Identification and differentiation of rice fields using semi-automatic in north Iran
Subject Areas :Ali Akbar Noroozi 1 , Mojtaba Saneie 2 , Zahra Rezghi 3
1 - Soil Conservation and Watershed Management Instituteو Tehran, Iran
2 - Soil Conservation and Watershed Management Institute
3 - Department of Soil Science,Tarbiat modares university, Tehran, Iran
Keywords: LANDSAT 8, NDVI, paddy field, remote sensing, LSWI,
Abstract :
As one of the world's major grains, rice has always been given special emphasis. Monitoring agricultural products plays a important role in providing global food supplies. In general, inorder to estimate product output, information such as area and production per unit area is important. Remote sensing technology can provide valuable and timely information on distribution, cropping, and production potential to decision makers in the agricultural sector. Detection by operator with field operation or visual detection cost too much, also associated with human errors. In addition, with regard to the possibility of changing land use, there is always the need for border control of rice paddies. So that the purpose of this study was to separate the rice fields with an automatic method in Gilan province. In recent years remote sensing has been raised as a valuable tool in the management and control of resources and agricultural products. By the same token using satellite images Landsat 8 and soil moisture index (LSWI) and leaves (LAI) and normalized differential vegetation index (NDVI) fully automated method is used to detect the rice paddies. According to this method, then, using the characteristics of the high water range in the early rice transplanting, these areas were identified using specific threshold. Then using another image in the harvest period, unwanted complications such as wetlands and permanent water and wet forests were removed. Final evaluation model with graphical borders of operator were taken from GEO Eye images. The results indicate conformity of 90% of the used method.
_||_