Search

Highlights • A Grapevine yellows (GY) disease detection system was developed. • It can distinguish GY disease from other diseases with similar symptoms. • Six neural network architectures were utilized and evaluated. • This automatic tool is crucial to avoid missing GY-positive plants. • It offers an end-to-end detection of GY. Abstract Grapevine yellows (GY) are a significant threat to grapes due to the severe symptoms and lack of treatments. Conventional diagnosis of the phytoplasmas associated to GYs relies on symptom identification, due to sensitivity limits of diagnostic tools (e.g. real time PCR) in asymptomatic vines, where the low concentration of the pathogen or its erratic distribution can lead to a high rate of false-negatives. GY's primary symptoms are leaf discoloration and irregular wood ripening, which can be easily confused for symptoms of other diseases making recognition a difficult task. Herein, we present a novel system, utilizing convolutional neural networks, for end-to-end detection of GY in red grape vine (cv. Sangiovese), using color images of leaf clippings. The diagnostic test detailed in this work does not require the user to be an expert at identifying GY. Data augmentation strategies make the system robust to alignment errors during data capture. When applied to the task of recognizing GY from digital images of leaf clippings—amongst many other diseases and a healthy control—the system has a sensitivity of 98.96% and a specificity of 99.40%. Deep learning has 35.97% and 9.88% better predictive value (PPV) when recognizing GY from sight, than a baseline system without deep learning and trained humans respectively. We evaluate six neural network architectures: AlexNet, GoogLeNet, Inception v3, ResNet-50, ResNet-101 and SqueezeNet. We find ResNet-50 to be the best compromise of accuracy and training cost. The trained neural networks, code to reproduce the experiments, and data of leaf clipping images are available on the internet. This work will advance the frontier of GY detection by improving detection speed, enabling a more effective response to the disease. [ABSTRACT FROM AUTHOR]

Highlights • A Laurel Wilt detection disease system was developed. • It can distinguish Laurel Wilt disease from other disorders with similar symptoms. • Early disease detection was achieved utilizing hyperspectral data. • Spectral signature reflectance and vegetation indices were used for disease detection. Abstract Laurel wilt (Lw) disease is an exotic and lethal disease that can kill laurel family trees very fast. It is vectored by the redbay ambrosia beetle that prefers to live and lay eggs inside avocado trees (among other plants). Lw disease continues to expand in Florida posing a major threat to the avocado industry. Early and accurate disease detection is very critical in this case to remove infected trees and distinguish Lw disease from other diseases or disorders with similar symptoms. Herein, we present a nondestructive remote sensing method to detect Lw-infected avocado trees (in early and late stage) and discriminate them from healthy and other factors that cause similar symptoms, such as iron and nitrogen deficiencies, by using a portable spectral data collection system (visible – near infrared; 400–970 nm). Two data sets were collected in 10 nm and 40 nm spectral resolution, and 23 vegetation indices (VIs) were calculated to detect Lw-affected trees by using two classification methods: decision tree (DT) and multilayer perceptron (MLP) neural networks. Additionally, the optimal wavelengths and VIs to discriminate healthy, Lw-infected and avocado trees with iron and nitrogen deficiencies were identified. The results showed that it was possible to detect Lw-infected trees at early stage and distinguish them from other biotic and abiotic factors with high accuracy (around 100%) using the MLP method. Poorer results were achieved with DTs. The optimum 10 nm wide bands and VIs selected for the Lw-detection were found in the red, red-edge and NIR bands. [ABSTRACT FROM AUTHOR]

• A feature extraction method for identifying plant diseases was developed. • Karhunen-Loeve Expansion was used to identify patterns in spectral data. • This method can define a reduced spectral dataset for early-stage disease detection. • It can be used in classical and deep-learning methods for classifying diseases. This study introduces a novel method using reduced datasets for extracting plant reflectance signatures. Plant reflectance data from hyperspectral imaging sensors were used to develop biomarker signatures for plant identification and early disease detection purposes. The Karhunen-Loeve Expansion (KLE) of spectral reflectance data, taken from healthy and diseased plants (two case studies of citrus canker in citrus and Laurel wilt in avocado), were used to identify a basis set of functions that represents the distribution of the reflected signal energy. By spectral decomposition, the eigenvalues were related to the KLE basis set. The eigenvalues can be used to identify the KLE eigenvectors, which comprise the highest variation in the data. These components can be interpreted as the weighted variables which carry with them most of the information on the reflectance spectrum of the plant. From indications presented by this multivariate KLE analysis, a frequency reconstruction was adapted to convert the eigenvector information to a wave function. This reconstruction via KLE and frequency transformation formed the signature identification process (i.e., unique biomarker). These frequency spectra can be used as average signature reflectance patterns for plant identification, classification, and disease biomarkers. Defining these spectral identification biomarkers or signatures is purposeful since it could lead to less invasive classification and disease diagnostics techniques. The techniques used to determine these reflectance spectra require a unique spectrum reconstruction method, using Fourier transform methods and KLE for data reduction. [ABSTRACT FROM AUTHOR]

• This study provides a rapid and low-cost approach for strawberry bruise inspection. • Two lighting conditions were applied respectively for image acquisition. • Four bruise severities were classified based on the Bruise Ratio. • UV light was more useful in classifying strawberry bruise severities. • The AI-based strawberry bruise detection system can be applied for commercial use. Bruising is one of the major defects of strawberries. Strawberry bruised areas are often visible as flattened, sunken, and discolored. Postharvest strawberries can be bruised by compression, impact, or vibration forces during harvesting, transportation, and packinghouse operations. These bruises make strawberries more vulnerable to rot or disease, thus shortening their shelf life. Bruises can also cause consumers to abandon purchase decisions. Therefore, inspecting strawberry bruises is an important procedure for commercial strawberry farms to guarantee the high quality of strawberries sent to the market. A novel technique was developed in this study to detect bruises on strawberry images captured by a color camera under incandescent and ultraviolet (UV) light. Mask Region-Based Convolutional Neural Networks (Mask R-CNN), a deep learning method, was utilized for the strawberry bruise detection. Four bruise severities were classified according to the estimated bruise ratio of every single strawberry (Minor Bruise, Light Bruise, Moderate Bruise, and Severe Bruise). The results showed that Mask R-CNN could accurately detect whether a strawberry is bruised or not under both lighting conditions. However, the strawberry bruise severity classification accuracy of the proposed machine vision technique was improved significantly under UV light compared to incandescent light. For whole strawberry bruise detection under UV light, the F1 score for detecting strawberry bruises was 0.99, and the highest F1 score for whole strawberry bruise severity classification was 0.92 for Minor Bruise. This strawberry bruise detection system can assist farmers in detecting bruised strawberries on the processing line or an automated strawberry harvester, improve the quality of strawberries that are sent to the market, and avoid shortening the shelf life of fresh strawberries. [ABSTRACT FROM AUTHOR]

The harvest process for specialty crops is generally one of intensive activity because many people are required for harvest and packing, and the harvest window is brief due to the high perishability of the produce. Herein we present a cloud-based Harvest Management Information System (HMIS) that combines a novel real-time Portable Labor Monitoring System (PLMS) with a cloud-based harvest management software. The PLMS comprised of three key elements (1) a self-leveling scale, (2) electronic control box, and (3) a frame that supports all hardware. The electronic control box includes: (i) a RFID reader, (ii) a LCD display, (iii) a thermal printer, (iv) a GPS module, and (v) a communication system. RFID tags, containing unique ID numbers, embedded within rubber wrist bands, are worn by pickers. This system can read a picker’s ID (RFID bracelet), measure the weight of fruit, and record the time and location (optional) of every fruit ‘transaction’ (i.e., every time a picker brings a bucket of fruit to the collection bin). The collected data can be transmitted wirelessly to the server in real-time. The cloud-based software receives and processes the PLMS data on labor activities, visualizes the collected data, and can extract the data necessary for management information and automated filling of documents (e.g. payroll, yield maps). The HMIS is unique in its ability to: (1) accurately credit pickers for the fruit they have harvested in the field without impeding or altering the harvest process, (2) streamline data entry to payroll, (3) provide real-time tracking of harvest, yield mapping, and traceability, and, (4) generate precise and reliable harvest efficiency data. This integrated system was evaluated in sweet cherry, blueberry and apple orchards in Washington, USA. The weight of harvested fruit, time and location of every fruit drop were calculated accurately; all the data were transmitted wirelessly to the server and no errors were recorded. [ABSTRACT FROM AUTHOR]

Abstract: This paper introduces a real-time labor monitoring system (LMS) with the ability to track and record individual picker efficiency during manual harvest of specialty crops. This system utilizes existing commercial harvest equipment and integrates a digital weighing scale, RFID reader, computational unit, and a portable datalogger carried by pickers. The RFID reader, digital scale and computational unit are assembled on a common portable chassis. As pickers transfer fruit into a standard collection bin, the system reads the picker’s ID (RFID tag) and the weight of fruit. Weight data can then be transmitted wirelessly to the picker’s datalogger which records and displays the incremental and total weight of harvested fruit. An algorithm was developed in Matlab® to record, process and store the data, as well as to transmit wirelessly the weight value to the wearable datalogger. System prototypes were assembled and field-tested for accuracy and reliability during commercial harvest of sweet cherries (Prunus avium L.) in the Pacific Northwest. The LMS reliably calculated the harvest rate, picking cost, weight of harvested fruit, number of harvested buckets, range in fruit weight per bucket, and mean fruit weight per bucket, in real time. The mean harvest rate (±standard error) in a ‘Chelan’/Mazzard sweet cherry orchard trained to a steep leader architecture was 0.53±0.13kg/person/min. Harvest rate was similar for the same genotype trained to a steep leader (3 leaders) at 0.50±0.10kg/person/min and for harvest of ‘Tieton’/‘Gisela®6’ trees trained to a central leader architecture (0.53±0.15kg/person/min). When surveyed after using the LMS, every picker indicated a preference for knowing the exact weight of fruit they harvested compared with the current system of reimbursement per bucket or bin. Using the LMS, reliable data on worker efficiency can be collected with minimal interference with standard commercial practices. [Copyright &y& Elsevier]

• A novel wearable technology was developed for horse foraging behavior analysis. • It utilizes sound recognition techniques and a recurrent neural network (RNN). • It can accurately detect and distinguish bite and chew events in horses. • It can provide valuable information to predict pasture use and animal performance. Wearable sensing technologies can be used for precision livestock production and to study foraging strategies to better understand the relationships between herbivores, vegetation, and landscape. In this context, monitoring grazing behavior (i.e. chew and bite events) can provide critical information for livestock management. This study presents a computational tool that utilizes wearable sensing and deep learning to distinguish chew and bite events in horses. A micro camera equipped with a microphone (0–18 kHz) was used to obtain video/audio data from horses during grazing. The collected audio data were treated in a pre-processing filtering step, then used to train a recurrent neural network (RNN) with a long short-term memory (LSTM) layer to detect and distinguish chews, bites, and noise events. A post-processing sliding window technique was used to filter events with low confidence levels and lengths. Initial evaluation of this system showed an accuracy of 88.64% for bite identification and an accuracy of 94.13% for chew identification. The distinction between events and evaluation of responses to different pasture species and structure can provide useful information on the plant-animal interface. That is aligned to information such as bite rate, bite mass, and grazing time, and can help determine management strategies that optimize intake and provide data for modeling foraging behavior to predict pasture use and animal performance. [ABSTRACT FROM AUTHOR]

• An automated system was developed for measuring pecan nut growth. • It utilizes machine vision and deep learning convolutional neural networks. • It provides rapid and precise measurements of different pecan nut parts throughout the season. • It can be used for more precise management of irrigation, pesticide, and harvest schedules. Knowing pecan nut growth dynamic and its thermal requirements from bloom to ripening are essential to predict nut size and harvest date. However, monitoring of pecan growth (shuck, shell, and embryo) is a time consuming manual task, requiring multiple measurements for each nut. This process also ignores any deformation of the nut. A computer vision system can facilitate such a task by performing pixel-by-pixel semantic segmentation for all parts of the nut. This study uses the Mask-RCNN algorithm for object detection and semantic segmentation for area estimation of shuck, shell, and embryo on pecans in multiple growth stages. The dataset was divided in two stages: small (young) and big (older) pecans. The network chosen achieved an F1 score of 95.3% to 100% on the models for the object detection task. The area estimation achieved a mean absolute percentage error of 10.14% to 28.06% on these tasks. The resulting growth plot presents a sigmoid curve for all parts of the nut, which demonstrates the precision of the measurement. This automated measuring system makes it possible to have a multidimensional understanding of each aspect of nut development, and gives us the ability of modeling nut growth phenology and predicting nut productions. [ABSTRACT FROM AUTHOR]

• A cloud- and AI-based application was developed to analyze and visualize UAV-collected data. • This interactive and user-friendly application can automatically create tree inventories and assess plant characteristics. • It simplifies the data processing and visualization procedures and promotes adoption of UAVs. • It can provide a low-cost, consistent, and rapid method for field survey and plant phenotyping. Traditional sensing technologies in specialty crops production, for pest and disease detection and field phenotyping, rely on manual sampling and are time consuming and labor intensive. Since availability of personnel trained for field scouting is a major problem, small Unmanned Aerial Vehicles (UAVs) equipped with various sensors can simplify the surveying procedure, decrease data collection time, and reduce cost. To accurate and rapidly process, analyze and visualize data collected from UAVs and other platforms (e.g. small airplanes, satellites, ground platforms), a cloud and artificial intelligence (AI) based application (named Agroview) was developed. This interactive and user-friendly application can: (i) detect, count and geo-locate plants and plant gaps (locations with dead or no plants); (ii) measure plant height and canopy size (plant inventory); (iii) develop plant health (or stress) maps. In this study, the use of this Agroview application to evaluate phenotypic characteristics of citrus trees (as a case study) is presented. It was found, that this emerging technology detected citrus trees with mean absolute percentage error (MAPE) of 2.3% in a commercial citrus orchard with 175,977 trees (1,871 acres; 39 normal and high-density spacing blocks). Furthermore, it accurately estimated tree height with 4.5% and 12.93% MAPE for normal and high-density spacing respectively, and canopy size with MAPE of 12.9% and 34.6% for normal and high-density spacing respectively. It provides a consistent, more direct, cost-effective and rapid method for field survey and plant phenotyping. [ABSTRACT FROM AUTHOR]

• A UAV-based high-throughput technique was developed for citrus tree. • Phenotypic characteristics of sweet orange trees grafted on 25 rootstock were evaluated. • Data collected by UAV were correlated significantly with manually collected data. • This cost-efficient tool can be used for monitoring phenotypic changes of plants at spatial and temporal resolution. • It provides an objective, accurate, and cost-efficient approach for plant monitoring and health assessment. The implementation of breeding methods requires the creation of a large and genetically diverse training population. Large-scale experiments are needed for the rapid acquisition of phenotypic data to explore the correlation between genomic and phenotypic information. Traditional sensing technologies for field surveys and field phenotyping rely on manual sampling and are time consuming and labor intensive. Since availability of personnel trained for phenotyping is a major problem, small UAVs (unmanned aerial vehicles) equipped with various sensors can simplify the surveying procedure, decrease data collection time, and reduce cost. In this study, we evaluated the phenotypic characteristics of sweet orange trees grafted on 25 rootstock cultivars with different influences on plant growth and productivity utilizing a UAV-based high throughput phenotyping system. Data collected by UAV were compared with data collected manually according to standard horticultural procedures. The UAV-based technique was able to detect and count citrus trees with high precision (99.9%) in an orchard of 4931 trees and estimate tree canopy size with a high correlation (R = 0.84) with the manual collected data. No correlation of UAV-based data and manually collected data was observed for yield. The reason for the observed deviation is the influence of different rootstock cultivars on yield efficiency. Despite the low vigor-inducing effect of some rootstocks, they are highly productive, whilst others are high in vigor but produce less fruit. Our study demonstrates the high accuracy of the UAV technique to assess tree size. When using these techniques, it is essential to recognize the limitations imposed by the biological system. [ABSTRACT FROM AUTHOR]

Highlights • A smart sprayer for precision pest management was developed. • This low-cost technology can distinguish target weeds from non-target objects. • It sprays only on a selected target (weed). • Neural networks were utilizing for target detection and classification. • It can significantly reduce the quantity of agrochemicals applied. Abstract Most conventional sprayers apply agrochemicals uniformly, despite the fact that distribution of weeds is typically patchy, resulting in wastage of valuable compounds, increased costs, crop damage risk, pest resistance to chemicals, environmental pollution and contamination of products. To reduce these negative impacts, a smart sprayer was designed and developed utilizing machine vision and artificial intelligence to distinguish target weeds from non-target objects (e.g. vegetable crops) and precisely spray on the desired target/location. Two different experimental scenarios were designed to simulate a vegetable field and to evaluate the smart sprayer's performance. The first scenario contained artificial weeds (targets) and artificial plants (non-targets). The second and most challenging scenario contained real plants; portulaca weeds as targets, and sedge weeds and pepper plants as non-targets. Two different embedded graphics processing unit (GPU) were evaluated as the smart sprayer processing unit (for image processing and target detection). The more powerful GPU (NVIDIA GTX 1070 Ti) achieved an overall precision of 71% and recall of 78% (for plant detection and target spraying accuracy) on the most challenging scenario with real plants, and 91% accuracy and recall for the first scenario with artificial plants. The less powerful GPU (NVIDIA Jetson TX2) achieved an overall precision and recall of 90% and 89% respectively on the first scenario with artificial plants, and 59% and 44% respectively on the second scenario with real plants. Finally, an RTK GPS was connected to the smart sprayer and an algorithm was developed to automatically generate weed maps and visualize the collected data (after every application). This smart technology integrates a state of the art (AI-based) weed detection system, a novel fast and precision spraying system, and a weed mapping system. It can significantly reduce the quantity of agrochemicals required, especially compared with traditional broadcast sprayers that usually treat the entire field, resulting in unnecessary application to areas that do not require treatment. It could also reduce costs, risk of crop damage and excess herbicide residue, as well as potentially reduce environmental impact. [ABSTRACT FROM AUTHOR]

Precision agriculture has emerged as a dominant force in the United States, with widespread adoption of advanced technologies and decision support systems (DSS) since the 1980s. Key tools such as variable rate application (VRA), autopilot systems, and remote sensing have become integral for U.S. farmers, offering invaluable insights from crop, soil, and weather information to optimize agricultural production while minimizing environmental impact. To synthesize and categorize the extensive research available on precision agriculture, a systematic review protocol has been designed. Our objective is to offer clear and authoritative insights into the nature, scope, and volume of this field. Implementing a rigorous search strategy, we utilized renowned databases such as Scopus® and Web of ScienceTM to gather relevant and significant materiality. The retrieval process involved the use of indexing terms and Boolean operators, with a focus on 'precision agriculture' and 'precision farming', striking a balance between specificity and comprehensiveness. To ensure the credibility of our findings, only peer-reviewed papers authored by individuals affiliated with U.S. institutions have been included. Expert reviewers with deep knowledge in the field independently assessed the selected papers, thoroughly evaluating titles, abstracts, keywords, methods, conclusions, and declarations. Consistency and eligibility were paramount in determining which papers met the criteria for inclusion. Any discrepancies or disagreements were resolved through rigorous consensus-building discussions among the reviewers. Through this comprehensive meta-review, we provide a scientific contribution that enhances our understanding of precision agriculture, highlighting focus areas for further research and development (R&D). By synthesizing and categorizing the existing literature, we offer authoritative insights into the research landscape, informing future investigations and fostering innovation. Focusing specifically on the U.S., we shed light on the unique aspects and pioneering advancements in precision agriculture within the country. Ultimately, our findings have the potential to drive progress, contributing to sustainable development, increased productivity, enhanced environmental sustainability, and responsible agricultural practices. [ABSTRACT FROM AUTHOR]

• A smart sensing system for tree crop sprayers was developed and evaluated. • It utilizes sensor fusion (e.g., LiDAR, machine vision, GPS) and artificial intelligence. • The low-cost prototype was evaluated in citrus orchards. • It was able to detect trees and spray based on tree size and leaf density. • It reduced spraying volume by 28% compared to conventional spraying. Delivering the appropriate amount of chemicals for each plant based on their needs is essential for reducing costs, waste, and labor. Smart agricultural machinery can provide variable-rate technologies using artificial intelligence (AI) and machine vision to optimize spraying applications. In this research, a low-cost smart sensing system for controlling airblast tree crop sprayers was designed and evaluated using citrus as a case study. The prototype comprised a LiDAR, machine vision, GPS, flow meters, sensor fusion, and AI to scan trees for tree height, tree classification, and fruit counting. Specifically, this smart sensing system can detect and classify objects to tree or non-tree (e.g., human, field constructions), measure tree height and canopy density, and detect and count fruit. Based on this information, it controls spraying nozzles to optimize spraying applications. A novel software was written in C++ and ran on an Nvidia Jetson Xavier NX embedded computer to process and control the data utilizing data fusion and AI techniques. The smart sensing system's results for tree height estimate indicated a relatively low average error of 6%. A convolutional neural network (CNN) was used to perform tree classification with an average accuracy of 84% in classifying the collected imagery into mature, young, dead, and non-tree objects. The fruit count module (also a CNN) had an F1 score of 89%, compared to ground-truth labeled images of mature and immature citrus fruits. Finally, the adoption of this new sensing method reduced spraying volume by 28%, compared to traditional spraying applications. [ABSTRACT FROM AUTHOR]

• An AI-based smartphone application was developed for pest detection in real-time. • A database was developed for pest data management. • A spatial interpolation method was used to generate spatial distribution maps for site-specific pest management. • This smartphone application worked faster to achieve higher pest detection accuracy compared with the traditional method. Pests can cause huge yield loss in crop production without timely and effective monitoring methods. The traditional method of manually counting pests is time-consuming and causes bias between different observers. Therefore, this study developed an Android-based smartphone application for automated pest counting using a deep learning model (YOLOv7 tiny). This study evaluated the smartphone application using twospotted spider mite (TSSM) detection as a case study and compared it with the manual counting method. The smartphone application records the total number of pests and the GPS coordinates into a MySQL database. The spatial interpolation method was used to generate a spatial distribution map of TSSM in real-time, which could guide growers to apply the right amount of pesticide at the right location. To train the deep learning model for TSSM detection, 3,348 images were collected using three smartphones. The YOLOv7 tiny model was trained at three input image sizes (160x160, 320x320, and 640x640 pixels). The model trained at 320x320 pixels was chosen to deploy on the smartphone application because of the superior performance for both detection accuracy and inference speed. The Android-based smartphone application was installed on two widely used smartphones (Samsung Galaxy S10e and LG Stylo 6) for field testing. The counting accuracy of TSSM motiles (all life stages except eggs) and TSSM eggs using the smartphone application was 78.1% and 75.3%, respectively. The average counting speed of the smartphone application was 233 s per strawberry (trifoliate) leaf. The smartphone application had higher counting accuracy (7.1% higher for TSSM motiles and 12.7% higher for TSSM eggs) and speed (29 s faster per leaf) than manual counting using a magnifying lens. The quick assessment of field conditions with smartphone technology can allow growers to implement quick management tactics and potentially reduce the amount of pesticides to manage TSSM, which can save growers money and protect the environment. [ABSTRACT FROM AUTHOR]

• A new visible NDVI (vNDVI) was developed using genetic algorithms. • It was evaluated in three different crops (citrus, grapes, and sugarcane). • The vNDVI accurately estimates NDVI values from RGB data. • It can provide a low-cost alternative for plant phenotyping. Several vegetation indices have been developed, with the normalized difference vegetation index (NDVI) been the most studied and commonly used. To generate an NDVI map, a relatively high-cost multispectral sensor is required; but currently, most UAVs are equipped with low-cost RGB cameras. For that reason, other indices that utilize RGB data have been developed to generate maps similar to NDVI and minimize the data acquisition cost, such as the triangular greenness index (TGI) and the visible atmospheric resistant index (VARI). However, several studies found that these indices cannot be recommended as reliable general-purpose crop health indicators. This study utilizes a genetic algorithm to develop a new visible index (visible NDVI; vNDVI) that estimates NDVI values of vegetation from uncalibrated RGB cameras mounted on UAVs (or other remote sensing platforms). Three experiments were conducted to create and validate the proposed index. First, the NDVI values generated from a multispectral camera were compared with the NDVI values generated by a hyperspectral camera. In the second experiment, the vNDVI formula was created using a genetic algorithm. The third experiment validates the proposed vNDVI, generated from two uncalibrated RGB cameras, in three different crops (citrus, grapes, and sugarcane). The proposed vNDVI proved to be highly accurate on estimating NDVI values by just using RGB cameras, with an overall mean percentage error of 6.89% and a mean average error of 0.052 in all three crops, providing a low-cost alternative for remote sensing and plant phenotyping. [ABSTRACT FROM AUTHOR]

• The use of Unmanned Aerial Vehicles (UAVs) and machine learning techniques for plant phenotyping has shown great potential for increasing the efficiency and accuracy of crop monitoring. • These technologies can provide high-resolution images and data that can be used to analyze plant traits and predict crop yield. • By automating the process of data collection and analysis, UAV-based imaging and machine learning techniques can reduce the time and labor required for traditional crop monitoring methods. • Furthermore, they can enable real-time decision making, which is critical for optimizing crop management strategies and achieving higher yields. The availability of timely and accurate information about plant conditions is critical for making informed decisions for maintaining or improving agricultural productivity. However, field data collection is often laborious and time-consuming, underscoring the need for accurate and efficient methods to monitor crop growth and development. Therefore, the objective of this study was to evaluate the effectiveness of combining unmanned aerial vehicle (UAV)-based imaging and machine learning (ML) techniques for monitoring sweet corn (Zea mays var. saccharata) height, biomass, and yield. The study was conducted at the Tropical Research and Education Center (TREC) during the winter (dry) season of 2020–2021 using 16 experimental plots planted with sweet corn. The treatments were set up in a completely randomized block design (RCBD) with four irrigation treatments of 25%, 50%, 75%, and 100% full irrigation with four replications each. Field data collection included plant height, fresh and dry biomass, and yield. In addition, UAV images were collected using the DJI Matrice 210 v2 UAV (SZ DJI Technology Co., Ltd., Shenzhen, China) equipped with a MicaSense RedEdge-MX multispectral sensor (MicaSense, Seattle, WA, USA). Image processing was done with Pix4Dmapper 4.7.5 (Pix4D S.A., Prilly, Switzerland). A crop surface model, representing estimated plant height (UAVH), was calculated based on pixel-to-pixel differences between digital surface and terrain models. A simple linear regression model was used to estimate sweet corn biomass and yield from UAV images estimated plant height (UAVH). In addition, two linear algorithms known as a linear model (LM) and Lasso and elastic-net regularized generalized linear model (GLMNET) and three non-linear ML algorithms including random forest (RF), support vector machine (SVM), and k-nearest neighbor (kNN) were used to predict plant height and biomass. These algorithms were chosen due to their reliable performance and ability to learn complex non-linear relationships. Eight vegetation indices with UAVH were also used to evaluate the ML models' performance to predict plant height and biomass. Results confirmed that UAV imaging could be effectively used to estimate plant height (d = 0.99, r2 = 0.99, and MAE = 5 cm). Estimated fresh and dry biomass from UAV imaging also had good agreements with measured data with r2 values of 0.75 and 0.70, respectively. A statistically significant linear correlation between measured fresh yield at harvest and UAVH was found with d, adjusted r2 , and MAE of 0.84, 0.66, and 67 g m−2, respectively. Evaluation of ML algorithms revealed that all models performed well for plant biomass estimation with d values between 0.88 and 0.99. However, the kNN and SVM outperformed all other models for biomass estimation. GMLNET performed better than other models for plant height estimation. Overall, results revealed that UAV imaging and ML models could be effectively used for monitoring plant phenotypic characteristics such as height, yield, and biomass. [ABSTRACT FROM AUTHOR]

• An automated system for monitoring ACP in groves was developed. • This low-cost technology can distinguish target insects from other insects and objects. • Neural networks were utilized for ACP detection and classification. • It develops a map to visualize the ACP numbers for each tree. • It can automate scouting procedures in citrus and to be extended to other crop insects. Specialty crop growers face challenges from numerous diseases and pests. For example, the Asian citrus psyllid (ACP) is a key pest of citrus due to its role as vector of huanglongbing (HLB) (greening disease). There is no known cure for HLB, but vector management is critical, both for slowing spread and attenuating symptoms in infected trees. Therefore, monitoring ACP population, as well as other pest populations, is an essential management component for timing and assessment of control actions. Manual crop scouting is often labor intensive and time consuming. In this work, an automated system was developed and evaluated utilizing machine vision and artificial intelligence to monitor ACP in groves. This system comprised a tapping mechanism to collect insects from the tree's branches and a board with a grid of cameras for image acquisition. A software was developed using two convolutional neural-networks to accurately detect and distinguish psyllids from other insects and debris fallen from the tree. A GPS was utilized to automatically record individual tree position to facilitate data assessment on large groves. A precision and recall of 80% and 95%, respectively, was obtained on detecting ACPs on a sample of 90 young citrus trees. The system proved a great potential to automate scouting procedures in citrus and to be extended to other crop insects. [ABSTRACT FROM AUTHOR]

Highlights • A remote sensing technique for Laurel Wilt detection in avocado was developed. • This low-cost technique can distinguish Laurel Wilt disease from other stressors. • Neural networks were utilizing for disease detection and classification. • Early disease detection was achieved utilizing multispectral imaging. Abstract Early and accurate disease detection is essential for implementing timely disease management practices. Current disease detection tactics, like visual detection through scouting, are labor intensive, expensive, requires a level of expertise in pest identification, and, may result in subjective disease identification. Diagnosis based on visual symptoms is often compromised by the inability to differentiate between similar symptoms caused by different biotic and abiotic factors. In this paper, an automated early disease detection technique for avocado trees is presented and evaluated. This remote sensing technique can detect an important avocado disease, the laurel wilt (Lw) disease, and differentiate it from healthy trees (H), trees infected by phytophthora root rot (Prr), and trees with iron (Fe) and nitrogen (N) deficiencies. Detection of Lw disease in avocado trees, in early stage, is very difficult, because it has similar symptoms with other stress factors such as nutrient deficiency, salt damage, phytophthora root rot, etc. The proposed disease detection procedure contains several steps including image acquisition, image pre-processing, image segmentation, feature extraction and classification. For image acquisition, two cameras were utilized and evaluated: (i) a Tetracamera (6 bands Tetracam) and (ii) a modified Canon camera (3 bands); and two classification methods were studied: (a) neural network multilayer perceptron (MLP), and (ii) K- nearest neighbors, to detect Lw in asymptomatic stage and in late (symptomatic) stage. Additionally, two segmentation methods, region of interest (OVROI) and polygon region of interest (PROI), were utilized. The MLP classification method with the Tetracam was able to successfully detect Lw with an accuracy of 99% in asymptomatic (early) stage. Hence, low-cost remote technique can be utilized to differentiate healthy and unhealthy plants. [ABSTRACT FROM AUTHOR]