A large-scale, outdoor pervasive computing system uses static and animal-borne nodes to measure the state of a complex system comprising climate, soil, pasture, and animals. Agriculture faces many challenges, such as climate change, water shortages, labor shortages due to an aging urbanized population, and increased societal concern about issues such as animal welfare, food safety, and environmental impact. Humanity depends on agriculture and water for survival, so optimal, profitable, and sustainable use of our land and water resources is critical.

Understanding the growth and development of individual plants is of central importance in modern agriculture, crop breeding, and crop science. To this end, using 3D data for plant analysis has gained attention over the last years. High-resolution point clouds offer the potential to derive a variety of plant traits, such as plant height, biomass, as well as the number and size of relevant plant organs. Periodically scanning the plants even allows for performing spatio-temporal growth analysis. However, highly accurate 3D point clouds from plants recorded at different growth stages are rare, and acquiring this kind of data is costly. Besides, advanced plant analysis methods from machine learning require annotated training data and thus generate intense manual labor before being able to perform an analysis. To address these issues, we present with this dataset paper a multi-temporal dataset featuring high-resolution registered point clouds of maize and tomato plants, which we manually labeled for computer vision tasks, such as for instance segmentation and 3D reconstruction, providing approximately 260 million labeled 3D points. To highlight the usability of the data and to provide baselines for other researchers, we show a variety of applications ranging from point cloud segmentation to non-rigid registration and surface reconstruction. We believe that our dataset will help to develop new algorithms to advance the research for plant phenotyping, 3D reconstruction, non-rigid registration, and deep learning on raw point clouds. The dataset is freely accessible athttps://www.ipb.uni-bonn.de/data/pheno4d/.

In this paper, a newly-developed direct georeferencing system for the guidance, navigation and control of lightweight unmanned aerial vehicles (UAVs), having a weight limit of 5 kg and a size limit of 1.5 m, and for UAV-based surveying and remote sensing applications is presented. The system is intended to provide highly accurate positions and attitudes (better than 5 cm and 0.5°) in real time, using lightweight components. The main focus of this paper is on the attitude determination with the system. This attitude determination is based on an onboard single-frequency GPS baseline, MEMS (micro-electro-mechanical systems) inertial sensor readings, magnetic field observations and a 3D position measurement. All of this information is integrated in a sixteen-state error space Kalman filter. Special attention in the algorithm development is paid to the carrier phase ambiguity resolution of the single-frequency GPS baseline observations. We aim at a reliable and instantaneous ambiguity resolution, since the system is used in urban areas, where frequent losses of the GPS signal lock occur and the GPS measurement conditions are challenging. Flight tests and a comparison to a navigation-grade inertial navigation system illustrate the performance of the developed system in dynamic situations. Evaluations show that the accuracies of the system are 0.05° for the roll and the pitch angle and 0.2° for the yaw angle. The ambiguities of the single-frequency GPS baseline can be resolved instantaneously in more than 90% of the cases.

Unmanned aerial vehicles (UAVs) open new opportunities in precision agriculture and phenotyping because of their flexibility and low cost. In this study, the potential of UAV imagery was evaluated to quantify lodging percentage and lodging severity of barley using structure from motion (SfM) techniques. Traditionally, lodging quantification is based on time-consuming manual field observations. Our UAV-based approach makes use of a quantitative threshold to determine lodging percentage in a first step. The derived lodging estimates showed a very high correlation to reference data (R² = 0.96, root mean square error (RMSE) = 7.66%) when applied to breeding trials, which could also be confirmed under realistic farming conditions. As a second step, an approach was developed that allows the assessment of lodging severity, information that is important to estimate yield impairment, which also takes the intensity of lodging events into account. Both parameters were tested on three ground sample distances. The lowest spatial resolution acquired from the highest flight altitude (100 m) still led to high accuracy, which increases the practicability of the method for large areas. Our new lodging assessment procedure can be used for insurance applications, precision farming, and selecting for genetic lines with greater lodging resistance in breeding research.

In viticulture, phenotypic data are traditionally collected directly in the field via visual and manual means by an experienced person. This approach is time consuming, subjective and prone to human errors. In recent years, research therefore has focused strongly on developing automated and non-invasive sensor-based methods to increase data acquisition speed, enhance measurement accuracy and objectivity and to reduce labor costs. While many 2D methods based on image processing have been proposed for field phenotyping, only a few 3D solutions are found in the literature. A track-driven vehicle consisting of a camera system, a real-time-kinematic GPS system for positioning, as well as hardware for vehicle control, image storage and acquisition is used to visually capture a whole vine row canopy with georeferenced RGB images. In the first post-processing step, these images were used within a multi-view-stereo software to reconstruct a textured 3D point cloud of the whole grapevine row. A classification algorithm is then used in the second step to automatically classify the raw point cloud data into the semantic plant components, grape bunches and canopy. In the third step, phenotypic data for the semantic objects is gathered using the classification results obtaining the quantity of grape bunches, berries and the berry diameter.