The efficient use of nitrogen fertilizer is a crucial problem in modern agriculture. Fertilization has to be minimized to reduce environmental impacts but done so optimally without negatively affecting yield. In June 2017, a controlled experiment with eight different nitrogen treatments was applied to winter wheat plants and investigated with the UAV-based hyperspectral pushbroom camera Resonon Pika-L (400–1000 nm). The system, in combination with an accurate inertial measurement unit (IMU) and precise gimbal, was very stable and capable of acquiring hyperspectral imagery of high spectral and spatial quality. Additionally, in situ measurements of 48 samples (leaf area index (LAI), chlorophyll (CHL), and reflectance spectra) were taken in the field, which were equally distributed across the different nitrogen treatments. These measurements were used to predict grain yield, since the parameter itself had no direct effect on the spectral reflection of plants. Therefore, we present an indirect approach based on LAI and chlorophyll estimations from the acquired hyperspectral image data using partial least-squares regression (PLSR). The resulting models showed a reliable predictability for these parameters (R²_LAI = 0.79, RMSE_{LAI [m²m⁻²]} = 0.18, R²_CHL = 0.77, RMSE_{CHL [µg cm⁻²]} = 7.02). The LAI and CHL predictions were used afterwards to calibrate a multiple linear regression model to estimate grain yield (R²_yield = 0.88, RMSE_{yield [dt ha⁻¹]} = 4.18). With this model, a pixel-wise prediction of the hyperspectral image was performed. The resulting yield estimates were validated and opposed to the different nitrogen treatments, which revealed that, above a certain amount of applied nitrogen, further fertilization does not necessarily lead to larger yield.

In order to meet the increasing demand for crops under challenging climate conditions, efficient and sustainable cultivation strategies are becoming essential in agriculture. Targeted herbicide use reduces environmental pollution and effectively controls weeds as a major cause of yield reduction. The key requirement is a reliable weed detection system that is accessible to a wide range of end users. This research paper introduces a self-built, low-cost, multispectral camera system and evaluates it against the high-end MicaSense Altum system. Pixel-based weed and crop classification was performed on UAV datasets collected with both sensors in maize using a U-Net. The training and testing data were generated via an index-based thresholding approach followed by annotation. As a result, the F1-score for the weed class reached 82% on the Altum system and 76% on the low-cost system, with recall values of 75% and 68%, respectively. Misclassifications occurred on the low-cost system images for small weeds and overlaps, with minor oversegmentation. However, with a precision of 90%, the results show great potential for application in automated weed control. The proposed system thereby enables sustainable precision farming for the general public. In future research, its spectral properties, as well as its use on different crops with real-time on-board processing, should be further investigated.

Pumpkins are economically and nutritionally valuable vegetables with increasing popularity and acreage across Europe. Successful commercialization, however, require detailed pre-harvest information about number and weight of the fruits. To get a non-destructive and cost-effective yield estimation, we developed an image processing methodology for high-resolution RGB data from Unmanned aerial vehicle (UAV) and applied this on a Hokkaido pumpkin farmer’s field in North-western Germany. The methodology was implemented in the programming language Python and comprised several steps, including image pre-processing, pixel-based image classification, classification post-processing for single fruit detection, and fruit size and weight quantification. To derive the weight from two-dimensional imagery, we calculated elliptical spheroids from lengths of diameters and heights. The performance of this processes was evaluated by comparison with manually harvested ground-truth samples and cross-checked for misclassification from randomly selected test objects. Errors in classification and fruit geometry could be successfully reduced based on the described processing steps. Additionally, different lighting conditions, as well as shadows, in the image data could be compensated by the proposed methodology. The results revealed a satisfactory detection of 95% (error rate of 5%) from the field sample, as well as a reliable volume and weight estimation with Pearson’s correlation coefficients of 0.83 and 0.84, respectively, from the described ellipsoid approach. The yield was estimated with 1.51 kg m⁻² corresponding to an average individual fruit weight of 1100 g and an average number of 1.37 pumpkins per m². Moreover, spatial distribution of aggregated fruit densities and weights were calculated to assess in-field optimization potential for agronomic management as demonstrated between a shaded edge compared to the rest of the field. The proposed approach provides the Hokkaido producer useful information for more targeted pre-harvest marketing strategies, since most food retailers request homogeneous lots within prescribed size or weight classes.

The determination of soil texture and organic carbon across agricultural areas provides important information to derive soil condition. Precise digital soil maps can help to till agricultural fields with more accuracy, greater cost-efficiency and better environmental protection. In the present study, the laboratory analysis of sand, silt, clay and soil organic carbon (SOC) content was combined with hyperspectral image data to estimate the distribution of soil texture and SOC across an agricultural area. The aim was to identify regions with similar soil properties and derive uniform soil regions based on this information. Soil parameter data and corresponding laboratory spectra were used to calibrate cross-validated (leave-one-out) partial least squares regression (PLSR) models, resulting in robust models for sand (R² = 0.77, root-mean-square error (RMSE) = 5.37) and SOC (R² = 0.89, RMSE = 0.27), as well as moderate models for silt (R² = 0.62, RMSE = 5.46) and clay (R² = 0.53, RMSE = 2.39). The regression models were applied to Airborne Imaging Spectrometer for Applications DUAL (aisaDUAL) hyperspectral image data to spatially estimate the concentration of these parameters. Afterwards, a decision tree, based on the Food and Agriculture Organization (FAO) soil texture classification scheme, was developed to determine the soil texture for each pixel of the hyperspectral airborne data. These soil texture regions were further refined with the spatial SOC estimations. The developed method is useful to identify spatial regions with similar soil properties, which can provide a vital information source for an adapted treatment of agricultural fields in terms of the necessary amount of fertilizers or water. The approach can also be adapted to wider regions with a larger sample size to create detailed digital soil maps (DSMs). Further, the presented method should be applied to future hyperspectral satellite missions like Environmental Mapping and Analysis Program (EnMap) and Hyperspectral Infrared Imager (HyspIRI) to cover larger areas in shorter time intervals. Updated DSMs on a regular basis could particularly support precision farming aspects.

The potential of field spectroscopy is shown by the quantitative assessment of summer barley canopy properties. A data set was acquired, comprising top of canopy reflectance (0.4–2.4 μm) and the corresponding canopy parameters of summer barley stands for two phenological stages. Canopy variables (leaf area index, canopy chlorophyll, water, and fresh matter contents) were predicted by partial least squares regression (PLSR) from field spectra resampled to HyMap spectral resolution with very high accuracy (r ² _cv > 0.88). PLSR models obtained from field spectra were applied to HyMap data to allow spatial prediction of summer barley canopy properties. Subsequently, to assess accuracy of the spatially distributed estimates, predictions obtained from the image data were validated with reference to the vegetation samples taken in the field. The predicted values reflect the spatial conditions of the investigated plot. The results indicate the potential of this method as a quick screening tool for spatial assessment of canopy variables.