Sent2Agri System Based Crop Type Mapping in Yellow River Irrigation Area
(2020) Journal of Geodesy and Geoinformation Science — Vol. 3, n° 4, p. 110-117 (2020)
Files
Sent2AgriSystemBasedCropTypeMappinginYellowRiverIrrigationArea.pdf
Open Access - Adobe PDF
- 9.13 MB
Details
- Authors
- Abstract
- Ａｇｒｉｃｕｌｔｕｒａｌ ｍｏｎｉｔｏｒｉｎｇ ｉｓ ｅｓｓｅｎｔｉａｌ ｆｏｒ ａｄｅｑｕａｔｅ ｍａｎａｇｅｍｅｎｔ ｏｆ ｆｏｏｄ ｐｒｏｄｕｃｔｉｏｎ ａｎｄ ｄｉｓｔｒｉｂｕｔｉｏｎ. Ｃｒｏｐ ｌａｎｄ ａｎｄ ｃｒｏｐ ｔｙｐｅ ｃｌａｓｓｉｆｉｃａｔｉｏｎꎬ ｕｓｉｎｇ ｒｅｍｏｔｅ ｓｅｎｓｉｎｇ ｔｉｍｅ ｓｅｒｉｅｓꎬ ｆｏｒｍ ａｎ ｉｍｐｏｒｔａｎｔ ｔｏｏｌ ｔｏ ｃａｐｔｕｒｅ ｔｈｅ ａｇｒｉｃｕｌｔｕｒａｌ ｐｒｏｄｕｃｔｉｏｎ ｉｎｆｏｒｍａｔｉｏｎ. Ｔｈｅ ｒｅｃｅｎｔｌｙ ｌａｕｎｃｈｅｄ Ｓｅｎｔｉｎｅｌ￣２ ｓａｔｅｌｌｉｔｅｓ ｐｒｏｖｉｄｅ ｕｎｐｒｅｃｅｄｅｎｔｅｄ ｍｏｎｉｔｏｒｉｎｇ ｃａｐａｃｉｔｉｅｓ ｉｎ ｔｅｒｍｓ ｏｆ ｓｐａｔｉａｌ ｒｅｓｏｌｕｔｉｏｎꎬ ｓｗａｔｈ ｗｉｄｔｈꎬ ａｎｄ ｒｅｖｉｓｉｔ ｆｒｅｑｕｅｎｃｙ. Ｔｈｅ Ｓｅｎｔｉｎｅｌ￣２ ｆｏｒ Ａｇｒｉｃｕｌｔｕｒｅ (Ｓｅｎ２￣Ａｇｒｉ) ｓｙｓｔｅｍ ｈａｓ ｂｅｅｎ ｄｅｖｅｌｏｐｅｄ ｔｏ ｆｕｌｌｙ ｅｘｐｌｏｉｔ ｔｈｏｓｅ ｃａｐａｃｉｔｉｅｓꎬ ｂｙ ｐｒｏｖｉｄｉｎｇ ｆｏｕｒ ｒｅｌｅｖａｎｔ ｅａｒｔｈ ｏｂｓｅｒｖａｔｉｏｎ ｐｒｏｄｕｃｔｓ ｆｏｒ ａｇｒｉｃｕｌｔｕｒａｌ ｍｏｎｉｔｏｒｉｎｇ. Ｕｎｄｅｒ ｔｈｅ Ｄｒａｇｏｎ ４ Ｐｒｏｇｒａｍꎬ ｔｈｅ ｃｒｏｐ ｍａｐｐｉｎｇ ｗｉｔｈ ｖａｒｉｏｕｓ ｓａｔｅｌｌｉｔｅ ｉｍａｇｅｓ ａｎｄ ａ ｓｐｅｃｉｆｉｃ ｆｏｃｕｓ ｏｎ ｔｈｅ Ｙｅｌｌｏｗ Ｒｉｖｅｒ ｉｒｒｉｇａｔｅｄ ａｇｒｉｃｕｌｔｕｒａｌ ａｒｅａ ｉｎ ｔｈｅ Ｎｉｎｇｘｉａ Ｈｕｉ Ａｕｔｏｎｏｍｏｕｓ Ｒｅｇｉｏｎ ｉｎ Ｃｈｉｎａ ｗａｓ ｃａｒｒｉｅｄ ｏｕｔ ｗｉｔｈ ｔｈｅ Ｓｅｎｔｉｎｅｌ￣２ ｆｏｒ Ａｇｒｉｃｕｌｔｕｒｅ ｓｙｓｔｅｍ (Ｓｅｎｔ２Ａｇｒｉ). ９ ｔｙｐｅｓ ｏｆ ｃｒｏｐｓ ｗｅｒｅ ｃｌａｓｓｉｆｉｅｄ ａｎｄ ｔｈｅ ｃｒｏｐ ｔｙｐｅ ｍａｐ ｉｎ ２０１７ ｗａｓ ｐｒｏｄｕｃｅｄ ｂａｓｅｄ ｏｎ ３５ ｓｃｅｎｅｓ Ｓｅｎｔｉｎｅｌ ２Ａ/ Ｂ ｉｍａｇｅｓ. Ｔｈｅ ｏｖｅｒａｌｌ ａｃｃｕｒａｃｙ ｃｏｍｐｕｔｅｄ ｆｒｏｍ ｔｈｅ ｅｒｒｏｒ ｃｏｎｆｕｓｉｏｎ ｍａｔｒｉｘ ｉｓ ８８％ꎬ ｗｈｉｃｈ ｉｎｃｌｕｄｅｓ ｔｈｅ ｃｒｏｐｐｅｄ ａｎｄ ｕｎｃｒｏｐｐｅｄ ｔｙｐｅｓ. Ａｆｔｅｒ ｔｈｅ ｒｅｍｏｖａｌ ｏｆ ｔｈｅ ｕｎｃｒｏｐｐｅｄ ａｒｅａꎬ ｔｈｅ ｏｖｅｒａｌｌ ａｃｃｕｒａｃｙ ｆｏｒ ａ ｃｒｏｐｐｅｄ ｄｅｃｒｅａｓｅ ｔｏ ７３％. Ｉｎ ｏｒｄｅｒ ｔｏ ｆｕｒｔｈｅｒ ｉｍｐｒｏｖｅ ｔｈｅ ｃｒｏｐ ｃｌａｓｓｉｆｉｃａｔｉｏｎ ａｃｃｕｒａｃｙꎬ ｔｈｅ ｔｒａｉｎｉｎｇ ｄａｔａｓｅｔ ｓｈｏｕｌｄ ｂｅ ｆｕｒｔｈｅｒ ｉｍｐｒｏｖｅｄ ａｎｄ ｔｕｎｅｄ.
- Affiliations
Citations
Fan, J., Defourny, P., Dong, Q., ZHANG, X., De Vroey, M., Bellemans, N., XU, Q., Li, Q., Zhang, L., & Gao, H. (2020). Sent2Agri System Based Crop Type Mapping in Yellow River Irrigation Area. Journal of Geodesy and Geoinformation Science, 3(4), 110-117. https://doi.org/10.11947/j.JGGS.2020.0411 (Original work published 2020)