Hybrid Random Forest and Support Vector Machine Classification for Benthic Habitat Mapping using Sentinel-2 Imagery
Huwaida Nur Salsabila (a*), Setiawan Djody Harahap (b), Abhista Fawwaz Sahitya (c)

(a) Department of Geography, Faculty of Mathematics and Natural Sciences, Universitas Negeri Makassar, Makassar, South Sulawesi 90224, Indonesia
(b) Master in Remote Sensing, Faculty of Geography, Universitas Gadjah Mada, Sekip Utara, Kab. Sleman, Daerah Istimewa Yogyakarta 55281, Indonesia
(c) Blue Carbon Research Group, Faculty of Geography, Universitas Gadjah Mada, Sekip Utara, Kab. Sleman, Daerah Istimewa Yogyakarta 55281, Indonesia

Abstract

Accurate benthic habitat mapping is essential for coastal management and ecosystem monitoring. However, remote sensing-based classification faces challenges due to the small size of benthic objects and their submerged nature, which increases the risk of misclassification. Machine learning algorithms such as Random Forest (RF) and Support Vector Machines (SVM) have been widely used to address these limitations, yet each has its drawbacks, RF may overfit. At the same time, SVM can misclassify and is sensitive to complex samples. This study proposes a hybrid classification method to overcome these limitations by fusing RF and SVM outputs within Google Earth Engine. The study area is located along the coast of Bontang City, East Kalimantan, Indonesia. Sentinel 2 imagery was classified using RF (ntree=50), SVM (gamma=10, cost=10), and a hybrid approach. The hybrid fusion rule applies RF and SVM agreement where available and resolves disagreement using a neighbourhood majority vote. Three benthic classes were mapped: coral/algae, seagrass, and bare substrate. RF yielded an overall accuracy of 0.785, while SVM reached 0.819. The hybrid method achieved the highest overall accuracy of 0.822, with producers and users accuracy outperforming both individual classifiers in most classes. Additional tests with varied RF and SVM parameters confirmed the robustness of the hybrid approach. Spatially, the hybrid classification reduced salt and pepper noise and improved coherence across benthic zones. These results demonstrate that the hybrid fusion method enhances benthic habitat mapping accuracy and offers a reliable solution for coastal monitoring applications.

Keywords: Benthic habitat- Remote sensing- Random forest- Support vector machine- Hybrid classification

Topic: Topic A: General Remote Sensing