Determination of planting crops using satellite data at Shonai Plain in Japan

5th International Symposium on LAPAN-IPB Satellite for Food Security and Environmental Monitoring 2018

生态环境科学

Saito, Genya^1 ; Uto, Kuniaki^2 ; Yonezawa, Chinatsu^3 ; Sasaki, Yuka^4 ; Yoshino, Kunihiko^1 ; Oda, Kunio^5 ; Sato, Jun^6 ; Oyoshi, Kei^7 ; Mizukami, Yosei^7

Graduate School of Agricultural and Life Sciences, University of Tokyo, Tokyo, Japan^1

School of Computing, Tokyo Institute of Technology, Yokohama, Japan^2

Graduate School of Agricultural Science, Tohoku University, Sendai, Japan^3

Faculty of Agriculture, Yamagata University, Tsuruoka, Japan^4

Shonai Branch Office, Yamagata Prefecture, Tsuruoka, Japan^5

Tsuruoka College, National Institute of Technology, Tsuruoka, Japan^6

Earth Observation Research Center, Japan Aerospace Exploration Agency, Tsukuba, Japan^7

关键词: Agricultural fields;    Asian countries;    Crop classification;    Evaluation criteria;    Optical satellites;    Satellite data;    Satellite images;    Time-series data;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/284/1/012009/pdf
DOI  :  10.1088/1755-1315/284/1/012009

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

Shonai Plain is famous paddy area in Japan. As a countermeasure overproduction of rice in Japan, there are many other crops not rice in the paddy fields. We try to determination of planted crops using satellite data. Recently, in Asian countries, GIS is spread widely and agricultural field polygon is commonly used. Crop determination using the polygon is becoming popular. Normally, fields are used without changing, but sometimes, one field is divided two or three fields, and two or three fields are integrated one field. It is necessary to overlay satellite image and the polygon, and to check with fitting or not. In case of mismatch, field polygon must be corrected. Simultaneously with this confirmation work for each field polygon, the optical satellite data were classified as water surface 'water', vegetation as 'vegetation', bare soil as 'soil' and 'cannot determination'. In the SAR image, we classify 4 degrees, as small scattering of the water surface '0', scattering slightly larger than the water surface '1', larger than the water surface '2', and large scattering '3'. The crop determination is performed according to the following standard using to the time series data. (a) Rice: Almost water surface at rice planting time and following several weeks. In summer, there is large vegetation. (b) Wheat: There is large vegetation at rice planting time, and small vegetation at summer. (c) Soybean: There is small vegetation at rice planting time, and large vegetation in summer. Using this evaluation criteria, we performed the crop classifications in Shonai Pain by ALOS2 and Sentinel2 data, and the correct answer rates are more than 95% at both satellite data.