Inversion Of Physiological Index Parameters Of Brassica Napus Based On UAV Image

As a modern,diversified and efficient remote sensing technology,UAV(Unmanned Aerial Vehicle)image can monitor the growth of agricultural crops and provide a good growth environment for crops.In recent years,UAV image remote sensing technology continues to achieve new breakthroughs,and are widely used in agricultural production and scientific and technological innovation and other fields.The use of UAV image remote sensing technology can provide a more accurate data support and decision-making basis for crop growth management.It become an effective means in the process of agricultural modernization.In this study with different shooting height of 11 varieties of napus rape as the research object,using the UAV with high-definition camera to obtain images,brassica napus were used comparative analysis of color features,combined with the artificial measurement of rape SPAD value,LAI index correlation analysis,based on UAV image in different height of the physiological indexes of brassica napus were used inversion model,The purpose of this paper is to explore the application of UAV image technology in crop growth information acquisition and growth monitoring.Research shows that:1.It is found that the color features of the rape images are significantly different at different heights by the UAV,and the accuracy of the inversion model is also different.The CMV(Crop Machine Vision)software system was used to obtain the color characteristic values of Brassica napus images taken by the UAV at different heights.The results of variance analysis showed that the color characteristics of the Brassica napus images taken by the UAV at different heights were significantly different,and the correlation analysis results between it and chlorophyll content and leaf area index also showed that there was a significant correlation between the color features of UAV images and chlorophyll content and leaf area index at different heights,but the correlation coefficients were significantly different.Moreover,according to the comparison of the important parameters obtained from the model established by color characteristics and physiological indexes at six heights,the optimal height was selected for the parameter inversion model of physiological indexes in Brassica napus.2.linear regression,robust linear regression,binomial regression,random forest and support vector regression,five inversion models were established between the chlorophyll content of Brassica napus and the most significantly correlated color features(2G-R-B)/V,(2G-R-B)/Y,(2G-R-B)/ 11,(G-B)/R and g-b of UAV Brassica napus images at a height of 7meters.The verification results show that linear regression model and binomial regression model are the most stable.3.linear regression,robust linear regression,binomial regression,random forest and support vector regression rapeseed leaf area index five inversion models were established between Brassica napus’ s leaf area index and the most significantly correlated color features Hv,Hi,(g-b)/(r-b),r /(g + b)and a*/b of UAV Brassica napus image at a height of 30 meters.The verification results show that the linear regression model and binomial regression model are the most stable.