Filters

2019 - 201922020 - 20212

More filters

2021 - 20224
Show more
Reset filters

Settings

DDC: 620 × Type: article × Publisher: Basel : MDPI AG × WGL Institute: ATB ×

Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Optimized Deep Learning Model as a Basis for Fast UAV Mapping of Weed Species in Winter Wheat Crops
    (Basel : MDPI AG, 2021) de Camargo, Tibor; Schirrmann, Michael; Landwehr, Niels; Dammer, Karl-Heinz; Pflanz, Michael
    Weed maps should be available quickly, reliably, and with high detail to be useful for site-specific management in crop protection and to promote more sustainable agriculture by reducing pesticide use. Here, the optimization of a deep residual convolutional neural network (ResNet-18) for the classification of weed and crop plants in UAV imagery is proposed. The target was to reach sufficient performance on an embedded system by maintaining the same features of the ResNet-18 model as a basis for fast UAV mapping. This would enable online recognition and subsequent mapping of weeds during UAV flying operation. Optimization was achieved mainly by avoiding redundant computations that arise when a classification model is applied on overlapping tiles in a larger input image. The model was trained and tested with imagery obtained from a UAV flight campaign at low altitude over a winter wheat field, and classification was performed on species level with the weed species Matricaria chamomilla L., Papaver rhoeas L., Veronica hederifolia L., and Viola arvensis ssp. arvensis observed in that field. The ResNet-18 model with the optimized image-level prediction pipeline reached a performance of 2.2 frames per second with an NVIDIA Jetson AGX Xavier on the full resolution UAV image, which would amount to about 1.78 ha h−1 area output for continuous field mapping. The overall accuracy for determining crop, soil, and weed species was 94%. There were some limitations in the detection of species unknown to the model. When shifting from 16-bit to 32-bit model precision, no improvement in classification accuracy was observed, but a strong decline in speed performance, especially when a higher number of filters was used in the ResNet-18 model. Future work should be directed towards the integration of the mapping process on UAV platforms, guiding UAVs autonomously for mapping purpose, and ensuring the transferability of the models to other crop fields.
  • Thumbnail Image
    Item
    Evaluating Soil-Borne Causes of Biomass Variability in Grassland by Remote and Proximal Sensing
    (Basel : MDPI AG, 2019) Vogel, Sebastian; Gebbers, Robin; Oertel, Marcel; Kramer, Eckart
    On a grassland field with sandy soils in Northeast Germany (Brandenburg), vegetation indices from multi-spectral UAV-based remote sensing were used to predict grassland biomass productivity. These data were combined with soil pH value and apparent electrical conductivity (ECa) from on-the-go proximal sensing serving as indicators for soil-borne causes of grassland biomass variation. The field internal magnitude of spatial variability and hidden correlations between the variables of investigation were analyzed by means of geostatistics and boundary-line analysis to elucidate the influence of soil pH and ECa on the spatial distribution of biomass. Biomass and pH showed high spatial variability, which necessitates high resolution data acquisition of soil and plant properties. Moreover, boundary-line analysis showed grassland biomass maxima at pH values between 5.3 and 7.2 and ECa values between 3.5 and 17.5 mS m−1. After calibrating ECa to soil moisture, the ECa optimum was translated to a range of optimum soil moisture from 7% to 13%. This matches well with to the plant-available water content of the predominantly sandy soil as derived from its water retention curve. These results can be used in site-specific management decisions to improve grassland biomass productivity in low-yield regions of the field due to soil acidity or texture-related water scarcity.
  • Thumbnail Image
    Item
    Analyzing temporal and spatial characteristics of crop parameters using Sentinel-1 backscatter data
    (Basel : MDPI AG, 2019) Harfenmeister, Katharina; Spengler, Daniel; Weltzien, Cornelia
    The knowledge about heterogeneity on agricultural fields is essential for a sustainable and effective field management. This study investigates the performance of Synthetic Aperture Radar (SAR) data of the Sentinel-1 satellites to detect variability between and within agricultural fields in two test sites in Germany. For this purpose, the temporal profiles of the SAR backscatter in VH and VV polarization as well as their ratio VH/VV of multiple wheat and barley fields are illustrated and interpreted considering differences between acquisition settings, years, crop types and fields. Within-field variability is examined by comparing the SAR backscatter with several crop parameters measured at multiple points in 2017 and 2018. Structural changes, particularly before and after heading, as well as moisture and crop cover differences are expressed in the backscatter development. Furthermore, the crop parameters wet and dry biomass, absolute and relative vegetation water content, leaf area index (LAI) and plant height are related to SAR backscatter parameters using linear and exponential as well as multiple regression. The regression performance is evaluated using the coefficient of determination (R2) and the root mean square error (RMSE) and is strongly dependent on the phenological growth stage. Wheat shows R2 values around 0.7 for VV backscatter and multiple regression and most crop parameters before heading. Single fields even reach R2 values above 0.9 for VV backscatter and for multiple regression related to plant height with RMSE values around 10 cm. The formulation of clear rules remains challenging, as there are multiple influencing factors and uncertainties and a lack of conformity. © 2019 by the authors.
  • Thumbnail Image
    Item
    Determination of Nutrients in Liquid Manures and Biogas Digestates by Portable Energy-Dispersive X-ray Fluorescence Spectrometry
    (Basel : MDPI AG, 2021) Horf, Michael; Gebbers, Robin; Vogel, Sebastian; Ostermann, Markus; Piepel, Max-Frederik; Olfs, Hans-Werner
    Knowing the exact nutrient composition of organic fertilizers is a prerequisite for their appropriate application to improve yield and to avoid environmental pollution by over-fertilization. Traditional standard chemical analysis is cost and time-consuming and thus it is unsuitable for a rapid analysis before manure application. As a possible alternative, a handheld X-ray fluorescence (XRF) spectrometer was tested to enable a fast, simultaneous, and on-site analysis of several elements. A set of 62 liquid pig and cattle manures as well as biogas digestates were collected, intensively homogenized and analysed for the macro plant nutrients phosphorus, potassium, magnesium, calcium, and sulphur as well as the micro nutrients manganese, iron, copper, and zinc using the standard lab procedure. The effect of four different sample preparation steps (original, dried, filtered, and dried filter residues) on XRF measurement accuracy was examined. Therefore, XRF results were correlated with values of the reference analysis. The best R2s for each element ranged from 0.64 to 0.92. Comparing the four preparation steps, XRF results for dried samples showed good correlations (0.64 and 0.86) for all elements. XRF measurements using dried filter residues showed also good correlations with R2s between 0.65 and 0.91 except for P, Mg, and Ca. In contrast, correlation analysis for liquid samples (original and filtered) resulted in lower R2s from 0.02 to 0.68, except for K (0.83 and 0.87, respectively). Based on these results, it can be concluded that handheld XRF is a promising measuring system for element analysis in manures and digestates.