Inversion Of Leaf Chlorophyll Content Of Ratoon Rice Based On Near-surface Remote Sensing

As one of the rice planting modes,ratoon rice is of great significance to improve the contradiction of cultivated land and increase grain yield.Chlorophyll content is one of the important indicators to evaluate crop growth,which is of great significance to guide crop growth management.Near-surface remote sensing has the advantages of fast and non-destructive information acquisition,and is an effective way to achieve crop chlorophyll content inversion.However,few studies pay attention to the inversion of chlorophyll content of ratoon rice,and few scholars compare and analyze the inversion results of chlorophyll content from different near-surface remote sensing.Based on this,this paper takes the leaf chlorophyll content of ratoon rice as the research object,and obtains leaf scale hyperspectral,canopy scale hyperspectral and UAV imaging multispectral images of ratoon rice.Based on the measured near-surface remote sensing data and leaf chlorophyll content,this paper analyzes the spectral response law of ratoon rice at leaf and canopy scales,builds physical models and empirical models of leaf chlorophyll content of ratoon rice,and explores different data enhancement methods(Derivative,texture information)on the leaf chlorophyll content inversion accuracy of hyperspectral and UAV multispectral,and further compares the effects of different model algorithms and near-surface remote sensing methods on the leaf chlorophyll content inversion accuracy of ratoon rice.The main results are as follows:(1)The trend of leaf scale hyperspectral reflectance and canopy scale hyperspectral reflectance is consistent,but the leaf scale hyperspectral reflectance is generally higher.In addition,under the condition of different leaf chlorophyll content,the difference between leaf scale and canopy scale hyperspectral responses is mainly in green band and red band.The leaf scale hyperspectral is highly correlated with leaf chlorophyll content in green band and red band,while the canopy scale hyperspectral is only moderately correlated with leaf chlorophyll content in those band.(2)Derivative of leaf scale hyspectral can not improve the inversion results of leaf chlorophyll content of ratoon rice;but the second derivative of canopy scale hyperspectral improves the inversion results of leaf chlorophyll content of ratoon rice.Based on the UAV platform,the multispectral vegetation index combined with texture can provide more accurate inversion results of leaf chlorophyll content of ratoon rice.(3)Comparing the inversion results of leaf chlorophyll content in ratoon rice with different models,the inversion accuracy and stability of the empirical model are higher as a whole,and the inversion accuracy of the physical model is slightly lower.In addition,the inversion results of the three empirical models in the original hyperspectral,canopy scale multispectral and UAV multispectral are not significantly different,the inversion results of the(PLSR)model are poor in the derivative hyperspectral.(4)The coefficient of determination(R~2),root mean square error(RMSE)and relative root mean square error(r RMSE)are used as the evaluation indexes for the inversion accuracy of leaf chlorophyll content of ratoon rice.The inversion results of leaf chlorophyll content of ratoon rice from different near-surface remote sensing data are compared,and the accuracy from high to low is as follows:leaf scale hyperspectral,UAV multispectral,canopy scale hyperspectral and canopy scale multispectral.Although the leaf scale hyperspectral data has the highest inversion accuracy,leaf scale hyperspectral data collection is time-consuming,labor-intensive and inefficient.In contrast,the UAV multispectral remote sensing takes into account both the data collection efficiency and inversion accuracy of leaf chlorophyll content,and is an effective way to achieve the chlorophyll content inversion of ratoon rice.In conclusion,this paper realizes the inversion of leaf chlorophyll content of ratoon rice based on near-surface remote sensing,which proves the feasibility of applying near-surface remote sensing to the inversion of leaf chlorophyll content of ratoon rice.The article provides reference and assistance to effectively utilize agricultural resources and ensure national food security.