, Bogor Agricultural University, Kampus IPB Darmaga, West Java, Bogor

16680, Indonesia^1

关键词: Field observations;    Habitat classification;    Influence of water;    Object-based classifications;    Remote sensing technology;    Satellite image classification;    Spatial resolution;    Water column correction;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/149/1/012052/pdf
DOI  :  10.1088/1755-1315/149/1/012052

来源: IOP

【 摘 要 】

The availability of satellite imagery with a variety of spatial resolution, both free access and commercial become as an option in utilizing the remote sensing technology. Variability of the water column is one of the factors affecting the interpretation results when mapping marine shallow waters. This study aimed to evaluate the influence of water column correction (depth-invariant index) on the accuracy of shallow water habitat classification results using OBIA. This study was conducted in North of Kepulauan Seribu, precisely in Harapan Island and its surrounding areas. Habitat class schemes were based on field observations, which were then used to build habitat classes on satellite imagery. The water column correction was applied to the three pairs of SPOT-7 multispectral bands, which were subsequently used in object-based classification. Satellite image classification was performed with four different approaches, namely (i) using DII transformed bands with single pair band input (B1B2), (ii) multi pairs bands (B1B2, B1B3, and B2B3), (iii) combination of multi pairs band and initial bands, and (iv) only using initial bands. The accuracy test results of the four inputs show the values of Overall Accuracy and Kappa Statistics, respectively 55.84 and 0.48; 68.53 and 0.64; 78.68 and 0.76; 77.66 and 0.74. It shows that the best results when using DII and initial band combination for shallow water benthic classification in this study site.