The Study On The Estimation Of The Crop Coverage Of Winter Wheat Based On Multi-Source Remote Sensing Data

Crop coverage is an important index of agricultural ecological remote sensing.It can reflect interception ability of the crop light, and the dynamic change of crop growth period, is the important parameters of crop growth status and biological yield and is also associated with the photosynthetic active radiation absorption of crops.It reflect crop photosynthesis and transpiration, is the result of environment’s comprehensive impact on crop growth, and is one of the condition monitoring by remote sensing ground test data.To estimate crop coverage has an important significance to guide agricultural production management and the evaluation of agricultural ecosystem.This paper studies around improving the winter wheat vegetation coverage estimation accuracy based on remote sensing data. Analyzing previous extraction method of wheat coverage based on digital photos feature, and find points which can be optimized, then add parameters and increase the filter way of improving the precision of wheat coverage extraction. Then, research results show that the short wave infrared, visible light and near infrared band have a certain relevance with vegetation coverage.Whether it that the combination the three important characteristics of the band to build a new vegetation index, can improve the effect of remote sensing estimation of vegetation coverage is surely worth a try, and at present this kind of research is still rare. Based on ground the hyperspectral data and measured wheat vegetation coverage data of the winter wheat of country of xiaotangshan, changping, Beijing in the 2013 and 2014, this paper will try to takeadvantage of the combination of short wave infrared water index and common vegetation index to create a new type of vegetation index to explore the possibility of improving the precision for remote sensing estimation of vegetation coverage. In addition, the study about the vegetation coverage is usually using regression model of single vegetation index,but because plant spectrum is under the influence of various factors such as the environment, atmosphere and vegetation reestablishment,vegetation index usually has the obvious regionality and timeliness.And mainly because all the vegetation indexes are saturated for vegetation coverage more or less,the best vegetation index determined is often unstable. Based on ground the hyperspectral data and measured wheat vegetation coverage data of the winter wheat of country of xiaotangshan, changping, Beijing in the 2014 and 2015 and remote sensing data of GF-1 in 2015, choose the typical vegetation index, and through the way of segment joint inversion attempt to determine the section points through the data analysis, and then select the best vegetation index of section, and finally with the method of joint inversion to estimate vegetation coverage of wheat aiming at improving the accuracy of estimating wheat coverage.Experiments show that the results of the research can improve the estimation accuracy of the winter wheat vegetation coverage, and also have a certain guiding role for other types of crops in the estimation of vegetation coverage.and provide new thought and reference for the inversion of vegetation coverage by remote sensing using vegetation index.