11-30, 13:30–13:50 (Asia/Seoul), Seoul Archive
A greenhouse is an important infrastructure for protecting out-of-season plants against extreme cold or hot weather. Remote sensing datasets are useful for identifying the greenhouse locations, measuring the greenhouse sizes and accessing the greenhouse directions. In this research, the greenhouses were detected from the Planetscope satellite imagery using the YOLO algorithm through the following steps. First, the training samples of the greenhouses were made using the Planetscope satellite imagery . Second, the YOLO model was trained using the generated training samples. Finally, the greenhouses were detected from the given Planetscope satellite image using the trained YOLO model. The trained YOLO model had the outstanding performance for detecting the greenhouse features from the given Planetscope satellite imagery with the medium spatial resolution (3m).
Acknowledgment: This work was carried out with the support of "Cooperative Research Program for Agriculture Science and Technology Development (Project No. PJ0162342022)" Rural Development Administration, Republic of Korea.
Dr. Yun-Jae Choung is the GIS & Remote Sensing Expert. He received his Ph.D. from the Ohio State University in 2014. He is the Chief Technology Officer (CTO) of Geo C&I Co., Ltd. He has joined the multiple research projects for developing the remote sensing and GIS application softwares for water resource, disaster prevention and agriculture. He can use ArcGIS, QGIS, SNAP, ENVI, Erdas Imagine and Python.