[go: up one dir, main page]
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
8.3
4.2
Journals
Remote Sensing
Special Issues
Remote Sensing in Viticulture II
remotesensing-logo
Submit to Special Issue Submit Abstract to Special Issue Review for Remote Sensing Propose a Special Issue

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Remote Sensing in Viticulture II

A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Remote Sensing in Agriculture and Vegetation".

Deadline for manuscript submissions: 25 January 2025 | Viewed by 14165

Special Issue Editors

Dr. Luís Pádua

E-Mail Website
Guest Editor
Centre for the Research and Technology of Agro-Environmental and Biological Sciences (CITAB), University of Trás-os-Montes and Alto Douro (UTAD), 5000-801 Vila Real, Portugal
Interests: digital image processing; unmanned aerial vehicles; precision viticulture; precision agriculture; photogrammetric processing; multi-temporal analysis; spectral imaging; machine learning
Special Issues, Collections and Topics in MDPI journals
Dr. Emanuel Peres

E-Mail Website
Guest Editor
Department of Engineering, School of Sciences and Technology, University of Trás-os-Montes e Alto Douro, 5000-801 Vila Real, Portugal
Interests: remote sensing; precision agriculture; in-field data processing; remote monitoring; UAV; UAS; precision forestry; sensors and data processing; human–computer interfaces; augmented reality; virtual reality; embedded systems
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Raul Morais dos Santos

E-Mail Website
Guest Editor
Centre for the Research and Technology of Agro-Environmental and Biological Sciences (CITAB), University of Trás-os-Montes and Alto Douro (UTAD), 5000-801 Vila Real, Portugal
Interests: sensor interfaces; microelectronics; wireless sensor networks; IoT; precision viticulture; energy harvesting; proximal sensing
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

The technological and scientific developments in the last several decades have allowed the emergence of new approaches for data acquisition and the processing of remote sensed data within the context of precision viticulture. Remote sensing contributes to improvements in decision-support systems, making it possible to retrieve a multitude of information. Vineyard mapping, monitoring phytosanitary issues, yield and quality estimation, and water status monitoring are among the viticulture applications that can be performed with remotely sensed data. In fact, current technology and methodologies enable vineyard monitoring at a parcel scale or at an individual plant scale, with the possibility of estimating different biophysical and geometrical plant parameters, including multi-sensor data fusion approaches.

Given the current widespread availability and accessibility for acquiring, processing and analyzing proximal and remotely sensed data, it is possible to employ these data within any specific period, regardless of the location. Constant updates to time series data with different spatial, spectral and temporal resolutions available from satellite systems enable continuous vineyard monitoring at, local, regional and global scales. The emergence of unmanned aerial systems and mobile or stationary proximal sensing platforms has made it possible to acquire huge amounts of data from various sensors. Considering environmental and economic sustainability, the use of tridimensional, multi- or hyperspectral, and thermal data opens new possibilities to promote more sustainable and efficient vineyard management, supporting the preservation of natural resources.

This Special Issue aims to encourage the publication of studies or review articles documenting recent advances in the viticulture sector using remote sensing and intelligent field monitoring. It aims to cover the development of novel methodologies, algorithms, and applications using remotely sensed data including, but not limited to: grapevine vegetation monitoring using unmanned aerial vehicles (UAVs), airborne and satellite data; vigor mapping and site-specific applications; time series and multi-temporal vineyard analysis; digital image processing, computer vision and machine learning methods applied in viticulture; precision viticulture methods; advances in proximal sensing in viticulture, including the use of image sensors; as well as the estimation and mapping of water status, irrigation demands, and phytosanitary issues. 

Dr. Luís Pádua
Dr. Emanuel Peres
Dr. Raul Morais
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • precision viticulture
  • vineyard management
  • vineyard mapping and classification
  • machine and deep learning
  • decision support
  • phenological modelling and yield prediction
  • estimation of biophysical and geometrical parameters
  • intelligent monitoring
  • multi-temporal analysis

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here.

Published Papers (6 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

Jump to: Review

21 pages, 19780 KiB  
Article
Characterisation of Two Vineyards in Mexico Based on Sentinel-2 and Meteorological Data
by Maria S. del Rio, Victor Cicuéndez and Carlos Yagüe
Remote Sens. 2024, 16(14), 2538; https://doi.org/10.3390/rs16142538 - 10 Jul 2024
Cited by 1 | Viewed by 582
Abstract
In Mexico, viticulture represents the second source of employment in the agricultural area after the fruit and vegetable sector. In developed countries, remote sensing is widely used for vineyard monitoring; however, this tool is barely used in the developing countries of Iberoamerica. In [...] Read more.
In Mexico, viticulture represents the second source of employment in the agricultural area after the fruit and vegetable sector. In developed countries, remote sensing is widely used for vineyard monitoring; however, this tool is barely used in the developing countries of Iberoamerica. In this research, our overall objective is to characterise two vineyards in the state of Queretaro (Mexico) using Sentinel-2 and meteorological data, specifically spectral and thermal indices. Results show that spectral indices obtained from Sentinel-2 bands have adequately characterised the phenological dynamics of the different varieties of the vineyards. The Modified Soil-Adjusted Vegetation Index (MSAVI) was adequately used to discriminate between the first stages of vineyards, while the Normalized Difference Vegetation Index (NDVI) was useful for monitoring vineyards during the rest stages of vineyards. Thermal indices have shown that the best grape varieties are those that can adapt to both cooler and warmer temperatures, have a reasonable ripening period, and can produce wines with balanced acidity and flavours. In conclusion, the combination of meteorological (including thermal indices) and remote sensing data (NDVI and MSAVI) provide information for choosing a suitable grape variety for this region. Full article
(This article belongs to the Special Issue Remote Sensing in Viticulture II)
►▼ Show Figures

Figure 1

Figure 1
<p>Queretaro in the world. Images were taken from the ESRI satellite for QGIS.</p> Full article ">Figure 2
<p>Images from Google Satellite of the studied vineyards: (<b>a</b>) Puerta del Lobo (PL) and (<b>b</b>) Vinaltura (VA).</p> Full article ">Figure 3
<p>Monthly rainfall and average maximum (red line) and minimum (blue line) monthly temperatures for the period 1981-2023: (<b>a</b>) Puerta del Lobo and (<b>b</b>) Vinaltura. The black triangles and the orange circles correspond to the values for the year 2022.</p> Full article ">Figure 3 Cont.
<p>Monthly rainfall and average maximum (red line) and minimum (blue line) monthly temperatures for the period 1981-2023: (<b>a</b>) Puerta del Lobo and (<b>b</b>) Vinaltura. The black triangles and the orange circles correspond to the values for the year 2022.</p> Full article ">Figure 4
<p>Dynamics of the studied meteorological variables in the two vineyards during year 2022: (<b>a</b>) dynamics of temperature (°C) and precipitation (mm) in Puerta del Lobo (PL), (<b>b</b>) dynamics of temperature (°C) and precipitation (mm) in VA, (<b>c</b>) dynamics of relative humidity (%) and dew Point (°C) in PL, and (<b>d</b>) dynamics of relative humidity (%) and dew point (°C) in Vinaltura (VA).</p> Full article ">Figure 4 Cont.
<p>Dynamics of the studied meteorological variables in the two vineyards during year 2022: (<b>a</b>) dynamics of temperature (°C) and precipitation (mm) in Puerta del Lobo (PL), (<b>b</b>) dynamics of temperature (°C) and precipitation (mm) in VA, (<b>c</b>) dynamics of relative humidity (%) and dew Point (°C) in PL, and (<b>d</b>) dynamics of relative humidity (%) and dew point (°C) in Vinaltura (VA).</p> Full article ">Figure 5
<p>Maximal and minimal daily temperatures in Puerta del Lobo (PL) (<b>a</b>) and in Vinaltura (VA) (<b>b</b>) for the year 2022.</p> Full article ">Figure 6
<p>Dynamics of the spectral vegetation indices, NDVI and MSAVI, of the mean of the plots for the two vineyards: (<b>a</b>) Puerta del Lobo (PL) and (<b>b</b>) Vinaltura (VA) for the year 2022. The letters are the different phenological stages identified in field, shown in <a href="#remotesensing-16-02538-t002" class="html-table">Table 2</a>: S—Sprouting, LA—Leaf Appearance, F—Flowering, V—Veraison, H—Harvest, B—Browning of leaves.</p> Full article ">Figure 7
<p>MSAVI maps showing every 15 days from January to June 2022 for Puerta del Lobo (PL).</p> Full article ">Figure 8
<p>NDVI maps showing every 15 days from July to December 2022 for Puerta del Lobo (PL).</p> Full article ">Figure 9
<p>MSAVI maps showing every 15 days from January to June 2022 for Vinaltura (VA).</p> Full article ">Figure 10
<p>NDVI maps showing every 15 days from July to December 2022 for Vinaltura (VA).</p> Full article ">Figure 11
<p>Dynamics of NDVI and precipitation obtained from CHIRPS in Puerta del Lobo (PL) (<b>a</b>) and Vinaltura (VA) (<b>b</b>) for the year 2022.</p> Full article ">
24 pages, 28702 KiB  
Article
Vineyard Zoning and Vine Detection Using Machine Learning in Unmanned Aerial Vehicle Imagery
by Milan Gavrilović, Dušan Jovanović, Predrag Božović, Pavel Benka and Miro Govedarica
Remote Sens. 2024, 16(3), 584; https://doi.org/10.3390/rs16030584 - 3 Feb 2024
Cited by 2 | Viewed by 1950
Abstract
Precision viticulture systems are essential for enhancing traditional intensive viticulture, achieving high-quality results, and minimizing costs. This study explores the integration of Unmanned Aerial Vehicles (UAVs) and artificial intelligence in precision viticulture, focusing on vine detection and vineyard zoning. Vine detection employs the [...] Read more.
Precision viticulture systems are essential for enhancing traditional intensive viticulture, achieving high-quality results, and minimizing costs. This study explores the integration of Unmanned Aerial Vehicles (UAVs) and artificial intelligence in precision viticulture, focusing on vine detection and vineyard zoning. Vine detection employs the YOLO (You Only Look Once) deep learning algorithm, achieving a remarkable 90% accuracy by analysing UAV imagery with various spectral ranges from various phenological stages. Vineyard zoning, achieved through the application of the K-means algorithm, incorporates geospatial data such as the Normalized Difference Vegetation Index (NDVI) and the assessment of nitrogen, phosphorus, and potassium content in leaf blades and petioles. This approach enables efficient resource management tailored to each zone’s specific needs. The research aims to develop a decision-support model for precision viticulture. The proposed model demonstrates a high vine detection accuracy and defines management zones with variable weighting factors assigned to each variable while preserving location information, revealing significant differences in variables. The model’s advantages lie in its rapid results and minimal data requirements, offering profound insights into the benefits of UAV application for precise vineyard management. This approach has the potential to expedite decision making, allowing for adaptive strategies based on the unique conditions of each zone. Full article
(This article belongs to the Special Issue Remote Sensing in Viticulture II)
►▼ Show Figures

Figure 1

Figure 1
<p>Area of interest, vineyards for testing and training.</p> Full article ">Figure 2
<p>Graphical representation of proposed model for the zoning of vineyards.</p> Full article ">Figure 3
<p>Details of the proposed model for the detection of living vines.</p> Full article ">Figure 4
<p>Defining bounding boxes and the determination of the final bounding box.</p> Full article ">Figure 5
<p>The appearance of the vine tree before the development of vegetation (<b>a</b>) and its shadow, and the identified shadow cast by the tree (<b>b</b>).</p> Full article ">Figure 6
<p>Example of training dataset with drawn bounding boxes for vine shadows and the coordinates of each bounding box.</p> Full article ">Figure 7
<p>Determining the positions of live vines (bounding boxes containing identified shadows (<b>a</b>), translated bounding boxes (<b>b</b>), extracted parts of the vine row (<b>c</b>), localized vine (<b>d</b>)).</p> Full article ">Figure 7 Cont.
<p>Determining the positions of live vines (bounding boxes containing identified shadows (<b>a</b>), translated bounding boxes (<b>b</b>), extracted parts of the vine row (<b>c</b>), localized vine (<b>d</b>)).</p> Full article ">Figure 8
<p>K-means clustering-based pixel classification.</p> Full article ">Figure 9
<p>Part of the results after applying the YOLO algorithm (<b>a</b>) and after completing all steps of the proposed algorithm for vine detection (<b>b</b>) in the analysed vineyard.</p> Full article ">Figure 10
<p>Results of the detection of vines for all three combinations: 2020–2020 (<b>a</b>), 2020–2022 (<b>b</b>), and 2022–2022 (<b>c</b>).</p> Full article ">Figure 11
<p>NDVI before (<b>a</b>) and after (<b>b</b>) inter-row removal.</p> Full article ">Figure 12
<p>The result of clustering using the K-means method (<b>a</b>) and final (filtered) management zones (<b>b</b>).</p> Full article ">Figure 13
<p>Box plot diagrams for the obtained clusters. Values of N, P, and K content are in % of dry wt.</p> Full article ">
22 pages, 4887 KiB  
Article
Multispectral and Thermal Sensors Onboard UAVs for Heterogeneity in Merlot Vineyard Detection: Contribution to Zoning Maps
by Luz K. Atencia Payares, Ana M. Tarquis, Roberto Hermoso Peralo, Jesús Cano, Joaquín Cámara, Juan Nowack and María Gómez del Campo
Remote Sens. 2023, 15(16), 4024; https://doi.org/10.3390/rs15164024 - 14 Aug 2023
Cited by 7 | Viewed by 1236
Abstract
This work evaluated the ability of UAVs to detect field heterogeneity and their influences on vineyard development in Yepes (Spain). Under deficit irrigation, vine growth and yield variability are influenced by soil characteristics such as water holding capacity (WHC). Over two irrigation seasons [...] Read more.
This work evaluated the ability of UAVs to detect field heterogeneity and their influences on vineyard development in Yepes (Spain). Under deficit irrigation, vine growth and yield variability are influenced by soil characteristics such as water holding capacity (WHC). Over two irrigation seasons (2021–2022), several vegetation indices (VIs) and parameters of vegetative growth and yield were evaluated in two field zones. Multispectral and thermal information was obtained from bare soils. The water availability showed annual differences; it was reduced by 49% in 2022 compared to 2021, suggesting that no significant differences were found for the parameters studied. The zone with higher WHC also had the higher vegetative growth and yield in 2021. This agreed with the significant differences among the VIs evaluated, especially the ratio vegetation index (RVI). Soil multispectral and thermal bands showed significant differences between zones in both years. This indicated that the soil spectral and thermal characteristics could provide more reliable information for zoning than vine vegetation itself, as they were less influenced by climatic conditions between years. Consequently, UAVs proved to be valuable for assessing spatial and temporal heterogeneity in the monitoring of vineyards. Soil spectral and thermal information will be essential for zoning applications due to its consistency across different years, enhancing vineyard management practices. Full article
(This article belongs to the Special Issue Remote Sensing in Viticulture II)
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of the different sources of information used to describe the spatial and temporal heterogeneity in a commercial vineyard in Yepes-Toledo.</p> Full article ">Figure 2
<p>Monthly rainfall, reference evapotranspiration, and minimum and maximum temperatures from October 2020 until September 2022 located in the experimental vineyard (Toledo, Spain).</p> Full article ">Figure 3
<p>(<b>a</b>) Aerial photograph of the experimental vineyard. Lines indicate the preliminary soil zoning of the commercial vineyard, the red rectangle indicates the experimental zone, the locations of the two soil trenches (ST1 and ST2) are indicated with red points; (<b>b</b>) soil map details corresponding to the experimental zone’s final zoning in the vineyard (Toledo, Spain).</p> Full article ">Figure 4
<p>Screenshot of the ImageJ program used to obtain the percentage value of the gaps of the vines by transforming the RGB images.</p> Full article ">Figure 5
<p>Multispectral image detail with the location of the canopy experimental vine (<b>a</b>) and soil (<b>b</b>) in one of the study soil plots (ST1).</p> Full article ">Figure 6
<p>Horizons of two soil trenches: (<b>a</b>) Soil Trench 1 (ST1); (<b>b</b>) soil Trench 2 (ST2).</p> Full article ">Figure 7
<p>Histogram of each multispectral band of the soil surface of nearby experimental vines (ST1 and ST2) in the vineyard (Toledo, Spain) based on 500 pixels for each site. The first two rows correspond to 2021 (−21), and the rest correspond to 2022 (−22). The bands are red edge (<b>a</b>,<b>e</b>); red (<b>b</b>,<b>f</b>); NIR (<b>c</b>,<b>g</b>); and green (<b>d</b>,<b>h</b>).</p> Full article ">Figure 8
<p>Vegetation indices of two soil trenches (ST1 and ST2) located near the experimental vines (Toledo, Spain). The first row corresponds to 2021, and the second row corresponds to 2022. The indices are NDVI (<b>a</b>,<b>d</b>); RVI (<b>b</b>,<b>e</b>); and GNDVI (<b>c</b>,<b>f</b>). Levels of statistical significance (Sig.): ns, non-significant; * <span class="html-italic">p</span> &lt; 0.05; ** <span class="html-italic">p</span> &lt; 0.01. Statistics are based on 500 pixels for each soil trench (ST1 and ST2).</p> Full article ">
25 pages, 4705 KiB  
Article
Modeling Phenology Combining Data Assimilation Techniques and Bioclimatic Indices in a Cabernet Sauvignon Vineyard (Vitis vinifera L.) in Central Chile
by Víctor García-Gutiérrez and Francisco Meza
Remote Sens. 2023, 15(14), 3537; https://doi.org/10.3390/rs15143537 - 14 Jul 2023
Cited by 1 | Viewed by 1628
Abstract
Phenology is a science that is fundamental to crop productivity and is especially sensitive to environmental changes. In Mediterranean and semi-arid climates, vineyard phenology is directly affected by changes in temperature and rainfall distribution, being highly vulnerable to climate change. Due to the [...] Read more.
Phenology is a science that is fundamental to crop productivity and is especially sensitive to environmental changes. In Mediterranean and semi-arid climates, vineyard phenology is directly affected by changes in temperature and rainfall distribution, being highly vulnerable to climate change. Due to the significant heterogeneity in soil, climate, and crop variables, we need fast and reliable ways to assess vineyard phenology in large areas. This research aims to evaluate the performance of the phenological data assimilation model (DA-PhenM) and compare it with phenological models based on meteorological data (W-PhenM) and models based on Sentinel-2 NDVI (RS-PhenM). Two W-PhenM approaches were evaluated, one assessing eco- and endo-dormancy, as proposed by Caffarra and Eccel (CaEc) and the widely used BRIN model, and another approach based on the accumulation of heat units proposed by Parker called the Grapevine Flowering Veraison model (GFV). The DA-PhenM evaluated corresponds to the integration between RS-PhenM and CaEc (EKF-CaEC) and between RS-PhenM and GFV (EKF-GFV). Results show that EKF-CaEc and EKF-GFV have lower root mean square error (RMSE) values than CaEc and GFV models. However, based on the number of parameters that models require, EKF-GFV performs better than EKF-CaEc because the latter has a higher Bayesian Index Criterion (BIC) than EKF-GFV. Thus, DA-PhenM improves the performance of both W-PhenM and RS-PhenM, which provides a novel contribution to the phenological modeling of Vitis vinifera L. cv Cabernet Sauvignon. Full article
(This article belongs to the Special Issue Remote Sensing in Viticulture II)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Data assimilation approach for phenology modeling in vineyards. Model-based on remote sensing data (RS-PhenM). Savitzki–Golay filter (SG-Filter). Grapevine Flowering Veraison model (GFV). Caffarra and Eccel Approach (CaEc). BRIN model. Extended Kalman Filter (EKF) algorithm. Assimilated GFV (EKF-GFV) and assimilated CaEc (EKF-CaEc). Ground data is the reference information that is used for the parameterization of RS-PhenM and W-PhenM. The RS-PhenMs are optimized with SG-Filter and a double Gaussian model. The EKF algorithm builds the DA-PhenM from the RS-PhenM and W-PhenM.</p> Full article ">Figure 2
<p>Vineyard study area. The yellow dot corresponds to the eddy covariance system (EC) position.</p> Full article ">Figure 3
<p>Average daily temperature for S<sub>1</sub> (2017–2018), S<sub>2</sub> (2018–2019), and S<sub>3</sub> (2019–2020). Vertical dotted lines represent the earlier and later observed dates for each stage.</p> Full article ">Figure 4
<p>Cumulative daily evapotranspiration (mm) in S<sub>1</sub> (2017–2018), S<sub>2</sub> (2018–2019), and S<sub>3</sub> (2019–2020). Vertical dotted lines refer to each phenological stage’s earliest and latest observed dates. Cumulative ET in S<sub>3</sub> is lower than in S<sub>2</sub> and S<sub>1</sub>. In the flowering window, cumulative ET in S<sub>3</sub> was about 20% less than cumulative ET in S<sub>2</sub> and S<sub>1</sub>, while in the veraison window, the difference in S<sub>3</sub> compared to S<sub>1</sub> and S<sub>2</sub> was about 34%.</p> Full article ">Figure 5
<p>NDVI time series in S<sub>1</sub> (2017–2018), S<sub>2</sub> (2018–2019), and S<sub>3</sub> (2019–2020) before (<b>a</b>–<b>c</b>) and after applying the Golay–Savitzki filter (<b>d</b>–<b>f</b>).</p> Full article ">Figure 6
<p>Time series of NDVI fitted to a Gaussian model and their first and second derivatives. The dotted lines identify the phenological metric: BB = budburst, FL = Flowering, SE = Setting, and VE = Veraison. (<b>a</b>–<b>c</b>) NDVI fitted to a double Gaussian model. (<b>d</b>–<b>f</b>) First derivative of the double Gaussian NDVI. (<b>g</b>–<b>i</b>) Second derivative of the double Gaussian NDVI.</p> Full article ">Figure 7
<p>Data assimilation skill in percentage (DA<sub>skill</sub> %). It represents the percentage reduction in the Root Mean Square Error (RMSE in days) of the assimilated models (EKF-GFV and EKF-CaEc) compared to the respective non-assimilated models (GFV and CaEc).</p> Full article ">Figure A1
<p>Eddy covariance system oriented in the prevailing direction of daytime winds over the vineyard alignment (north–south). Irgason (infrared gas analyzer and sonic anemometer); PAR (photosynthetically active radiation sensor); Net Radiation (net radiation sensor); T &amp; HR (temperature and relative humidity sensor); TDR (time-domain reflectometry system); Flux Soil (soil sensible heat flux plate system and thermocouples).</p> Full article ">
26 pages, 23625 KiB  
Article
Swin-Transformer-YOLOv5 for Real-Time Wine Grape Bunch Detection
by Shenglian Lu, Xiaoyu Liu, Zixuan He, Xin Zhang, Wenbo Liu and Manoj Karkee
Remote Sens. 2022, 14(22), 5853; https://doi.org/10.3390/rs14225853 - 18 Nov 2022
Cited by 28 | Viewed by 6276
Abstract
Precise canopy management is critical in vineyards for premium wine production because maximum crop load does not guarantee the best economic return for wine producers. The growers keep track of the number of grape bunches during the entire growing season for optimizing crop [...] Read more.
Precise canopy management is critical in vineyards for premium wine production because maximum crop load does not guarantee the best economic return for wine producers. The growers keep track of the number of grape bunches during the entire growing season for optimizing crop load per vine. Manual counting of grape bunches can be highly labor-intensive and error prone. Thus, an integrated, novel detection model, Swin-transformer-YOLOv5, was proposed for real-time wine grape bunch detection. The research was conducted on two varieties of Chardonnay and Merlot from July to September 2019. The performance of Swin-T-YOLOv5 was compared against commonly used detectors. All models were comprehensively tested under different conditions, including two weather conditions, two berry maturity stages, and three sunlight intensities. The proposed Swin-T-YOLOv5 outperformed others for grape bunch detection, with mean average precision (mAP) of up to 97% and F1-score of 0.89 on cloudy days. This mAP was ~44%, 18%, 14%, and 4% greater than Faster R-CNN, YOLOv3, YOLOv4, and YOLOv5, respectively. Swin-T-YOLOv5 achieved an R2 of 0.91 and RMSE of 2.4 (number of grape bunches) compared with the ground truth on Chardonnay. Swin-T-YOLOv5 can serve as a reliable digital tool to help growers perform precision canopy management in vineyards. Full article
(This article belongs to the Special Issue Remote Sensing in Viticulture II)
►▼ Show Figures

Graphical abstract

Graphical abstract
Full article ">Figure 1
<p>Grape dataset acquisition on (<b>a</b>) Chardonnay (white color of berry skin when mature; 10–33 grape bunches per plant) with the close-up views of grape bunch during (<b>b</b>) immature and (<b>c</b>) mature stages, and (<b>d</b>) Merlot (red color of berry skin when mature; 12–32 grape bunches per plant) with the close-up views of grape bunch during (<b>e</b>) immature and (<b>f</b>) mature stages.</p> Full article ">Figure 2
<p>Illustrations of the dataset augmentation: (<b>a</b>) original image; (<b>b</b>) rotation; (<b>c</b>) channel enhancement; (<b>d</b>) Gaussian blur/noise; and (<b>e</b>) rectangle pixel discard.</p> Full article ">Figure 3
<p>(<b>a</b>) Overall architecture of Swin-transformer (H and W refer to height and width of the images; C refers to the number of feature channels) and (<b>b</b>) two successive Swin-transformer blocks (W-MSA and SW-MSA refer to Window-based multi-head self-attention and shifted window-based MSA; MLP refers to multi-layer perceptron).</p> Full article ">Figure 4
<p>(<b>a</b>) Integrated architecture of Swin-transformer-YOLOv5; layers of (<b>b</b>) focus; (<b>c</b>) CBL and cross-stage partial (CSP) bottleneck with 3 convolutions (C3); and (<b>d</b>) spatial pyramid pooling (SPP), where CONV, Concat, BN, SiLU, and ADD (⊕) refer to convolutional, concatenate, batch normalization, activation function of sigmoid linear unit, and feature fusion with the number of channels unchanged. P refers to the specific layer of feature map.</p> Full article ">Figure 5
<p>Precision–recall (P–R) curves of Faster R-CNN, YOLOv3, YOLOv4, YOLOv5, and Swin-transformer-YOLOv5 (Swin-T-YOLOv5) in detecting grape bunches.</p> Full article ">Figure 6
<p>The number of grape bunches comparison between in-field manual counting (gTruth), manual label, and detection using Swin-transformer-YOLOv5 (Swin-T-YOLOv5) with (<b>a</b>) Chardonnay (sunny); (<b>b</b>) Merlot (sunny); (<b>c</b>) Chardonnay (cloudy); and (<b>d</b>) Merlot (cloudy). RMSE refers to root mean square error. Plant# refers to the number of plants.</p> Full article ">Figure 7
<p>The number of grape bunches comparison between in-field manual counting (gTruth), manual label, and detection using Swin-transformer-YOLOv5 (Swin-T-YOLOv5) for (<b>a</b>,<b>b</b>) immature and (<b>c</b>,<b>d</b>) mature berries with Chardonnay (left) and Merlot (right). RMSE refers to root mean square error. Plant# refers to the number of plants.</p> Full article ">Figure 8
<p>The number of grape bunches comparison between in-field manual counting (gTruth), manual label, and detection using Swin-transformer-YOLOv5 (Swin-T-YOLOv5) during (<b>a</b>,<b>b</b>) morning; (<b>c</b>,<b>d</b>) noon; (<b>e</b>,<b>f</b>) afternoon with Chardonnay (left) and Merlot (right). RMSE refers to root mean square error. Plant# refers to the number of plants.</p> Full article ">Figure 9
<p>Detection results on Chardonnay (white variety; two columns on the left) and Merlot (red variety; two columns on the right) using (<b>a</b>–<b>d</b>) generic YOLOv5 (in magenta color) and (<b>e</b>–<b>h</b>) Swin-transformer-YOLOv5 (Swin-T-YOLOv5; in cyan color) in zoomed-in views.</p> Full article ">Figure 10
<p>Illustrations of failures (i.e., true negatives (TN) or false positives (FP) highlighted in red bounding boxes; true positives (TP) in cyan bounding boxes) in grape bunch detection using Swin-transformer-YOLOv5 (Swin-T-YOLOv5) during early (left column), mid- (middle column), and harvest-stage (right column) for (<b>a</b>–<b>c</b>) “Chardonnay (white variety)” and (<b>d</b>–<b>f</b>) “Merlot (red variety)” in zoomed-in views.</p> Full article ">Figure A1
<p>Demonstrations of detection results on the test set of Chardonnay (white variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 (in red color); (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) under sunny (left) and cloudy (right) weathers.</p> Full article ">Figure A1 Cont.
<p>Demonstrations of detection results on the test set of Chardonnay (white variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 (in red color); (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) under sunny (left) and cloudy (right) weathers.</p> Full article ">Figure A2
<p>Demonstrations of detection results on the test set of Merlot (red variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 in red color; (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) under sunny (left) and cloudy (right) weathers.</p> Full article ">Figure A2 Cont.
<p>Demonstrations of detection results on the test set of Merlot (red variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 in red color; (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) under sunny (left) and cloudy (right) weathers.</p> Full article ">Figure A3
<p>Demonstrations of detection results on the test set of Chardonnay (white variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 (in red color); (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) at immature (left) and mature (right) stages.</p> Full article ">Figure A3 Cont.
<p>Demonstrations of detection results on the test set of Chardonnay (white variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 (in red color); (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) at immature (left) and mature (right) stages.</p> Full article ">Figure A4
<p>Demonstrations of detection results on the test set of Merlot (red variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 (in red color); (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) at immature (left) and mature (right) stages.</p> Full article ">Figure A4 Cont.
<p>Demonstrations of detection results on the test set of Merlot (red variety) using (<b>a</b>,<b>b</b>) Faster R-CNN (bounding boxes in green color); (<b>c</b>,<b>d</b>) YOLOv3 (in blue color); (<b>e</b>,<b>f</b>) YOLOv4 (in red color); (<b>g</b>,<b>h</b>) YOLOv5 (in magenta color); and (<b>i</b>,<b>j</b>) Swin-transformer-YOLOv5 (in cyan color) at immature (left) and mature (right) stages.</p> Full article ">Figure A5
<p>Demonstrations of detection results on the test set of Chardonnay (white variety) using (<b>a</b>–<b>c</b>) Faster R-CNN (bounding boxes in green color); (<b>d</b>–<b>f</b>) YOLOv3 (in blue color); (<b>g</b>–<b>i</b>) YOLOv4 (in red color); (<b>j</b>–<b>l</b>) YOLOv5 (in magenta color); and (<b>m</b>–<b>o</b>) Swin-transformer-YOLOv5 (in cyan color) under morning (left), noon (middle), and afternoon (right) sunlight directions/intensities.</p> Full article ">Figure A6
<p>Demonstrations of detection results on the test set of Merlot (red variety) using (<b>a</b>–<b>c</b>) Faster R-CNN (bounding boxes in green color); (<b>d</b>–<b>f</b>) YOLOv3 (in blue color); (<b>g</b>–<b>i</b>) YOLOv4 (in red color); (<b>j</b>–<b>l</b>) YOLOv5 (in magenta color); and (<b>m</b>–<b>o</b>) Swin-transformer-YOLOv5 (in cyan color) under morning (left), noon (middle), and afternoon (right) sunlight directions/intensities.</p> Full article ">

Review

Jump to: Research

27 pages, 4080 KiB  
Review
Satellite Remote Sensing Tools for Drought Assessment in Vineyards and Olive Orchards: A Systematic Review
by Nazaret Crespo, Luís Pádua, João A. Santos and Helder Fraga
Remote Sens. 2024, 16(11), 2040; https://doi.org/10.3390/rs16112040 - 6 Jun 2024
Viewed by 1515
Abstract
Vineyards and olive groves are two of the most important Mediterranean crops, not only for their economic value but also for their cultural and environmental significance, playing a crucial role in global agriculture. This systematic review, based on an adaptation of the 2020 [...] Read more.
Vineyards and olive groves are two of the most important Mediterranean crops, not only for their economic value but also for their cultural and environmental significance, playing a crucial role in global agriculture. This systematic review, based on an adaptation of the 2020 PRISMA statement, focuses on the use of satellite remote sensing tools for the detection of drought in vineyards and olive groves. This methodology follows several key steps, such as defining the approach, selecting keywords and databases, and applying exclusion criteria. The bibliometric analysis revealed that the most frequently used terms included “Google Earth Engine” “remote sensing” “leaf area index” “Sentinel-2”, and “evapotranspiration”. The research included a total of 81 articles published. The temporal distribution shows an increase in scientific production starting in 2018, with a peak in 2021. Geographically, the United States, Italy, Spain, France, Tunisia, Chile, and Portugal lead research in this field. The studies were classified into four categories: aridity and drought monitoring (ADM), agricultural water management (AWM), land use management (LUM), and water stress (WST). Research trends were analysed in each category, highlighting the use of satellite platforms and sensors. Several case studies illustrate applications in vineyards and olive groves, especially in semi-arid regions, focusing on the estimation of evapotranspiration, crop coefficients, and water use efficiency. This article provides a comprehensive overview of the current state of research on the use of satellite remote sensing for drought assessment in grapevines and olive trees, identifying trends, methodological approaches, and opportunities for future research in this field. Full article
(This article belongs to the Special Issue Remote Sensing in Viticulture II)
►▼ Show Figures

Figure 1

Figure 1
<p>Average grape production (t) by country between 2018 and 2022 (<b>a</b>); worldwide grape production evolution between 1961 and 2022 (<b>b</b>); and percentage of vineyard plantation area by continent in 2022 (<b>c</b>). Adapted from FAOSTAT.</p> Full article ">Figure 2
<p>Average olive production (t) by country between 2018 and 2022 (<b>a</b>); worldwide olive production evolution between 1961 and 2022 (<b>b</b>); and percentage of olive plantation area by continent in 2022 (<b>c</b>). Adapted from FAOSTAT.</p> Full article ">Figure 3
<p>PRISMA flow diagram of the systematic literature review search adapted from Moher et al. [<a href="#B34-remotesensing-16-02040" class="html-bibr">34</a>].</p> Full article ">Figure 4
<p>Network connection graph between the top 20 most frequently used keywords in the selected studies. The size of each circle corresponds to the frequency of a keyword’s usage, with larger circles indicating higher usage. The top 5 most used keywords are distinguished in red. The proximity between circles, connected by lines, identifies the degree of connection between the corresponding keywords.</p> Full article ">Figure 5
<p>Temporal distribution of articles included in the systematic review by publication year (<b>a</b>) and the spatial distribution by country (<b>b</b>).</p> Full article ">Figure 6
<p>Distribution of studies expressed as a percentage, categorised by crop type.</p> Full article ">Figure 7
<p>Percentage of publications between 1 January 2003 and 30 December 2023, according to categorical classification by country (<b>a</b>) and by year (<b>b</b>). ADM: aridity and drought monitoring; AWM: agricultural water management; LUM: land use management; and WST: water stress. AF: Afghanistan; AU: Australia; CL: Chile; CN: China; FR: France; GR: Greece; IT: Italy; LB: Lebanon; MD: Moldova; MA: Morocco; PT: Portugal; KSA: Saudi Arabia (Kingdom of); ES: Spain; TN: Tunisia; TR: Türkiye; USA: United States of America.</p> Full article ">Figure 8
<p>Percentage distribution of studies by satellite platform (<b>a</b>) and by sensor (<b>b</b>) based on their respective categorical classification. ADM: aridity and drought monitoring; AWM: agricultural water management; LUM: land use management; and WST: water stress. Note that one study can appear several times, as it may include more than one index.</p> Full article ">Figure 9
<p>The distribution of studies by the use of indices according to their respective categorical classifications. ADM: aridity and drought monitoring; AWM: agricultural water management; LUM: land use management; and WST: water stress. Others correspond to the sum of vegetation indices that have been used only once. Note that one study can appear several times, as it may include more than one index.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-6
Back to TopTop