•We computed Normalized Difference Texture Index (NDTI) from fixed-wing UAS images.•Random Forest was employed to combine NDTI, vegetation index, and color index.•The optimal image resolution for ...texture extraction may depend on crop size.
Precision crop management in modern agriculture requires timely and effective acquisition of crop growth information. Recently, unmanned aerial systems (UASs) have rapidly developed and are now widely used in crop remote sensing (RS). Vegetation index (VI) and color index (CI) are commonly used RS methods to monitor crops. Texture is intrinsic information of the images, which can reflect the crop canopy structure and be used for vegetation classification. The objective of this study was to explore the potential of combining VI, CI, and texture to improve the estimation accuracy of wheat growth parameters based on fixed-wing UAS imagery. Wheat field experiments were carried out at the Xinghua Experimental Station for two consecutive years of 2017–2019 on three wheat cultivars under five nitrogen fertilization rates. Two commonly used wheat growth parameters, leaf area index (LAI) and leaf dry matter (LDM), synchronized with wheat field UAS images, were obtained at key growth stages. Simple regression (SR) was used to determine quantitative relationships between RS variables (VI, CI, and texture) and LAI, LDM. The data showed that individual texture does not correlate well with wheat growth parameters, while a texture index (TI), containing two texture measurements, showed stronger correlation with LAI and LDM. With the utilization of simple regression (SR), VI (R2 > 0.65, RRMSE < 21.87%) exhibited the best accuracy in estimating LAI and LDM, followed by TI (R2 > 0.51, RRMSE < 26.28%) and CI (R2 > 0.34, RRMSE < 27.74%). Multiple linear regression (MLR) and random forest (RF) were further employed to develop LAI and LDM estimation models using different input variable sets (VIs, VIs + CIs, and VIs + CIs + TIs). Compared with SR and MLR, the RF models that combined VIs, CIs, and TIs greatly improved the estimation accuracy of LAI and LDM, and the validated R2 of the best RF models for LAI and LDM estimation reached 0.78 and 0.78 (RRMSE = 17.32% and 13.83%) in pre-heading stages, 0.81 and 0.77 (RRMSE = 17.86% and 16.08%) in post-heading stages, and 0.76 and 0.75 (RRMSE = 18.13% and 16.79%) in all stages, respectively. This study demonstrated that image textures can assist wheat monitoring to achieve higher estimation accuracy of LAI and LDM, and fixed-wing UAS is a promising platform that can provide reliable data for large-scale crop management.
Display omitted
•Stable co-digestion of up to 20% chicken manure on VS basis with maize silage.•Threshold and total inhibition of biogas production occurred at 7 and 9gNL−1.•Dominant hydrogenotrophic ...methanogenesis reverts to acetoclastic on lower N input.•Loss of methanogenesis led to total VFA>60gL−1 before buffering failed.•Sub-optimal solids breakdown as a result of multi-inhibition.
The feasibility of co-digestion of chicken manure (CM) and maize silage (MS) without water dilution was investigated in 5-L digesters. Specific methane production (SMP) of 0.309LCH4g−1 volatile solids (VS) was achieved but only at lower %CM. Above a critical threshold for total ammonia nitrogen (TAN), estimated at 7gNL−1, VFA accumulated with a characteristic increase in acetic acid followed by its reduction and an increase in propionic acid. During this transition the predominant methanogenic pathway was hydrogenotrophic. Methanogenesis was completely inhibited at TAN of 9gNL−1. The low digestibility of the mixed feedstock led to a rise in digestate TS and a reduction in SMP over the 297-day experimental period. Methanogenesis appeared to be failing in one digester but was recovered by reducing the %CM. Co-digestion was feasible with CM ⩽20% of feedstock VS, and the main limiting factor was ammonia inhibition.
Deep-learning-based object detection algorithms have significantly improved the performance of wheat spike detection. However, UAV images crowned with small-sized, highly dense, and overlapping ...spikes cause the accuracy to decrease for detection. This paper proposes an improved YOLOv5 (You Look Only Once)-based method to detect wheat spikes accurately in UAV images and solve spike error detection and miss detection caused by occlusion conditions. The proposed method introduces data cleaning and data augmentation to improve the generalization ability of the detection network. The network is rebuilt by adding a microscale detection layer, setting prior anchor boxes, and adapting the confidence loss function of the detection layer based on the IoU (Intersection over Union). These refinements improve the feature extraction for small-sized wheat spikes and lead to better detection accuracy. With the confidence weights, the detection boxes in multiresolution images are fused to increase the accuracy under occlusion conditions. The result shows that the proposed method is better than the existing object detection algorithms, such as Faster RCNN, Single Shot MultiBox Detector (SSD), RetinaNet, and standard YOLOv5. The average accuracy (AP) of wheat spike detection in UAV images is 94.1%, which is 10.8% higher than the standard YOLOv5. Thus, the proposed method is a practical way to handle the spike detection in complex field scenarios and provide technical references for field-level wheat phenotype monitoring.
Crop aboveground biomass (AGB) is one of the most important indicators in crop breeding and crop management, and can be used for crop yield prediction. A number of vegetation indices (VIs) have been ...proposed to estimate crop biomass, but they perform poorly at high biomass levels and are easily affected by background materials. Texture analysis has been proved to be an efficient approach in forest biomass estimation, but has never been applied to crops with low-altitude unmanned aerial vehicle (UAV) images. The objective of this study was to improve rice AGB estimation by combining textural and spectral analysis of UAV imagery. A two-year rice experiment was conducted in 2015 and 2016, involving different nitrogen (N) rates, planting densities and rice cultivars with three replicates. A six-band multispectral (MS) camera was mounted on a UAV to acquire rice canopy images at critical stages during the rice growing seasons and concurrent field samplings were taken. Simple regression and stepwise multiple linear regression models were developed between biomass data from the two-year experiment and image parameters derived from four different types of feature sets. These features represented commonly used VIs, texture parameters, normalization of texture measurements (normalized difference texture index, NDTI) and combinations of VIs and NDTIs. Finally, all the regression models were evaluated by cross-validation over pooled data with the coefficient of determination (R
2
) and the root mean square error (RMSE). Results demonstrated that the optimized soil adjusted vegetation index (OSAVI) exhibited the best relationship with AGB for the whole season (R
2
= 0.63) and post-heading stages (R
2
= 0.65). Red-edge-based indices yielded best performance (R
2
> 0.70) only for the growth stages before heading. The texture measurement
mean
(MEA) from the NIR band was the best among the eight candidates in AGB estimation. Texture index (NDTI (MEA
800
, MEA
550
)) was superior to all the evaluated VIs in estimating AGB for the whole season (R
2
= 0.75) and pre-heading stages (R
2
= 0.84). Further improvement was obtained across the whole season by combining NDTIs and VIs through a multiple linear regression. This multivariate model produced the highest estimation accuracy for all stages (R
2
= 0.78 and RMSE = 1.84 t ha
−1
) and different stage groups (R
2
= 0.84 and RMSE = 1.06 t ha
−1
for pre-heading stages and R
2
= 0.65 and RMSE = 1.94 t ha
−1
for post-heading stages). The findings imply that the integration of textural information with spectral information significantly improves the accuracy for rice biomass estimation compared to the use of spectral information alone.
Simulated wheat yield (kgha (1) in 2000s (A) and yield change in 2030s (B), 2050s (C), and 2070s (D) under three scenarios (A1: world markets, high-greenhouse-gas-emission scenario; A2: provincial ...enterprise, a medium-emission scenario; B1: global sustainability, a low-emissions scenario) of climate-change and rain-fed condition. Positive values in figure B–D means the reduced wheat yield compare to 2000s. Display omitted
► The changes in the grain filling period are not main factor for yield reduction. ► Rain-fed wheat yields showed obvious differences between north and south China. ► The spatial pattern of ET change is quite similar to that of yield change.
Wheat is the second primary crop in China. Wheat production in China is an important component for national food security. The combination of high-resolution Global Climate Model (GCM) and WheatGrow model was used to assess the effects of climate change on wheat yields in the main wheat production regions of China. With the application of many techniques including the downscaling of meteorological data, rasterizing of sowing date, parameterization of region cultivar and vectorization of soil data, the spatial data in study area is divided into homogeneous grids with the resolution of 0.1°×0.1°. The grid is taken as the basic simulation unit, and each grid has a complete set of input data (meteorological, soil, management and varieties). Regional productivities are simulated with WheatGrow for each grid cell under scenarios of climate-change. There is an advance in flowering date in future climate compare to 2000s, but with a more homogeneous pattern for the whole producing region. The changes in grain filling period are relatively stable. Under rain-fed conditions, wheat yield is reduced in the north regions of China in three future periods, while wheat yield increases in the south regions of China. Under full-irrigation conditions, irrigated wheat yields will increase in almost all regions of whole producing region. The spatial pattern of evapotranspiration change is quite similar to that of yield change under rain-fed and full-irrigation conditions. The correlation between wheat yield and evapotranspiration (ET) increases to 0.96 and 0.51 (p<0.01) under rain-fed and full-irrigation conditions, respectively. The irrigation water use efficiency (IWUE) will decrease under three time slices in 2030s, 2050s and 2070s in western Shandong, southern Sichuan, as well as northern Henan, Shanxi and Shaanxi, while IWUE will increase under scenarios of climate-change in other areas. The results revealed that the increase in effective irrigation in the future would help to increase the ET and further improve the wheat yield in the northern regions of China, and the limited water should be mad full use of in the regions with relatively high IWUE under scenarios of climate-change.
•An algorithm for fusing vegetation indices with different spatial and temporal resolution was developed.•The optimal spectral index for predicting grain yield is RVI(Nir, Red) at initial filling ...stage.•The optimal spectral index for predicting grain protein content is RNir/(RRed+RGreen) during anthesis.•The accumulated spectral index gives higher prediction accuracy for grain yield and protein content than at a single period.
Non-destructive and quick assessment of grain yield and protein content is needed in modern wheat production. This study was undertaken to determine the optimal spectral index and the best time for predicting grain yield and grain protein content in wheat by fusing multi-sensor and multi-temporal remote-sensing images. Four field experiments were carried out at different locations, cultivars and nitrogen rates in two growing seasons of winter wheat (Triticum aestivum L.). During the experiment periods, data were obtained on time series RS images fused with high temporal and spatial resolutions, along with grain yields and protein contents at maturity. The results showed that the normalized difference vegetation index (NDVI) estimated by fusion exhibits high consistency with the SPOT-5 NDVI, which confirmed the usefulness of related algorithm. The periods around initial gain filling and anthesis stages were identified as the best periods for estimating wheat grain yield and protein content, respectively. The use of ratio vegetation index (RVI) (Nir, Red) at the initial filling stage obtained enhanced accuracy in wheat yield prediction, while the index RNir/(RRed+RGreen) during anthesis predicted grain protein content more accurately than that at other growth stages. In addition, the accumulated spectral index ∑RVI (Nir, Red) and ∑(RNir/(RRed+RGreen)) from jointing to initial filling stage gave higher prediction accuracy for grain yield and protein content, respectively, than the spectral index at a single period. These results help provide a technical approach to the prediction of grain yield and grain protein content in wheat with remote sensing at a large scale.
Two floating treatment wetlands (FTWs) in experimental tanks were compared in terms of their effectiveness on removing nutrients. The results showed that the FTWs were dominated by emergent wetland ...plants and were constructed to remove nutrients from simulated urban stormwater. Iris pseudacorus and Thalia dealbata wetland systems were effective in reducing the nutrient. T. dealbata FTWs showed higher nutrient removal performance than I. pseudacorus FTWs. Nitrogen (N) and phosphorous (P) removal rates in water by T. dealbata FTWs were 3.95 ± 0.19 and 0.15 ± 0.01 g/m
/day, respectively. For I. pseudacorus FTWs, the TN and TP removal rates were 3.07 ± 0.15 and 0.14 ± 0.01 g/m
/day, respectively. The maximum absolute growth rate for T. dealbata corresponded directly with the maximum mean nutrient removal efficiency during the 5th stage. At harvest, N and P uptak of T. dealbata was 23.354 ± 1.366 g and 1.489 ± 0.077 g per plant, respectively, approximate twice as high as by I. pseudacorus.
Aboveground biomass (AGB) is a widely used agronomic parameter for characterizing crop growth status and predicting grain yield. The rapid and accurate estimation of AGB in a non-destructive way is ...useful for making informed decisions on precision crop management. Previous studies have investigated vegetation indices (VIs) and canopy height metrics derived from Unmanned Aerial Vehicle (UAV) data to estimate the AGB of various crops. However, the input variables were derived either from one type of data or from different sensors on board UAVs. Whether the combination of VIs and canopy height metrics derived from a single low-cost UAV system can improve the AGB estimation accuracy remains unclear. This study used a low-cost UAV system to acquire imagery at 30 m flight altitude at critical growth stages of wheat in Rugao of eastern China. The experiments were conducted in 2016 and 2017 and involved 36 field plots representing variations in cultivar, nitrogen fertilization level and sowing density. We evaluated the performance of VIs, canopy height metrics and their combination for AGB estimation in wheat with the stepwise multiple linear regression (SMLR) and three types of machine learning algorithms (support vector regression, SVR; extreme learning machine, ELM; random forest, RF).
Our results demonstrated that the combination of VIs and canopy height metrics improved the estimation accuracy for AGB of wheat over the use of VIs or canopy height metrics alone. Specifically, RF performed the best among the SMLR and three machine learning algorithms regardless of using all the original variables or selected variables by the SMLR. The best accuracy (
= 0.78, RMSE = 1.34 t/ha, rRMSE = 28.98%) was obtained when applying RF to the combination of VIs and canopy height metrics.
Our findings implied that an inexpensive approach consisting of the RF algorithm and the combination of RGB imagery and point cloud data derived from a low-cost UAV system at the consumer-grade level can be used to improve the accuracy of AGB estimation and have potential in the practical applications in the rapid estimation of other growth parameters.
Leaf area index (LAI) and leaf dry matter (LDM) are important indices of crop growth. Real-time, nondestructive monitoring of crop growth is instructive for the diagnosis of crop growth and ...prediction of grain yield. Unmanned aerial vehicle (UAV)-based remote sensing is widely used in precision agriculture due to its unique advantages in flexibility and resolution. This study was carried out on wheat trials treated with different nitrogen levels and seeding densities in three regions of Jiangsu Province in 2018–2019. Canopy spectral images were collected by the UAV equipped with a multi-spectral camera during key wheat growth stages. To verify the results of the UAV images, the LAI, LDM, and yield data were obtained by destructive sampling. We extracted the wheat canopy reflectance and selected the best vegetation index for monitoring growth and predicting yield. Simple linear regression (LR), multiple linear regression (MLR), stepwise multiple linear regression (SMLR), partial least squares regression (PLSR), artificial neural network (ANN), and random forest (RF) modeling methods were used to construct a model for wheat yield estimation. The results show that the multi-spectral camera mounted on the multi-rotor UAV has a broad application prospect in crop growth index monitoring and yield estimation. The vegetation index combined with the red edge band and the near-infrared band was significantly correlated with LAI and LDM. Machine learning methods (i.e., PLSR, ANN, and RF) performed better for predicting wheat yield. The RF model constructed by normalized difference vegetation index (NDVI) at the jointing stage, heading stage, flowering stage, and filling stage was the optimal wheat yield estimation model in this study, with an R2 of 0.78 and relative root mean square error (RRMSE) of 0.1030. The results provide a theoretical basis for monitoring crop growth with a multi-rotor UAV platform and explore a technical method for improving the precision of yield estimation.
Crop production will likely face enormous challenges against the occurrences of extreme climatic events projected under future climate change. Heat waves that occur at critical stages of the ...reproductive phase have detrimental impacts on the grain yield formation of rice (Oryza sativa). Accurate estimates of these impacts are essential to evaluate the effects of climate change on rice. However, the accuracy of these predictions by crop models has not been extensively tested. In this study, we evaluated 14 rice growth models against four year phytotron experiments with four levels of heat treatments imposed at different times after flowering. We found that all models greatly underestimated the negative effects of heat on grain yield, suggesting that yield projections with these models do not reflect food shocks that may occur under short‐term extreme heat stress (SEHS). As a result, crop model ensembles do not help to provide accurate estimates of grain yield under heat stress. We examined the functions of grain‐setting rate response to temperature (TRF_GS) used in eight models and showed that adjusting the effective periods of TRF_GS improved the model performance, especially for models simulating accumulative daily temperature effects. For TRF_GS which uses daily maximum temperature averaged for the effective period, the models provided better grain yield estimates by using maximum temperatures averaged only when daily maximum temperatures exceeded the base temperature (Tbase). An alternative method based on heating‐degree days and stage‐dependent heat sensitivity parameters further decreased the prediction uncertainty of grain yield under heat stress, where stage‐dependent heat sensitivity was more important than heat dose for model improvement under SEHS. These results suggest the limitation of the applicability of existing rice models to variable climatic conditions and the urgent need for an alternative grain‐setting function accounting for the stage‐dependent heat sensitivity.
Current rice models greatly underestimated the negative and acute effects of short‐term extreme heat stress on grain yields, while crop model ensembles did not help to provide accurate estimates of grain yield under heat stress. In this study, we adjusted the effective periods and inputs of temperature functions for grain‐setting rate response to temperature, which improved the model performance significantly. An alternative method based on heat dose and stage‐dependent heat sensitivity parameters further decreased the prediction uncertainty of grain yield greatly under heat stress.
PDF
