Precision viticulture: Tool for production of quality grape

Introduction

Precision agriculture (PA), a relatively new discipline in agronomy developed in the mid-1980s, has been listed among the top ten developments in agriculture in recent decades (Santesteban, 2017) [38] . PA includes a wide range of techniques and technologies aimed at adapting crop management to field spatial variability. In view of the fact that the main characteristics of a crop production environment, such as water and nutrients, often vary considerably over space and time within a single agricultural area, it is necessary. It is therefore necessary to make management decisions taking into account these changes in order to meet the crop requirements properly. Precision viticulture (PV) corresponds to a part of the PA devoted to vineyard management. It is also important to distinguish between precision viticulture and digital viticulture. Precision viticulture has a specific application in viticulture derived from precision agriculture (Buss et al., 2005) [12] . The more accepted PA definition is as follows: precision agriculture is a management strategy that gathers, processes, and analyzes temporal, spatial, and individual data and combine it with other information to support management decision according to estimated variability for improved resource use efficiency, productivity, quality, profitability, and sustainability of agricultural production (Matese and Di Gennaro, 2015) [30] . In general, PA focuses on the implementation of technological advances in the assessment of spatial and temporal differences within a field. These technologies can assess soil-plantatmosphere conditions using different approaches, including wireless sensor networks Bramley and Hamilton, 2004 [8] . These proximal sensors can be in direct contact with the soil (soil moisture/salinity sensors), based on the plant trunk (e.g., sap flow sensors, dendrometers, acoustic sensors), at the canopy level (i.e., leaf temperature, psychrometry, and leaf wetness) or within the environment. Most of these proximal and contact sensor technologies offer high temporal resolution data that can be used to assess grapevine growth, water status, irrigation requirements, pest and disease detection, among others (Bramley, 2003; Bramley, 2005) [6,7] . Precision viticulture optimizes vineyard performance, maximizes grape yield and quality and minimizes environmental impacts and risks. This can be achieved by measuring local variations in factors affecting the yield and quality of grapes (soil, topography, microclimate, vine health, etc.) and using appropriate viticulture management practices (trellis design, pruning, fertilizer application, irrigation, harvest timing). Vineyards are characterized by a high degree of heterogeneity due to structural factors, such as morphological characteristics, as well as other factors such as cultivation practices and seasonal weather. This variability causes a different physiological response of the vine, which directly affects the quality of the grapes. Therefore, a specific agronomic management is required to satisfy the real needs of the crop in relation to spatial variability within the vineyard (Ozdemir et al., 2017) [32] . The introduction of new technologies aimed at supporting vineyard management makes it possible to improve the efficiency and quality of production and to reduce environmental impacts. With the help of the available tools, we can manage the unpredictability that makes precision horticulture practical by providing site-specific agronomic suggestions. Finally, evaluation is a crucial component of every precision farming system. The detailed steps involved in each process are clearly depicted as fallows

Objectives of precision farming in grape

Assessing variability

The crucial first step in precision farming is assessing variability. Since it is obvious that one cannot control what they do not know. There are numerous variables that affect crop performance in terms of yield over time and space. The challenge for precision farming is quantifying the variability of these factors and processes and figuring out when and where different combinations are accountable for the spatial and temporal variation in crop yield. Techniques for assessing spatial variability are readily available and have been applied extensively in precision agriculture. Assessing spatial variability is a key component of precision agriculture. There are methods for evaluating temporal variability as well, but the simultaneous reporting a spatial and temporal variation is rare. Both the spatial and the temporal statistics are necessary. We can observe the variation in yield of a crop in space but we cannot predict the reasons for the variation. It requires observations of crop growth and development over the course of the growing season, which is nothing but the temporal variation. To implement precision farming techniques, we therefore require both the space statistics and the time statistics.

Managing variability

Once variation is adequately assessed, farmers must match agronomic inputs to known conditions employing management recommendations. They are site-specific and employ precise application control technology. We can use the technology most effectively. In site-specific variability management we can utilize GPS instrument, so that the site specificity is pronounced and control will be easy and affordable. We must record the sample location coordinates while taking soil or plant samples so that we can use them later for management. This leads to the efficient utilization of inputs and the avoidance of waste, which is what we want.

Steps in Managing the variability a) Precision Soil Fertility Management b) Precision Pest Management c) Crop Management d) Water Management e) Soil Management

Evaluation

There are three important issues regarding precision agriculture evaluation Economics Environment Technology transfer

Economics:

The most important fact regarding the analysis of profitability of precision agriculture is that the value comes from the application of the data and not from the use of the technology.

Environment: Precision agriculture is frequently advocated for because of potential improvements in environmental quality. Decreased usage of agrochemicals, higher nutrient use efficiencies, increased efficiency of managed inputs, and increased soil yield from deterioration are frequently cited as potential benefits to the environment. Enabling technologies can make precision agriculture feasible, agronomic principles and decision rules can make it applicable and enhanced production efficiency or other forms of value can make it profitable.

Technology transfer:

The phrase "technology transfer" may indicate that precision agriculture happens when individuals or firms simply acquire and use the enabling technologies. While using enabling technologies and agronomic principles in precision agriculture does include managing spatial and temporal variability, the essential word here is manage. Much of the attention in what is called technology transfer has focused on how to communicate with the farmer. These issues associated with the managerial capability of the operator, the spatial distribution of infrastructure and the compatibility of technology to individual farms will change radically as precision agriculture continues to develop.

According to numerous prior studies, generally, the implementation of precision agriculture (PA) has shown to increase profits across different crops and with various technologies adopted (Schimmelpfennig and Ebel, 2016; Sumiahadi et al., 2019). There are numerous proven benefits of precision agriculture, which can be summarized as follows.  Improve crop Yield  Identification of plant stress  Constant monitoring of crops with the possibility to implement targeted actions.  Reduction of intra-field variability  Lower the environmental impact of agricultural operations  Give information to make better management decisions.  Reduction of costs and time of agricultural operations;  Almost no useful crop production resources are wasted since each must be used specifically according to the needs of the production system.  Optimization of the use of fertilizers, pesticides and water  Increase of products quality.

Major constraint in implementation of Precision Farming under Indian Conditions

Geographic Information System (GIS)

The GIS is an organized collection of computer hardware, software, geographical data, and personnel designed to efficiently capture, store, update, manipulate, analyze and display all forms of geographically referenced information. GIS is considered the brain of Precision Agriculture. Unlike traditional maps, computerized GIS maps have multiple layers of information (such as yield, soil survey maps, rainfall, crops, soil nutrient levels, and pests). Information on field topography, soil types, surface and subsurface drainage, irrigation, chemical application rates, soil testing, and crop production can all be found in a farming GIS database.

Sensor technologies

A range of technologies, including electromagnetic, conductivity, photoelectric, ultrasonic, and others, are employed to measure a variety of environmental parameters, including temperature, humidity, vegetation, texture, structure and physical character. Data from remote sensing is used to track drought, soil, and plant conditions, identify pests and weeds, and distinguish crop types (Di Gennaro et al., 2016) [17] . Sensors enable the collection of immense quantities of data without laboratory analysis.

Remote Sensing

It is a group of techniques used to gather data about land, water, object or region without coming into direct contact between the sensor and the subject of analysis. (Remote -Not in contact with an object and Sensing -Getting information). Remote sensing is a multidisciplinary field which encompasses a variety of scientific fields, including optics, spectroscopy, computer, satellite launching, photography and electronics. Remotely sensed data, obtained either by aircraft or satellite, containing electromagnetic emittance and reflectance data of crop can provide information useful for soil condition, plant growth, weed infestation etc. It is a valuable technology for precision agriculture because it can provide information on the field's parameters relatively easily. In general, we perceive the sun's reflected light, which is composed of infrared, red, green, and blue wavelengths as well as ultraviolet wave lengths. The green plants reflect the green and infrared waves while absorbing the red and blue waves (Hall et al., 2002) [21] . With multispectral cameras, we can measure the wavelengths that are reflected to determine the vigour of the plants or any problems like illness, nutrient deficiency, water logging, etc. We can relate the organic content, moisture, to the soil color (Hoff et al., 2017) [23] .

Satellites

Satellites have played a vital role in precision farming for over four decades, starting with the launch of Landsat 1 in 1972. This satellite, equipped with a multispectral sensor, provided imagery with a spatial resolution of 80 meters per pixel and revisited locations approximately every 18 days. Subsequent to this, Landsat 5, launched in 1984, enhanced imaging capabilities by capturing data in multiple spectral bands including blue, green, red, near-infrared, and thermal, at a spatial resolution of 30 meters. The initial utilization of remote sensing in precision agriculture involved analyzing Landsat imagery of exposed soil to discern spatial variations in soil organic matter content (Mulla, 2013) [31] .

VRT

Precision viticulture employs VRT to differentiate agronomic management and dosage inputs in both time and location. Automation technologies are used by modern agricultural machinery to regulate movement in the vineyard, including speed, direction of travel, and steering angle as well as to handle agronomic managements. With this technique, prescription maps created for individual operations can be combined with position data acquired by a GPS module. The agronomic inputs will now be applied based on the actual demands of the vines, which are determined by the vineyard's heterogeneity, rather than in average amounts per hectare. This technology meets the current needs of the food industry, ensuring adequate productivity and profitability in the vineyard. The resulting benefits are a substantial reduction of the work and speededup operations. The guidance systems can reduce operating stress, while the VRT provides a rational use of agronomic inputs, with direct impact on costs, quality, and environmental sustainability.

Application of precision viticulture

Precision viticulture, together with other tools such as different global positioning systems (GPS) and geographical information systems (GIS), promotes the capacity of grape and wine producers to acquire detailed geo-referenced information about vineyard performance and to start using this information to tailor the production of both grapes and wine according to expectations of vineyard performance and desired objectives in terms of both yield, quality and environment (Ozdemir et al., 2017) [32] . The process of viticulture precision begins with the mapping of yields and the acquisition of complementary information, followed by the interpretation and evaluation of this information, which leads to the implementation of targeted management strategies. It is followed by further observation. As a result, the data acquisition and use process is continuous, and the management improvements are incremental (Arno et al., 2017) [3] . At present, various precision viticulture applications help grape growers to produce high-quality grapes (Goldammer, 2015) [19] .

Terroir Management

Precision agriculture improves terroir management of vineyards (Bouma, 2015). Precision viticulture tools make it possible for growers, wine producers and researchers to understand that vineyard terroir may vary (Jose Maria et al., 2017) [24] . In fact, vineyards which produce wines characteristic of a particular region may be able to produce different wines from different areas within the same management unit (Bramley and Hamilton, 2007) [9] . Mobile platforms equipped with proximal soil sensors can be deployed across vineyards to gather geo-referenced soil data, enabling the creation of high-resolution maps that highlight spatial variations in soil properties and topography. This information is crucial for the establishment or redevelopment of vineyards, as noted by Bramley (2010) [10] . Recently, advancements have integrated electromagnetic induction sensors and multispectral imaging to assess variability in vineyard soil and vine vigor, as detailed by Hubbard et al. (2021) [22] .

Canopy management

Canopy and vigour control are the areas where growers are mostly used for a number of reasons. It is possible to get timely and high-resolution information during the growing period, which may be relevant for canopy management, fertilization, and irrigation (Brunori et al., 2022). Remote and proximal sensing technologies have been extensively utilized to assess vine canopy growth and health in commercial vineyards, as highlighted by Darra et al. (2021) [16] . Portable fluorescence-based sensors are employed for evaluating leaf nitrogen levels and can be mounted on mobile platforms (Diago et al., 2016a) [18] . Additionally, light detection and ranging (LiDAR) technology has demonstrated its effectiveness in rapidly and nondestructively evaluating canopy and leaf characteristics in vineyards (Arno et al., 2013) [5] . Arno et al. (2017) [3] used mobile terrestrial laser scanners (MTLS) to study the significance of mapping the leaf area index (LAI) for viticulture. According to Luo et al. (2016) [27] , when harvesting grape clusters, grapes are likely to collide and be damaged by manipulation, binocular stereo vision was used to locate the spatial coordinates of the cutting points on a peduncle of grape clusters for the end-effector and determine the bounding volume of the grape clusters for the motion planner of the manipulator to conduct undamaged robotic harvesting. The cutting-point detection success rate is approximately 87%, indicating that this method can be used on harvesting robots. In order to monitor vineyards, numerous methods have been created. These systems combine a GPS system for data georeferencing with a highresolution screening of the canopy side across the row. Sensors is called GrapeSense (Lincoln Ventures Ltd, Hamilton, New Zealand), which measures the height and texture of the vines along the row by taking a highfrequency digital picture of the canopy side.

Water management

Given the increasing scarcity and regulation of water, improved management is essential. Variations in soil type and topography within vineyards mean that different blocks have varying water requirements. Advanced irrigation systems have been developed to precisely deliver water where it is needed. Monitoring vineyard water status is crucial for controlling yield and quality, and it directly influences irrigation management. Monitoring stem and leaf water potential at different times of the day has been effectively used to monitor, control, and manage irrigation, showing strong correlations with soil and plant water status as well as vegetation indices (Cancela et al., 2017) [13] . Thermal imaging has emerged as a practical tool for managing agricultural water use by providing rapid assessment of canopy surface temperature, which correlates with transpiration and indicates crop water status. Recently, thermal imaging systems have improved in resolution and decreased in weight, facilitating their integration onto Unmanned Aerial Vehicles (UAVs) for civil and agricultural engineering applications. This approach addresses many limitations of on-site thermal imaging by enabling the mapping of plant water status over large areas, accounting for natural or artificially induced variability at field or farm scales. Santesteban et al. (2017) [38] investigated the extent to which high-resolution thermal imaging enables the assessment of both instantaneous and seasonal variations in water status within a vineyard.

Disease and insect management

Grape cultivation in India is significantly threatened by various diseases and insect pests. Key fungal diseases include downy mildew, powdery mildew, and anthracnose, while major insect pests such as mealybugs and thrips lead to substantial economic losses. Downy mildew is particularly severe in several horticultural crops, including grapes. The effective management of pests hinges on the precise application of the correct pesticide, at the appropriate dosage, timing, and location. Remote sensing and monitoring technologies have revolutionized the identification of diseases and pests in vineyards, offering precise, swift, cost-effective, and reliable diagnostics compared to the time-consuming, subjective, risky, and expensive methods of visual inspection (Lee and Tardaguila, 2021) [26] . These advancements enable the mapping of pests and diseases across vineyards, facilitating targeted application of fungicides using Variable Rate Technology (VRT) (Chen et al., 2020) [14] . Gutierrez et al. (2021b) [20] utilized computer vision and deep learning techniques to successfully detect and distinguish between downy mildew and spider mite in commercial vineyards. Oberti et al.

(2014) [33] researched the automated identification of powdery mildew on grapevine leaves through image analysis. They highlighted that powdery mildew constitutes a significant fungal threat to grapevines (Vitis vinifera L.) and other critical specialty crops, leading to substantial damage such as reduced yield and diminished quality of wine or produce. Oberti et al. (2016) [34] developed an agricultural robot equipped with a new precision-spraying end-effector with an integrated disease-sensing system based on R-G-NIR multispectral imaging.

In India, the primary insect pests affecting grapes, ranked by economic impact, include mealybugs, thrips, flea beetles, leafhoppers, stem borers, and mites. Excessive pesticide use not only escalates cultivation costs but also poses environmental and human health risks. Therefore, effective pest management in viticulture should adopt an integrated approach encompassing optimal farming practices, advanced forecasting models, decision support systems (DSSs), biological controls, and judicious use of chemical sprays to minimize pesticide usage (Pertot et al., 2017) [37] . Chougule et al. (2019) [15] developed "PDMGrapes," a decision support system for grape crop protection in India's hot tropical regions, employing ontology, semantic web rule language, and image processing techniques to manage insect pests and diseases effectively.

Crop load monitoring

In order to ensure a uniform production of high-quality fruits and mature wood, it is necessary to manage the volume of crops in vineyard. "Crop load" is the ratio of the exposed leaf area to fresh fruit weight. A large quantity of leaves contributes to shading and reduces the quality of the fruit and, in some cases, the bud fruit. Too little leaf space per unit of fruit delays ripening and reduces the size of vines. Measurements of crop load are useful for growers to assess the success of vineyard management. The Ravaz index is a common metric used for estimating crop load according to the ratio of yield to pruning weight. Current and future research on precision viticulture has many different priorities, such as environmental economics, production quality assessment methods and new technologies for crop monitoring (Arno et al., 2009) [4] .

Berry quality monitoring NDVI image is an excellent tool for designing quality sampling zones based on NDVI classification. The sourceto-sink size ratio, i.e. the relative abundance of photosynthetically active organs (leaves) relative to photosynthate-demanding organs (mainly bunches), is widely known to be one of the main drivers of grape oenological quality (Kavak et al., 2014) [25] . However, precision viticulture (PV) has been mainly based on withinfield zone delineation using vegetation indices due to the difficulty of remote sink size estimation. This approach has given only moderately satisfactory results for discriminating zones with differential quality. Urretavizcaya et al. (2017) investigated an approach to delineate within-vineyard quality zones that included an estimator of sink size in the dataset set. We performed zone delineation using normalized difference vegetation index (NDVI), soil apparent electrical conductivity (ECa), and bunch number (BN) data.

Harvest management

During grape harvesting, the measurement of yield is commonly achieved by weighing berries as they pass over load cells integrated into mechanical harvesters (Taylor et al., 2015) [39] . Recently, computer vision systems have been employed to estimate grape yields based on factors such as cluster density (Palacios et al., 2019) [35] , berry count per cluster (Aquino et al., 2018) [2] , cluster weight (Liu et al., 2020b) [29] , and berry size (Roscher et al., 2014). The proper ripening of grapes is the key to obtain a high-quality wine and another grape product. The spatial distribution of the vineyard and planted variety, in addition to uncontrollably occurring climate conditions, affects the ripening of grape. It is a multifaceted process that cannot be characterized by a single parameter; instead, the profile of the grape's components is altered (Aquino et al., 2015) [1] . Several Android-based applications, including vitisFlower® (Aquino et al., 2015) [1] , vitisBerry® (Aquino et al., 2015) [1] , and 3DBunch® (Liu et al., 2020a) [28] , have been developed to measure parameters related to flowers, berries, and bunches. For large-scale commercial vineyards, mobile vehicles equipped with automatic RGB image capturing devices are utilized to capture images, and grape yield predictions are generated using computer vision technology (Palacios et al., 2020) [36] .

Conclusion

References

1. Aquino A, Millan B, Gutiérrez S, Tardáguila J. Grapevine flower estimation by applying artificial vision techniques on images with uncontrolled scene and multi-model analysis. Comput Electron Agric. 2015;119:92-104.
2. Aquino A, Barrio I, Diago MP, Millan B, Tardaguila J. VitisBerry®: an Android-smartphone application to early evaluate the number of grapevine berries by means of image analysis. Comput Electron Agric. 2018;148:19-28.
3. Arno J, Martinez Casasnovas JA, Ribes Dasi M, Rosell JR. Precision Viticulture. Research topics, challenges and opportunities in site-specific vineyard management. Span Agric Res. 2009;7(4):779-790.
4. Arno J, Escola A, Valles JM, Liorens J, Saniz R, Masip J, et al. Leaf area index estimation in vineyards using a ground-based LiDAR scanner. Precis Agric. 2013;14:290-306.
5. Bramley R, Pearse B, Chamberlain P. Being profitable precisely -a case study of precision viticulture from Margaret River. Aust N Z Grapegrower Winemaker. 2003;473a:84-87.
6. Bramley RGV. Understanding variability in winegrape production systems. 2. Within vineyard variation in quality over several vintages. Aust J Grape Res. 2005;11:33-42.
7. Bramley RGV, Hamilton RP. Understanding variability in winegrape production systems. 1. Within vineyard variation in yield over several vintages. Aust J Grape Wine Res. 2004;10:32-45.
8. Bramley RGV, Hamilton RP. Terroir and Precision Viticulture: are they compatible? J Int Sci Vigne Vin. 2007;41(1):1-8.
9. Brunori E, Moresi FV, Maesano M, Horatis M, Salvati R, Mugnozza G, et al. Field survey and UAV remote sensing as tools for evaluating the canopy management effects in smallholder grapevine farm. BIO Web Conf. 2022;44:05001.
10. Cancela JJ, Fandino M, Rey BJ, Dafonte J, Gonzalez XP. Discrimination of irrigation water management effects in pergola trellis system vineyards using a vegetation and soil index. Agric Water Manag. 2017;183:70-77.
11. Chen P, Xiao Q, Zhang J, Xie C, Wang B. Occurrence prediction of cotton pests and diseases by bidirectional long short-term memory networks with climate and atmosphere circulation. Comput Electron Agric. 2020;176:105612.
12. Chougule A, Jha VK. Decision support for grape crop protection using ontology. Int J Reasoning-based Intell Syst. 2019;11:24-37.
13. Darra N, Psomiadis E, Kasimati A, Anastasiou A, Anastasiou E, Fountas S. Remote and Proximal Sensing-Derived Spectral Indices and Biophysical Variables for Spatial Variation Determination in Vineyards. Agronomy. 2021;11:741.
14. Di Gennaro SF, Battiston E, Di Marco S, Facini O, Matese A, Nocentini M, et al. Unmanned Aerial Vehicle (UAV)-based remote sensing to monitor grapevine leaf stripe disease within a vineyard affected by esca complex. Phytopathol Mediterr. 2016;55:262.
15. Diago MP, Rey-Carames C, Le Moigne M, Fadaili EM, Tardaguila J, Cerovic ZG. Calibration of noninvasive fluorescence-based sensors for the manual and on-thego assessment of grapevine vegetative status in the field. Aust J Grape Wine Res. 2016;22:438-449.
16. Gutierrez S, Hernandez I, Ceballos S, Barrio I, DiezNavajas AM, Tardaguila J. Deep learning for the differentiation of downy mildew and spider mite in grapevine under field conditions. Comput Electron Agric. 2021;182:105991.
17. Hall A, Lamb DW, Holzapfel B, Louis J. Optical remote sensing applications in viticulture -a review. Aust J Grape Wine Res. 2002;8:36-47.
18. Hubbard SS, Schmutz M, Balde A, Falco N, Peruzzo L, Dafflon B, et al. Estimation of soil classes and their relationship to grapevine vigour in a Bordeaux vineyard: advancing the practical joint use of electromagnetic induction (EMI) and NDVI datasets for precision viticulture. Precis Agric. 2021;22:1353-1376.
19. Hoff R, Ducati J, Farias A. GIS and Remote Sensing to Support Precision Viticulture for Analysis of Vineyards in the Campanha Wine Region, Brazil. J Environ Agric Sci. 2017;10:20-32.
20. Kavak MT, Karadogan S, Ozdemir G. Investigating Vineyard Areas of Egil County of Diyarbakır Using Remote Sensing and GIS Techniques. Int Mesopotamia Agric Congr. 2014;307-316.
21. Luo L, Tang Y, Zou X, Ye M, Li G. Vision based extraction of spatial information in grape clusters for harvesting Engineering. 2016;151:90-104.
22. Liu S, Zeng X, Whitty M. 3DBunch: a Novel iOSSmartphone application to evaluate the number of grape berries per bunch using image analysis techniques. IEEE Access. 2020;8:114663-114674.
23. Liu S, Zeng X, Whitty M. A vision-based robust grape berry counting algorithm for fast calibration-free bunch weight estimation in the field. Comput Electron Agric. 2020;173:105360.
24. Matese A, Di Gennaro SF. Technology in precision viticulture: A state of the art review. Int J Wine Res. 2015;7:69-81.
25. Mulla DJ. Twenty five years of remote sensing in precision agriculture: key advances and remaining knowledge gaps. Biosyst Eng. 2013;114(4):358-371.
26. Ozdemir G, Sessiz A, Pekitkan FG. Precision Viticulture tools to production of high quality grapes. Sci Pap Ser B Hortic. 2017;61.
27. Oberti R, Marchi M, Tirelli P, Calcante A, Iriti M, Borghese AN. Automatic detection of powdery mildew on grapevine leaves by image analysis: Optimal viewangle range to increase the sensitivity. Comput Electron Agric. 2014;104:1-8.
28. Oberti R, Marchi M, Tirelli P, Calcante A, Iriti M, Tona E, et al. Selective spraying of grapevines for disease control using a modular agricultural robot. Biosyst 2016;146:203-215.
29. Palacios F, Diago MP, Tardaguila J. A non-invasive method based on computer vision for grapevine cluster compactness assessment using a mobile sensing platform under field conditions. Sensors (Basel). 2019;19:3799.
30. Palacios F, Bueno G, Salido J, Diago MP, Hernandez I, Tardaguila J. Automated grapevine flower detection and quantification method based on computer vision and deep learning from on-the-go imaging using a mobile sensing platform under field conditions. Comput Electron Agric. 2020;178:105796.
31. Pertot TC, Rossi V, Mugnai L, Hoffmann C, Grando MS, Gary C, et al. A critical review of plant protection tools for reducing pesticide use on grapevine and new perspectives for the implementation of IPM in viticulture. Crop Prot. 2017;97:70-84.
32. Santesteban LG, Di Gennaro SF, Herrero-Langreo A, Miranda C, Royo JB, Matese A. High-resolution UAVbased thermal imaging to estimate the instantaneous and seasonal variability of plant water status within a vineyard. Agric Water Manag. 2017;183:49-59.
33. Taylor J, Aubrecht C, Alonso P, Sietz D, Hill J. Dynamics and intensity of shifting cultivation at the southern margins of the Amazon forest. Int Arch Photogram Remote Sens Spat Inf Sci. 2015;40(7):927932.
34. Taylor J, Lowery B, Numata I, Llerena C, Coletta L, David C. Monitoring shifting cultivation at forest margins using traditional and autonomous remote sensing techniques. Geocarto Int. 2021;36(10):11211140.
35. Vitone A, Linsalata M, Lovreglio R, Castellini M, Fiori N, Mazzone F, et al. Modeling Grape Phenology Using UAV Multispectral Imaging in a Precision Viticulture Framework. Sustainability. 2021;13:13071.
36. Whitty M, Rogge D, Ford C. VitiCanopy: A smartphone app for grapevine canopy management. Comput Electron Agric. 2018;151:320-324.
37. Zhang J, Rowland DL, Whitty M. Automated counting of grapevine berries using deep learning algorithms: a comparative study. Comput Electron Agric. 2019;162:871-880.
38. Zhang J, Whitty M, Heras J. Detecting berry number and size in a vineyard using automated image analysis algorithms: A comparative study of a convolutional neural network and traditional image analysis techniques. Comput Electron Agric. 2020;168:105124.