Abstract: 【Background】In recent years, with the rapid development of remote sensing technology, real-time and non-destructive monitoring of crop growth status has become a research hotspot. The agricultural information obtained by remote sensing will provide guidance for the precise management of large areas of crops. Among many remote sensing monitoring platforms, UAVs have attracted wide attention due to their simple operation and low cost. UAVs equipped with multi-spectral cameras can quickly obtain crop growth conditions.【Objective】This study attempted to combine the texture information and spectral information of multispectral images of fixed-wing UAVs to explore the monitoring effect of "atlas" information on rice growth indicators.【Method】A 2-year rice field experiment involving different sowing dates, varieties, planting methods and nitrogen levels was conducted. Remote sensing images of the rice canopy were obtained by RedEdge multispectral camera mounted on a fixed-wing UAV during the key growth period of rice. Shoot destructive sampling was conducted simultaneously to obtain leaf area index (LAI), aboveground biomass (AGB), plant nitrogen content (PNC) and other agronomic indexes of rice. Simple regression, partial least squares regression and artificial neural network were used to construct a rice growth index monitoring model based on multispectral images of fixed-wing UAV. The monitoring effect of spectral texture information in different models is compared and analyzed.【Result】In this study, the quantitative relationship between vegetation index, single-band texture features and rice LAI, AGB, and PNC was first explored using simple linear regression. The results showed that there was a good correlation between vegetation indexes (VIs) and LAI and AGB, and the best vegetation indexes were CIRE and NDRE. The R squared values were 0.80 and 0.76, respectively. However, for PNC monitoring, the vegetation indexes did not achieve the ideal result, and the determination coefficient between RESAVI and NDRE and PNC was only 0.13, and it was found that the single-band texture is not as good as VIs in the monitoring of rice growth indicators. In order to further analyze the monitoring effect of image texture on the above three indexes, the normalized texture indexes (NDTI), the ratio texture indexes (RTI), and the difference texture indexes (DTI) were constructed with reference to the construction method of VIs in this paper. Correlation analysis showed that the newly constructed texture index (TIs) improved the monitoring accuracy of rice growth index compared with the single band texture feature, but the results were not better than the vegetation index. In order to achieve the combination of spectrum and texture, partial least squares and artificial neural network modeling methods were adopted in this paper. VIs and VIs+TIs were used as different input parameter combinations to construct rice LAI, AGB and PNC monitoring models. The results showed that compared with simple regression modeling, partial least squares and artificial neural network modeling methods can effectively improve the monitoring accuracy of the model. Among them, VIs+TIs were used as input parameters, and artificial neural network was used to construct the model, and the verification accuracy of the model reaches the optimum. The model verification accuracy of LAI has increased from 75% to 86%, while the model verification accuracy of AGB and PNC has also increased from 72% and 26% to 92% and 86%, and the RMSE of the model has been significantly decreased.【Conclusion】This study showed that the monitoring accuracy of rice LAI, AGB and PNC could be effectively improved by using the fixed-wing UAV to collect multispectral images of rice canopy and using the texture features and reflectance information as input parameters of the model through the model construction method of artificial neural network. The research results would provide a theoretical basis for the rapid monitoring of large area crop growth.

Key words: unmanned aerial vehicle (UAV), vegetation index, texture feature, growth, rice