Extraction of inland Nypa fruticans (Nipa Palm) using Support Vector Machine

Biagtan, A. R.; Alberto, Ronaldo T.; Quibuyen, J. S.; Serrano, S. C.; Damian, Guiller. B.; Camaso, E. E.; Panuyas, N. Z.

doi:10.1117/12.2277522

Cited by 1 publication

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Utilization of very high-resolution images has become a new trend in forest management, particularly in the detection and identification of forest stand variables. Support Vector Machine (SVM) classification process was used to extract Nipa (Nypa fructicans) with a high accuracy (95%) in inland areas by using the available LiDAR data [17].…”

Section: Discussionmentioning

confidence: 99%

Assessments of Nipa Forest Using Landsat Imagery Enhanced with Unmanned Aerial Vehicle Photography

et al. 2020

View full text Add to dashboard Cite

Nipa palms are exposed by the transformation of land use and land cover changes (LULCC) due to changes to aquaculture and orchards. Modern remote sensing for environmental monitoring of LULCC has been made easier by the use of high spatial resolution images, innovative image processing and Geographic Information Systems (GIS). The expense of high-resolution satellite imagery has resulted in investigators moving to open sources (e.g., Landsat), therefore, the interpretation of images at a medium resolution can be classified simply as LULCC classes and are constrained by the detection of small-scale disturbances. This research applied Landsat imagery with very high-resolution imagery from Unmanned Aerial Vehicles (UAVs). In order to be useful for real-world applications, the accuracy of remote sensing data must be validated using proven ground-based methods. UAVs equipped with multispectral sensors were flown over the Nipa palms at the Prasae River, Rayong Province, Thailand. The main advantage of UAV-based remote sensing is that it reduces costs and immediate availability of high-resolution data. The UAV imagery was expensed as “drone truthing data” to train image classification algorithms. These results show that UAV data can be used effectively to support and categorize similar land-cover/use classes (aquaculture vs. mangrove forest vs. nipa forest) with consistently high identification of over 87.6% on the generated thematic map, where the mangrove forest detection rate was as high as 86%. For that reason, UAVs are engaged successively in management and conservation tasks, which can be used for regional or local scale studies to compare the achieved accuracy to a general regional land cover map. This approach can be used for the variability of plants to rectify land-cover classification. Therefore, UAV images are a very useful tool to fill the gap between remote sensing information and expensive ground field campaigns.

show abstract

Section: Discussionmentioning

confidence: 99%