아리랑 2호 영상을 이용한 당진군 벼 재배면적 추정

Abstract

The purpose of this study is to estimate rice acreage and forecast agricultural outlook by using satellite images to forecast agricultural outlook. Rice was chosen because the acreage of the crop can easily be identified by using remote sensing technology. Dangjin, a rural county with a large rice acreage, was chosen as the sample region. Kompsat-2 satellite images were used since they were inexpensive to obtain. The method used to classify satellite images is based on three key concepts: segmentation, object-based classification, and accuracy verification. We found that the accuracy of analyzing satellite images was not different from the accuracy of on-site inspections. The accuracy of identifying rice paddy fields was very high at 90.8% on average. This study showed that remote sensing technology can actually be used to estimate rice acreage and forecast agricultural outlook. However, since it is difficult to use high-resolution satellite images because they are not readily available, and thus expensive to process, it is necessary to consider using medium-resolution satellite images.

The purpose of this study is to estimate rice acreage and forecast agricultural outlook by using satellite images to forecast agricultural outlook. Rice was chosen because the acreage of the crop can easily be identified by using remote sensing technology. Dangjin, a rural county with a large rice acreage, was chosen as the sample region. Kompsat-2 satellite images were used since they were inexpensive to obtain. The method used to classify satellite images is based on three key concepts: segmentation, object-based classification, and accuracy verification. We found that the accuracy of analyzing satellite images was not different from the accuracy of on-site inspections. The accuracy of identifying rice paddy fields was very high at 90.8% on average. This study showed that remote sensing technology can actually be used to estimate rice acreage and forecast agricultural outlook. However, since it is difficult to use high-resolution satellite images because they are not readily available, and thus expensive to process, it is necessary to consider using medium-resolution satellite images.