Detecting Soil Moisture Content Based On Multispectral Images From Unmanned Aerial Vehicle

China is a country with serious water shortage,and the development of agriculture has always been constrained by water shortage.How to use new technology to detect the soil moisture content of farmland in real time is the key to realize reasonable planning of agricultural water use.Traditional soil moisture content detection methods are time-consuming and labor-intensive,and the detection range is limited.The use of UAV low-altitude remote sensing for soil moisture content detection in agricultural fields can make up for these shortcomings.Therefore,this study uses UAVs to collect canopy multispectral images of three crops,corn,cotton and peanut,extract spectral reflectance and calculate vegetation indices to construct data sets respectively,and use machine learning algorithms to establish a soil water content detection model based on UAV multispectral images to achieve rapid and accurate detection of soil moisture content in agricultural fields.The main work of this study is as follows:(1)Acquisition and processing of UAV multispectral images and soil moisture content data.Through comprehensive consideration of the experimental field corn,cotton,peanut planting area,soil moisture conditions and the flight conditions of the UAV,a reasonable data collection program was set up for data collection and a large amount of effective data was obtained.The acquired multispectral images were preprocessed,and the data sets for the three crops were constructed using eight parameters,R,B,G,RE,NIR,MSR,RVI,and SIPI,as independent variables and soil letter moisture content as dependent variables,and the training and test sets were divided in the ratio of 7:3.(2)Multispectral soil moisture content detection model extended detection study.To investigate whether the soil moisture content detection model established using the maize dataset can be extended to other crops,datasets collated based on multispectral images of cotton and peanut were brought into the model for testing.The results show that the detection effect of peanut is better than that of cotton,but there is still a big difference with that of maize.Soil moisture content detection models based on multispectral data were established for cotton and peanut,respectively,and the best detection models for both crops were GBDT models.The R~2,RMSE and RPD of the best detection model for cotton soil moisture content reached 0.752,0.065 and 2.009,respectively,while the R~2,RMSE and RPD of the best detection model for peanut soil moisture content reached 0.738,0.061 and 1.954,respectively,with high detection accuracy.(3)Multispectral soil moisture content detection model extended detection study.To investigate whether the soil moisture content detection model established by using the maize dataset can be extended to other crops,datasets based on multispectral images of cotton and peanut were brought into the model for testing.The results show that the detection effect of peanut is better than that of cotton,but there is still a big difference with that of maize.Soil moisture content detection models based on multispectral data were established for cotton and peanut,respectively,and the best detection models for both crops were GBDT models.The R2,RMSE and RPD of the best detection model for cotton soil moisture content reached 0.752,0.065 and 2.009,respectively,while the R2,RMSE and RPD of the best detection model for peanut soil moisture content reached 0.738,0.061 and 1.954,respectively,with high detection accuracy.(4)Implementation of a soil moisture content analysis system based on UAV multispectral imagery.To enable practical application of soil moisture content detection models based on three crops,a soil moisture content detection system with functions such as reflectance extraction,soil moisture content detection,and moisture distribution visualization was developed in this study.