PFuji-Size dataset: A collection of images and photogrammetry-derived 3D point clouds with ground truth annotations for Fuji apple detection and size estimation in field conditions.

The PFuji-Size dataset is comprised of a collection of 3D point clouds of Fuji apple trees (Malus domestica Borkh. cv. Fuji) scanned at different maturity stages and annotated for fruit detection and size estimation. Structure-from-motion and multi-view stereo techniques were used to generate the 3D point clouds of 6 complete Fuji apple trees containing a total of 615 apples. The resulting point clouds were 3D segmented by identifying the 3D points corresponding to each apple (3D instance segmentation), obtaining a single point cloud for each apple. All segmented apples were labelled with ground truth diameter annotations. Since the data was acquired in field conditions and at different maturity stages, the set includes different fruit diameters -from 26.9 mm to 94.8 mm- and different fruit occlusion percentages due to foliage. In addition, 25 apples were photographed 360° in laboratory conditions, obtaining high resolution 3D point clouds of this sub-set. To the best of the authors' knowledge, this is the first publicly available dataset for apple size estimation in field conditions. This dataset was used to evaluate different fruit size estimation methods in the research article titled "In-field apple size estimation using photogrammetry-derived 3D point clouds: comparison of 4 different methods considering fruit occlusion" (Gené-Mola et al., 2021).

Assessing the Performance of RGB-D Sensors for 3D Fruit Crop Canopy Characterization under Different Operating and Lighting Conditions.

The use of 3D sensors combined with appropriate data processing and analysis has provided tools to optimise agricultural management through the application of precision agriculture. The recent development of low-cost RGB-Depth cameras has presented an opportunity to introduce 3D sensors into the agricultural community. However, due to the sensitivity of these sensors to highly illuminated environments, it is necessary to know under which conditions RGB-D sensors are capable of operating. This work presents a methodology to evaluate the performance of RGB-D sensors under different lighting and distance conditions, considering both geometrical and spectral (colour and NIR) features. The methodology was applied to evaluate the performance of the Microsoft Kinect v2 sensor in an apple orchard. The results show that sensor resolution and precision decreased significantly under middle to high ambient illuminance (>2000 lx). However, this effect was minimised when measurements were conducted closer to the target. In contrast, illuminance levels below 50 lx affected the quality of colour data and may require the use of artificial lighting. The methodology was useful for characterizing sensor performance throughout the full range of ambient conditions in commercial orchards. Although Kinect v2 was originally developed for indoor conditions, it performed well under a range of outdoor conditions.

LFuji-air dataset: Annotated 3D LiDAR point clouds of Fuji apple trees for fruit detection scanned under different forced air flow conditions.

This article presents the LFuji-air dataset, which contains LiDAR based point clouds of 11 Fuji apples trees and the corresponding apples location ground truth. A mobile terrestrial laser scanner (MTLS) comprised of a LiDAR sensor and a real-time kinematics global navigation satellite system was used to acquire the data. The MTLS was mounted on an air-assisted sprayer used to generate different air flow conditions. A total of 8 scans per tree were performed, including scans from different LiDAR sensor positions (multi-view approach) and under different air flow conditions. These variability of the scanning conditions allows to use the LFuji-air dataset not only for training and testing new fruit detection algorithms, but also to study the usefulness of the multi-view approach and the application of forced air flow to reduce the number of fruit occlusions. The data provided in this article is related to the research article entitled "Fruit detection, yield prediction and canopy geometric characterization using LiDAR with forced air flow" [1].

Application of light detection and ranging and ultrasonic sensors to high-throughput phenotyping and precision horticulture: current status and challenges.

Ultrasonic and light detection and ranging (LiDAR) sensors have been some of the most deeply investigated sensing technologies within the scope of digital horticulture. They can accurately estimate geometrical and structural parameters of the tree canopies providing input information for high-throughput phenotyping and precision horticulture. A review was conducted in order to describe how these technologies evolved and identify the main investigated topics, applications, and key points for future investigations in horticulture science. Most research efforts have been focused on the development of data acquisition systems, data processing, and high-resolution 3D modeling to derive structural tree parameters such as canopy volume and leaf area. Reported applications of such sensors for precision horticulture were restricted to real-time variable-rate solutions where ultrasonic or LiDAR sensors were tested to adjust plant protection product or fertilizer dose rates according to the tree volume variability. More studies exploring other applications in site-specific management are encouraged; some that integrates canopy sensing data with other sources of information collected at the within-grove scale (e.g., digital elevation models, soil type maps, historical yield maps, etc.). Highly accurate 3D tree models derived from LiDAR scanning demonstrate their great potential for tree phenotyping. However, the technology has not been widely adopted by researchers to evaluate the performance of new plant varieties or the outcomes from different management practices. Commercial solutions for tree scanning of whole groves, orchards, and nurseries would promote such adoption and facilitate more applied research in plant phenotyping and precision horticulture.

Spatial variability in orchards after land transformation: Consequences for precision agriculture practices.

The change from traditional to a more mechanized and technical agriculture has involved, in many cases, land transformations. This has supposed alteration of landforms and soils, with significant consequences. The effects of induced soil variability and the subsequent implications in site-specific crop management have not been sufficiently studied. The present work investigated the application of a resistivity soil sensor (Veris 3100), to map the apparent electrical conductivity (ECa), and detailed multispectral airborne images to analyse soil and crop spatial variability to assist in site-specific orchard management. The study was carried out in a peach orchard (Prunus persica (L.) Stokes), in an area transformed in the 1980 decade to change from rainfed arable crops to irrigated orchards. A total of 40 soil samples at two depths (0-30cm and 30-60cm) were analysed and compared to ECa and the normalised difference vegetation index (NDVI). Two types of statistical analysis were performed between ECa or NDVI classes with soil properties: a linear correlation analysis and multivariate analysis of variance (MANOVA). The results showed that the land transformation altered the spatial distribution and continuity of soil properties. Although a relationship between ECa and peach tree vigour could be expected, it was not found, even in the case of trees planted in soils with salts content above the tolerance threshold. Two types of management zones were proposed: a) zones delineated according to ECa classes to leach salts in the high ECa zones, and b) zones delineated according to NDVI classes to regulate tree vigour and yield. These strategies respond to the alteration of the original soil functions due to the land transformation carried out in previous years.

Mapping Vineyard Leaf Area Using Mobile Terrestrial Laser Scanners: Should Rows be Scanned On-the-Go or Discontinuously Sampled?

The leaf area index (LAI) is defined as the one-side leaf area per unit ground area, and is probably the most widely used index to characterize grapevine vigor. However, LAI varies spatially within vineyard plots. Mapping and quantifying this variability is very important for improving management decisions and agricultural practices. In this study, a mobile terrestrial laser scanner (MTLS) was used to map the LAI of a vineyard, and then to examine how different scanning methods (on-the-go or discontinuous systematic sampling) may affect the reliability of the resulting raster maps. The use of the MTLS allows calculating the enveloping vegetative area of the canopy, which is the sum of the leaf wall areas for both sides of the row (excluding gaps) and the projected upper area. Obtaining the enveloping areas requires scanning from both sides one meter length section along the row at each systematic sampling point. By converting the enveloping areas into LAI values, a raster map of the latter can be obtained by spatial interpolation (kriging). However, the user can opt for scanning on-the-go in a continuous way and compute 1-m LAI values along the rows, or instead, perform the scanning at discontinuous systematic sampling within the plot. An analysis of correlation between maps indicated that MTLS can be used discontinuously in specific sampling sections separated by up to 15 m along the rows. This capability significantly reduces the amount of data to be acquired at field level, the data storage capacity and the processing power of computers.

Georeferenced scanning system to estimate the leaf wall area in tree crops.

This paper presents the use of a terrestrial light detection and ranging (LiDAR) system to scan the vegetation of tree crops to estimate the so-called pixelated leaf wall area (PLWA). Scanning rows laterally and considering only the half-canopy vegetation to the line of the trunks, PLWA refers to the vertical projected area without gaps detected by LiDAR. As defined, PLWA may be different depending on the side from which the LiDAR is applied. The system is completed by a real-time kinematic global positioning system (RTK-GPS) sensor and an inertial measurement unit (IMU) sensor for positioning. At the end, a total leaf wall area (LWA) is computed and assigned to the X, Y position of each vertical scan. The final value of the area depends on the distance between two consecutive scans (or horizontal resolution), as well as the number of intercepted points within each scan, since PLWA is only computed when the laser beam detects vegetation. To verify system performance, tests were conducted related to the georeferencing task and synchronization problems between GPS time and central processing unit (CPU) time. Despite this, the overall accuracy of the system is generally acceptable. The Leaf Area Index (LAI) can then be estimated using PLWA as an explanatory variable in appropriate linear regression models.

Advanced technologies for the improvement of spray application techniques in spanish viticulture: an overview.

Spraying techniques have been undergoing continuous evolution in recent decades. This paper presents part of the research work carried out in Spain in the field of sensors for characterizing vineyard canopies and monitoring spray drift in order to improve vineyard spraying and make it more sustainable. Some methods and geostatistical procedures for mapping vineyard parameters are proposed, and the development of a variable rate sprayer is described. All these technologies are interesting in terms of adjusting the amount of pesticides applied to the target canopy.

Discriminating crop, weeds and soil surface with a terrestrial LIDAR sensor.

In this study, the evaluation of the accuracy and performance of a light detection and ranging (LIDAR) sensor for vegetation using distance and reflection measurements aiming to detect and discriminate maize plants and weeds from soil surface was done. The study continues a previous work carried out in a maize field in Spain with a LIDAR sensor using exclusively one index, the height profile. The current system uses a combination of the two mentioned indexes. The experiment was carried out in a maize field at growth stage 12-14, at 16 different locations selected to represent the widest possible density of three weeds: Echinochloa crus-galli (L.) P.Beauv., Lamium purpureum L., Galium aparine L.and Veronica persica Poir.. A terrestrial LIDAR sensor was mounted on a tripod pointing to the inter-row area, with its horizontal axis and the field of view pointing vertically downwards to the ground, scanning a vertical plane with the potential presence of vegetation. Immediately after the LIDAR data acquisition (distances and reflection measurements), actual heights of plants were estimated using an appropriate methodology. For that purpose, digital images were taken of each sampled area. Data showed a high correlation between LIDAR measured height and actual plant heights (R2 = 0.75). Binary logistic regression between weed presence/absence and the sensor readings (LIDAR height and reflection values) was used to validate the accuracy of the sensor. This permitted the discrimination of vegetation from the ground with an accuracy of up to 95%. In addition, a Canonical Discrimination Analysis (CDA) was able to discriminate mostly between soil and vegetation and, to a far lesser extent, between crop and weeds. The studied methodology arises as a good system for weed detection, which in combination with other principles, such as vision-based technologies, could improve the efficiency and accuracy of herbicide spraying.

Performance of an ultrasonic ranging sensor in apple tree canopies.

Electronic canopy characterization is an important issue in tree crop management. Ultrasonic and optical sensors are the most used for this purpose. The objective of this work was to assess the performance of an ultrasonic sensor under laboratory and field conditions in order to provide reliable estimations of distance measurements to apple tree canopies. To this purpose, a methodology has been designed to analyze sensor performance in relation to foliage ranging and to interferences with adjacent sensors when working simultaneously. Results show that the average error in distance measurement using the ultrasonic sensor in laboratory conditions is ±0.53 cm. However, the increase of variability in field conditions reduces the accuracy of this kind of sensors when estimating distances to canopies. The average error in such situations is ±5.11 cm. When analyzing interferences of adjacent sensors 30 cm apart, the average error is ±17.46 cm. When sensors are separated 60 cm, the average error is ±9.29 cm. The ultrasonic sensor tested has been proven to be suitable to estimate distances to the canopy in field conditions when sensors are 60 cm apart or more and could, therefore, be used in a system to estimate structural canopy parameters in precision horticulture.

Innovative LIDAR 3D Dynamic Measurement System to estimate fruit-tree leaf area.

In this work, a LIDAR-based 3D Dynamic Measurement System is presented and evaluated for the geometric characterization of tree crops. Using this measurement system, trees were scanned from two opposing sides to obtain two three-dimensional point clouds. After registration of the point clouds, a simple and easily obtainable parameter is the number of impacts received by the scanned vegetation. The work in this study is based on the hypothesis of the existence of a linear relationship between the number of impacts of the LIDAR sensor laser beam on the vegetation and the tree leaf area. Tests performed under laboratory conditions using an ornamental tree and, subsequently, in a pear tree orchard demonstrate the correct operation of the measurement system presented in this paper. The results from both the laboratory and field tests confirm the initial hypothesis and the 3D Dynamic Measurement System is validated in field operation. This opens the door to new lines of research centred on the geometric characterization of tree crops in the field of agriculture and, more specifically, in precision fruit growing.

Ultrasonic and LIDAR sensors for electronic canopy characterization in vineyards: advances to improve pesticide application methods.

Canopy characterization is a key factor to improve pesticide application methods in tree crops and vineyards. Development of quick, easy and efficient methods to determine the fundamental parameters used to characterize canopy structure is thus an important need. In this research the use of ultrasonic and LIDAR sensors have been compared with the traditional manual and destructive canopy measurement procedure. For both methods the values of key parameters such as crop height, crop width, crop volume or leaf area have been compared. Obtained results indicate that an ultrasonic sensor is an appropriate tool to determine the average canopy characteristics, while a LIDAR sensor provides more accuracy and detailed information about the canopy. Good correlations have been obtained between crop volume (C(VU)) values measured with ultrasonic sensors and leaf area index, LAI (R(2) = 0.51). A good correlation has also been obtained between the canopy volume measured with ultrasonic and LIDAR sensors (R(2) = 0.52). Laser measurements of crop height (C(HL)) allow one to accurately predict the canopy volume. The proposed new technologies seems very appropriate as complementary tools to improve the efficiency of pesticide applications, although further improvements are still needed.