Spot Spraying Research Articles

Advancing precision agriculture: domain-specific augmentations and robustness testing for convolutional neural networks in precision spraying evaluation

AbstractModern agriculture relies heavily on the precise application of chemicals such as fertilisers, herbicides, and pesticides, which directly affect both crop yield and environmental footprint. Therefore, it is crucial to assess the accuracy of precision sprayers regarding the spatial location of spray deposits. However, there is currently no fully automated evaluation method for this. In this study, we collected a novel dataset from a precision spot spraying system to enable us to classify and detect spray deposits on target weeds and non-target crops. We employed multiple deep convolutional backbones for this task; subsequently, we have proposed a robustness testing methodology for evaluation purposes. We experimented with two novel data augmentation techniques: subtraction and thresholding which enhanced the classification accuracy and robustness of the developed models. On average, across nine different tests and four distinct convolutional neural networks, subtraction improves robustness by 50.83%, and thresholding increases by 42.26% from a baseline. Additionally, we have presented the results from a novel weakly supervised object detection task using our dataset, establishing a baseline Intersection over Union score of 42.78%. Our proposed pipeline includes an explainable artificial intelligence stage and provides insights not only into the spatial location of the spray deposits but also into the specific filtering methods within that spatial location utilised for classification.

Agronomic and Technical Evaluation of Herbicide Spot Spraying in Maize Based on High-Resolution Aerial Weed Maps-An On-Farm Trial.

Spot spraying can significantly reduce herbicide use while maintaining equal weed control efficacy as a broadcast application of herbicides. Several online spot-spraying systems have been developed, with sensors mounted on the sprayer or by recording the RTK-GNSS position of each crop seed. In this study, spot spraying was realized offline based on georeferenced unmanned aerial vehicle (UAV) images with high spatial resolution. Studies were conducted in four maize fields in Southwestern Germany in 2023. A randomized complete block design was used with seven treatments containing broadcast and spot applications of pre-emergence and post-emergence herbicides. Post-emergence herbicides were applied at 2-4-leaf and at 6-8-leaf stages of maize. Weed and crop density, weed control efficacy (WCE), crop losses, accuracy of weed classification in UAV images, herbicide savings and maize yield were measured and analyzed. On average, 94% of all weed plants were correctly identified in the UAV images with the automatic classifier. Spot-spraying achieved up to 86% WCE, which was equal to the broadcast herbicide treatment. Early spot spraying saved 47% of herbicides compared to the broadcast herbicide application. Maize yields in the spot-spraying plots were equal to the broadcast herbicide application plots. This study demonstrates that spot-spraying based on UAV weed maps is feasible and provides a significant reduction in herbicide use.

Development and evaluation of a machine vision and deep learning-based smart sprayer system for site-specific weed management in row crops: An edge computing approach

Traditional weed management often involves blanket herbicide spraying, resulting in substantial herbicide wastage, environmental concerns, and herbicide resistant issues. Smart spraying systems utilizing robotics and sensors technologies can minimize herbicide usage and provide a sustainable solution for site-specific weed management. A machine vision-based spraying system was designed and developed for weed identification and precise spray application onto the target weeds. The sprayer platform utilizes a deep learning YOLOv4 model to accurately recognize multiple weed species, facilitating targeted spray application. The platform is equipped with an FLIR RGB camera for real-time image acquisition and Nvidia Jetson AGX Orin edge device for deploying weed detection deep-learning model. The GPIO pins of Nvidia Jetson were utilized to activate relay, providing precise on/off control over the TeeJet solenoid valves for spot spraying. Both indoor and field experiments were conducted to evaluate and compare the performance of vision-based sprayer system for weed identification and precise spraying onto the target weeds. In the indoor experiment, the sprayer system showed the average effective spraying rate of 93.33 %, with the precision of 100 % and recall of 92.8 %. Conversely, the field experiment resulted in a slightly lower average effective spraying rate of 90.6 %, while maintaining a precision of 95.5 % and a recall of 89.47 %. The reduced accuracy of the spraying system in field experiment was due to varying outdoor conditions such as lighting, shadows, and wind velocity. Overall, the result of this study demonstrates the spraying system's potential for targeted herbicide application onto the grid cells containing weeds, effectively reducing herbicide usage and overall weed management costs.

Time-Efficient Low-Resolution RGB Aerial Imaging for Precision Mapping of Weed Types in Site-Specific Herbicide Application

An efficient method for weed detection and the precise generation of spraying maps is crucial to optimising weed management strategies and minimising the costs associated with herbicide use. This study presents a method that leverages low-resolution UAV images to accurately detect and map weeds in sugarcane fields. The proposed approach integrates RGB images, vegetation indices and the HSV colour space, enhancing segmentation through histogram equalization (HE) and object-based image analysis (OBIA). The models developed in this study demonstrate exceptional performance in weed detection, with the most suitable dataset, achieving a detection capability of 95.78% for Broad-leaved Weeds (BLW) and the highest accuracy of 94.45% for Narrow-leaved Weeds (NLW). When considering multiple targets such as BLW, NLW, soil and sugarcane, the models exhibit a detection accuracy of 89.69%. Furthermore, the precision spraying maps generated by the coverage model method (CM) demonstrate remarkable accuracy, reaching 97.50% for weed control using agricultural drones. This method offers an efficient, cost-effective, and timely solution for precise weed detection, leading to improved weed control outcomes by enabling the selection of appropriate chemical substances tailored to each weed species. It reduces repetitive spraying costs and minimises chemical usage through spot spraying.

Optimizing fungicide programs for peanut leaf spot management in Southeast Alabama: A case for reducing chlorothalonil use

Early leaf spot (caused by Passalora arachidicola) and late leaf spot (Nothopassalora personata) are the most damaging fungal diseases of peanuts (Arachis hypogaea L.) in the southeastern United States. Producers rely heavily on single-site fungicides, which pose a moderate to high risk for resistance development. To mitigate this risk, chlorothalonil, the predominate multi-site fungicide used to manage leaf spot diseases in peanuts, is the backbone of leaf spot spray programs. However, the future availability of chlorothalonil in agriculture is at risk due to regulatory decisions and toxicity issues. Thus, this study was undertaken to evaluate the efficacy of copper sulfate, dodine, and sulfur alone or in combination with other single-site fungicides against leaf spot diseases as chlorothalonil alternatives as well as their influence on yield. Under low disease pressure, all the alternative fungicide programs provided statistically similar leaf spot control when compared to chlorothalonil alone. In contrast, under high disease pressure in 2022, dodine alone and dodine + penthiopyrad did not significantly decrease leaf spot severity when compared to the nontreated control. Additionally, dodine alone and the 2022 premium fungicide program had significantly higher leaf spot severity when compared to chlorothalonil alone under favorable weather patterns for rapid disease development. Sulfur + copper sulfate significantly reduced leaf spot severity as compared to chlorothalonil alone only in 2022. All other remaining fungicide programs provided statistically similar control as chlorothalonil alone. Overall, dodine, sulfur, and copper sulfate have potential to serve as alternatives to chlorothalonil in leaf spot fungicide spray programs in peanuts.

Reducing environmental exposure to PPPs in super-high density olive orchards using UAV sprayers.

The increasing demand for optimizing the use of agricultural resources will require the adoption of cutting-edge technologies and precision farming management. Unmanned Aerial Vehicle (UAV) sprayers seem promising due to their potential to perform precision or spot spraying, particularly in woody crop environments where total surface spraying is unnecessary. However, incorporating this technology is limited by the lack of scientific knowledge about the environmental risks associated with UAV sprayers and the strict legal framework. Nonetheless, these spraying systems' characteristic downwash airflow and the limited swath width can potentially mitigate drift in hedgerow crops. During our study we performed comparative studies aimed to compare the airborne drift, soil, and crop depositions between a conventional orchard sprayer and a UAV sprayer in a commercial superhigh-density orchard in the South Iberian Peninsula in 2022. Our findings reveal that, in superhigh-density olive orchards, the UAV sprayer presents a substantial reduction in airborne drift, while soil depositions showed no significant differences compared to those of a conventional terrestrial orchard sprayer. Crop depositions were significantly lower when utilizing the UAV sprayer. These results suggest that introducing UAV spraying technology in Mediterranean agricultural systems, under specific scenarios, can effectively reduce the environmental impact of crop spraying and encourage the responsible use of plant protection products (PPPs).

Public acceptance and sustainability perceptions of food produced with chemical, digital and mechanical weed control measures

This study investigated public acceptance and perceptions of the sustainability of food produced with different weed control measures. An online survey with a within-subject design was conducted with 485 respondents from the French- and German-speaking parts of Switzerland. Acceptance of food produced using the investigated measures and social, economic and environmental sustainability perceptions were assessed for full-surface herbicide spraying, herbicide reduction (spot spraying, precise spraying) and herbicide-free mechanical technology (hoeing machine). The importance of food’s naturalness, chemophobia, perceptions of farmers and sociodemographic variables were also recorded. The results revealed changes in social, economic and environmental sustainability perceptions based on the weed control measures, indicating that laypeople assess the impacts of these measures based on their type (chemical, digital, mechanical). The amount of herbicide sprayed is relevant for people’s judgements, as well as the precise spraying that keeps crops intact. The use of hoeing machine is perceived to be the most sustainable, natural and acceptable compared to the other investigated measures. Overall, the findings suggest that communicating information on weed control measures, the quantity of herbicide applied and the precision of the spraying might help increase public acceptance of plant protection measures applied by farmers, promoting sustainability.

Assessing clethodim spot spraying applications for control of problematic weedy rice and other grasses in California rice fields

AbstractSpot spraying applications offer the opportunity to target specific weeds in a field, while simultaneously reducing herbicide usage and increasing the long‐term efficacy of chemical control options. The study is focused on controlling California weedy rice accessions (Oryza spp.) and problematic grass weeds with a spot spray application of clethodim in a flooded rice system. The efficacy of incorporating nonionic surfactant to clethodim applications was also assessed. Dose‐response experiments were carried out in a greenhouse on five weedy rice accessions, common grass rice weeds, and cultivated rice varieties L207, M105, M206, M209, M211, and S102 to determine the dose needed to affect these populations. Clethodim was applied in a field setting to assess spot spraying efficacy, the possibility of herbicide dispersion in the water, and crop injury. Clethodim successfully controlled weedy rice and grasses in the greenhouse. The effective rates to control 90% of the five test populations (ED90) were between 51 and 74 g ai ha−1 clethodim for weedy rice accessions. Adding nonionic surfactant to clethodim increased its efficacy by 1.6‐ to 1.9‐fold. Cultivated rice varieties did not exhibit any tolerance to clethodim, however, spot spraying applications at 150 g ai ha−1 clethodim did not cause any dispersion in the field. Clethodim spot spray application was effective both at the three‐ to four‐leaf growth stage and tillering growth stage for weedy rice.

Leaching of herbicides and their metabolites in lysimeters filled with soils from railway tracks

Glyphosate currently is the most-widely used herbicide for weed control on railway tracks. However, railway companies look for non-chemical and chemical alternatives. Glyphosate has a low potential for leaching to groundwater and alternative herbicides should ideally share this behavior. We investigated the leaching of glyphosate and five potential alternatives in lysimeters filled with three different soils from railway tracks. Herbicides were applied in three consecutive years to 4–7 lysimeters. Leachate was collected and analysed for herbicides and major metabolites. Highest annual average concentrations of 23 μg/L were found for atrazine, which was selected as mobile reference compound. 2,4-D, flazasulfuron, and its major metabolite DTPU reached annual average concentrations of 0.60, 0.43, and 0.50 μg/L, respectively. Pelargonic-acid was detected in single samples shortly after application and heavy rainfall. Quizalofop-P, glyphosate, and metabolites were only sporadically found in leachate at annual average concentrations ≪0.1 μg/L. Highest leachate concentrations were generally observed in lysimeters with a high proportion of sandy fine material and little organic carbon, except for the quickly-degradable pelargonic-acid and 2,4-D that reached highest concentrations in lysimeters with high gravel content and thus small effective transport volume. After three years, up to 30, 6.5, and 0.56 % of the applied mass of flazasulfuron, atrazine, and 2,4-D, respectively, was recovered in leachate in form of the active substances and/or their metabolites (expressed in parent equivalents). Although recoveries of flazasulfuron exceeded those of atrazine, concentrations were lower due to the 100-fold lower application rate. From the perspective of groundwater protection, the currently used glyphosate still is by far the best (chemical) option while all tested possible glyphosate alternatives did show some potential for leaching which may, however, be manageable with certain restrictions (e.g., spot spraying). These findings are likely applicable to other uses in non-agricultural settings with gravel or paving stone surfaces.

Ecological and economic evaluation of conventional and new weed control techniques in row crops

The Farm to Fork strategy of the European Union aims to reduce pesticide use. In weed control, chemical measures are to be replaced mainly by mechanical ones. However, a comprehensive evaluation of mechanical methods with respect to ecological and economic parameters is not yet available. The aim of the study was to quantify relevant ecological and economic parameters of mechanical, mechanical-chemical and chemical weed control in sugar beet as a model crop. An overall consideration of the multidimensional risks for different conventional and new methods revealed that currently no method can be appraised as fully environmentally friendly. Toxicological risk assessment using the model SYNOPS-GIS indicated that broadcast spraying of conventional herbicides is associated with low toxicological risks in most soil-climate-regions and years. Broadcast spraying with herbicides such as CONVISO ONE, while beneficial for energy use, climatic effects, and toxicity to soil organisms, has the highest acute aquatic toxicity. Conventional mechanical methods, while toxicologically safe and, according to field trials, without negative effects on soil biota, soil erosion, or sugar yield, are associated with 100–150% higher greenhouse gas emissions compared to conventional herbicide application. Overall, hoeing with a solar powered robot combined with herbicide spot spraying has the greatest advantages. This technique requires relatively low resource and energy consumption, produces about 5% lower greenhouse gas emissions and has by 87% lower toxicological risk than conventional herbicide spraying. Among pure mechanical techniques, the combination of a solar-driven hoe robot with hand-hoeing appears to be most economic and environmentally sustainable.

Effect of Demeter Adjuvant (Rizospray Extremo) in Enhancing the Effectiveness of Herbicide Mixtures for Effective Weed Control in a Matured Oil Palm Plantation in Western Ghana

An Adjuvant is an additive chemical that is known to enhance the effectiveness of pesticides like herbicides by helping to improve the absorption of the herbicides by weeds and may eventually help to reduce the dosage of the herbicides used. In the current trial, the adjuvant (Rizospray Extremo) was applied at 3 rates, i.e., 73 ml/ha, 109 ml/ha and 145 ml/ha. The herbicide mixtures used for the trial were glyphosate (glyphader) and triclopyr (garlon) mixtures at 436 ml/ha and 182 ml/ha respectively (60 mls glyphosate + garlon at 25 mls mixed in 15 litres of water). The herbicide control was glyphosate at 436 ml/ha + garlon at 182 ml/ha (with no adjuvant) whiles one plot was used as the no herbicide control. Weeds on the site include both soft weeds and woodies. Observations on weed mortality were made on weekly basis. Generally, herbicide application (with and without the adjuvant) effectively controlled the soft weeds at the site. Percentage weed growth reduction and herbicide efficacies generally increased with increasing rates of the adjuvant. Highest herbicide efficacy of 96% was obtained by the use of the adjuvant at 145 ml/ha. The treatments differed significantly mostly in their levels of control of the woody weeds at the site with adjuvant application at 145 ml/ha inducing the highest control of woody weeds (herbicide efficacy of 95.7%), whiles Adjuvant application at 73 ml/ha induced an efficacy of 88.91% on the woodies at the site. Glyphosate plus garlon application (without the use of adjuvant) induced an efficacy of 78.39%. The adjuvant was very effective in controlling weeds with waxy cuticles like the Thaumatococcus spp which were not effectively controlled without the use of theand adjuvant. The adjuvant (Rizospray Extremo) is therefore recommended for use in herbicide mixtures for spot spraying of woodies or noxious weeds at the site.

Looking beyond Glyphosate for Site-Specific Fallow Weed Control in Australian Grain Production

Summer annual weed species in northern Australian summer fallows are frequently present at low densities and, increasingly, are glyphosate-resistant, creating the need for alternative herbicides for site-specific weed control. Alternative non-selective herbicide treatments are effective on problematic summer fallow weeds; however, many are yet to be evaluated as site-specific (spot spraying) treatments. This study aimed to identify herbicides that could be used in place of glyphosate to control larger/mature Chloris virgata and Sonchus oleraceus plants. The response of these weed species to 12 herbicide treatments was evaluated in pot experiments conducted over summer/autumn 2022. Despite herbicide treatments not being consistently effective across both species, there were instances where control was achieved by some herbicide treatments. S. oleraceus was controlled (i.e., ≤10% plant survival) by glufosinate-ammonium, paraquat and also with protoporphyrinogen-oxidase (PPO)-inhibiting herbicides saflufenacil, tiafenacil and trifludimoxazin. However, these results were not consistent in repeated studies or for C. virgata. Glyphosate was the only herbicide that controlled C. virgata. A glyphosate replacement as a spot-spraying treatment was not identified, and until further studies are more successful, alternative approaches are needed to preserve the ongoing effectiveness of this herbicide.

Optimal use of agricultural robot in arable crop rotation: A case study from the Netherlands

Acute shortages and rising costs of labour in the food and farming sector across Europe exhibit the fragility of agrifood systems. Agricultural robots present an opportunity to strengthen agrifood systems by addressing labour shortages and reduce CO2 emissions. This study proposes a method to investigate the potential use of an autonomous robotic system based on a case study on a typical high-tech Dutch farm that implements both an agricultural robot (AGR) and a conventional tractor (TRC) on a farm of 200 hectares in the region of Oldambt. To demonstrate the optimal use of an AGR, five farming operations (seeding, catch crop seeding, tine weeding, harrowing, inter-row hoeing, and spot spraying) in a crop rotation system of five crops (sugar beets, pumpkins, onions, spring barley and winter wheat) was chosen. The agricultural robot is here considered as a supplement (not substitute) to the farms overall cropping capability. It is found that price of fuel and labour are critical factors where higher fuel and labour price increases the benefit and use of the robot. Besides, time needed for remote supervision of the robot plays an important role.

Reducing herbicide input and optimizing spray method can minimize nontarget impacts on native grassland plant species.

Invasive plants threaten biodiversity worldwide and effective management must control the target invader while conserving biodiversity. Herbicide is often used to control invasive plants, but potential negative impacts on biodiversity have led to spot spraying being recommended over boom spraying to minimize the exposure of nontarget species to chemicals. We examined the influence of herbicide application methods on off-target plant communities in threatened temperate grasslands of southeastern Australia, where spraying with the broadleaf herbicide fluroxypr is commonly used to control St. John's wort, Hypericum perforatum L. It is well established that fluroxypr effectively controls H. perforatum but few studies have examined its impact on native forbs. A spray drift experiment using water-sensitive cards indicated that ground surface coverage was higher for spot spraying (91%-99%) than for boom spraying (5%-31%). We established a replicated, 3-year, before-after-control-impact experiment across 48 1-m2 quadrats to determine how three herbicide application methods (spot spray, fine boom and coarse boom) affected nontarget native forbs, the group most likely to be affected by broadleaf herbicides. This experiment was conducted in grasslands where H. perforatum was almost absent, so responses would reflect the direct impacts of the chemical, rather than structural changes resulting from removal of the target invader. Spot spraying decreased the probability of occurrence of native leguminous forbs, while increasing the occurrence of exotic leguminous forbs and the richness of all exotic species and exotic annual forbs. Spot spraying reduced the occurrence of the native Desmodium varians and the abundance of the native Chrysocephalum apiculatum. During this 3-year study, native species appeared to be impacted either directly by fluroxypr or indirectly by increased competition with exotic species. Where herbicide application is deemed crucial in these grasslands, we recommend boom spraying when H. perforatum density is moderate to high. Spot spraying should only be used when the density of H. perforatum is very low. Given the regional variation in H. perforatum density, the spatial scale of invasion, soil depth, and conservation values, we present a decision tree to assist managers in evaluating the costs and benefits of chemical control, indicating situations where alternative or modified methods could be used.

Development of robust communication algorithm between machine vision and boom sprayer for spot application via ISO 11783

A communication algorithm to send data via the standard communication protocol for agricultural implements, ISO 11783, was developed and tested in this research work. The goal was to enable a universal method to integrate machine vision systems to sprayers so that sensor-based real-time spot spraying becomes possible in the future. The algorithm receives pest detection results from a machine vision and constructs messages compatible with ISO 11783 (also known as ISOBUS), which gets sent to the sprayer. This allows for simultaneous opening of different nozzles on the sprayer even when pests are detected by different cameras at different locations across the boom. The results on our spraying apparatus which consisted of 6 nozzles showed that correct spraying commands were sent to the nozzles 100% of the times for all possible 64 simultaneous nozzle opening/closing combinations. The algorithm was designed to construct CAN frames based on arithmetic rather than logic operations. Also, the increase of data load on the ISOBUS of the sprayer was limited to 5.2%. Moreover, no data loss between the machine vision system and the sprayer was observed even when the data frames were sent every 10 milliseconds. The electronic control unit (ECU) on which the algorithm was written has two interfaces, one communicates with the machine vision system using serial communication, and the other communicates with the sprayer using the ISOBUS. The universal serial communication protocol used in this development allows for immediate integration to the ECU which creates the messages according to the specifications of the nozzle control unit of the sprayer.

Image‐based weed recognition and control: Can it select for crop mimicry?

AbstractAs highly adaptable plants, weeds have evolved numerous mechanisms to evade control in agroecosystems. For example, reliance on herbicides has resulted in widespread evolution of resistance in many species. Minimising weed adaptation is a major driver for integrated weed management strategies. Crop mimicry is a notable example of weed adaptation, where weed species evolve to avoid control by mimicking aspects of the crop phenotype. Visual selection by hand weeding has been documented to select for crop mimics that are difficult to distinguish from the crop at the vegetative stage. With recent advancements in weed recognition technologies, image‐based weed recognition for in‐crop, site‐specific weed control is on the cusp of becoming widely adopted. Whilst the control methods used in site‐specific weed control will be varied (e.g., spot spraying or lasers), they will share weed recognition technology. Visual selection via image‐based deep learning represents a selection pressure for weeds that can evade detection by mimicking crops. This mimicry may reduce weed recognition accuracy and thus weed control efficacy over time and result in difficult to manage mimetic weed phenotypes. Therefore, it is timely to explore the potential for selection of crop mimics by image‐based weed recognition algorithms.

Field evaluation of glutathione and glutamic acid as host marking pheromones for control of Tephritid fruit flies in a mango orchard in Kenya.

Tephritid fruit flies are a major constraint to the global production of horticultural crops. In Africa, an array of native and alien invasive fruit fly species represents a key challenge to the horticultural industry. In an effort to develop a safer management tool for these pests, we previously identified glutathione (GSH) and glutamic acid (GA), as the host marking pheromones (HMPs) of Ceratitis cosyra and Ceratitis rosa, respectively. Here, we report on the effectiveness of these compounds (GSH and GA) in reducing natural fruit fly population infestations in mango orchards. Mango trees at two different agroecological zones in Kenya were sprayed with HMPs, and the fruits sampled periodically and assessed for fruit fly emergence. Fruit fly emergence data were compared to two controls, the positive control using spot spraying of food bait (SS), and the negative control using water (W). The two HMPs and SS substantially decreased fruit fly emergence from the sampled mangoes. GSH and GA treated mangoes showed reduced C. cosyra and C. rosa infestation by ~70-75% relative to control (W), and with variation noted in the pheromone applied and the recovered fruit fly species. The adult emergence in pheromone-treated plants compared favorably with the positive control SS. However, the HMPs had little impact on reducing Bactrocera dorsalis infestation of mangoes. The decrease in fruit fly emergence in sampled mango fruits from HMP treated trees corroborate previous laboratory results and support the prospect of using HMPs in the management of African fruit fly species. © 2022 Society of Chemical Industry.

An Efficient Real-Time Weed Detection Technique using YOLOv7

Since farming is becoming increasingly more expensive, efficient farming entails doing so without suffering any losses, which is what the current situation desires. Weeds are a key issue in agriculture since they contribute significantly to agricultural losses. To control the weed, pesticides are now evenly applied across the entire area. This approach not only costs a lot of money but also harms the environment and people's health. Therefore, spot spray requires an automatic system. When a deep learning embedded system is used to operate a drone, herbicides can be sprayed in the desired location. With the continuous advancement of object identification technology, the YOLO family of algorithms with extremely high precision and speed has been applied in a variety of scene detection applications. We propose a YOLOv7-based object detection approach for creating a weed detection system. Finally, we used the YOLOv7 model with different parameters for training and testing analyzed on the early crop weed dataset and 4weed dataset. Experimental results revealed that the YOLOv7 model achieved the mAP@0.50, f1score, Precision, and Recall values for the bounding boxes as 99.6,97.6, 99.8, and 95.5 respectively on the early crop weed dataset and 78.53, 79.83, 86.34, and 74.24 on 4weed dataset. The Agriculture business can benefit from using the suggested YOLOv7 model with high accuracy in terms of productivity, efficiency, and time.

Digital technology adoption for plant protection: Assembling the environmental, labour, economic and social pieces of the puzzle

The prototype of a hoeing and spot spraying device was used to evaluate the impacts of digital technologies in vegetable production on labour, economic and social issues. These impacts were analysed in a case study conducted on a farm in Switzerland. Working-time requirements were modelled on the basis of time studies conducted during field trials. The results were used in a cost analysis and a comparison with a conventional plant-protection strategy. Furthermore, an in-depth interview with the farm manager revealed the personal, intellectual and social characteristics of successful technology adopters and the impact of applying a new technology to the working process. While one might expect digital technologies to substitute labour with capital, the prototype, in its current form, requires both higher investments and a higher input of labour per hectare. This study shows that the success of new technologies relies on enthusiastic farm managers in addition to public and consumer support, which can be gained by promoting the reduced amount of pesticides applied by these digital technologies.

Deep Learning Based Detector YOLOv5 for Identifying Insect Pests

Insect pests are a major element influencing agricultural production. According to the Food and Agriculture Organization (FAO), an estimated 20–40% of pest damage occurs each year, which reduces global production and becomes a major challenge to crop production. These insect pests cause sooty mold disease by sucking the sap from the crop’s organs, especially leaves, fruits, stems, and roots. To control these pests, pesticides are frequently used because they are fast-acting and scalable. Due to environmental pollution and health awareness, less use of pesticides is recommended. One of the salient approaches could be to reduce the wide use of pesticides by spraying on demand. To perform spot spraying, the location of the pest must first be determined. Therefore, the growing population and increasing food demand emphasize the development of novel methods and systems for agricultural production to address environmental concerns and ensure efficiency and sustainability. To accurately identify these insect pests at an early stage, insect pest detection and classification have recently become in high demand. Thus, this study aims to develop an object recognition system for the detection of crops damaging insect pests and their classification. The current work proposes an automatic system in the form of a smartphone IP- camera to detect insect pests from digital images/videos to reduce farmers’ reliance on pesticides. The proposed approach is based on YOLO object detection architectures including YOLOv5 (n, s, m, l, and x), YOLOv3, YOLO-Lite, and YOLOR. For this purpose, we collected 7046 images in the wild under different illumination and background conditions to train the underlying object detection approaches. We trained and test the object recognition system with different parameters from scratch. The eight models are compared and analyzed. The experimental results show that the average precision (AP@0.5) of the eight models including YOLO-Lite, YOLOv3, YOLOR, and YOLOv5 with five different scales (n, s, m, l, and x) reach 51.7%, 97.6%, 96.80%, 83.85%, 94.61%, 97.18%, 97.04%, and 98.3% respectively. The larger the model, the higher the average accuracy of the detection validation results. We observed that the YOLOv5x model is fully functional and can correctly identify the twenty-three species of insect pests at 40.5 milliseconds (ms). The developed model YOLOv5x performs the state-of-the-art model with an average precision value of (mAP@0.5) 98.3%, (mAP@0.5:0.95) value of 79.8%, precision of 94.5% and a recall of 97.8%, and F1-score with 96% on our IP-23 dataset. The results show that the system works efficiently and was able to correctly detect and identify insect pests, which can be employed for realistic application while farming.