Detecting vineyards using multispectral UAV imagery and artificial intelligence: A case study from Northern Greece

Estimating mangrove leaf area index based on red-edge vegetation indices: A comparison among UAV, WorldView-2 and Sentinel-2 imagery

Rice lodging may result in serious yield loss. Manual in-situ assessments of lodging are inefficient and inaccurate. Therefore, this paper explores the potential of unmanned aerial vehicle (UAV) image in evaluating rice lodging. Multispectral and red–green–blue (RGB) cameras mounted on UAV platforms were used to acquire images of two rice paddies in Shanghai, China. The image features of non-lodged and lodged rice, including their spectral reflectance, vegetation indices, texture, and colour, were extracted and analysed to optimize the indicators for lodging detection. Rice lodging detection models based on the selected image features were built to discriminate between non-lodged and lodged rice. Results revealed that the reflectances at the green and red-edge bands are the most common optimal indicators of lodged rice in the multispectral images. Moreover, the mean texture of the green channel and lightness value were determined to be the optimal indicators of rice lodging in the RGB image. The accuracies of the proposed lodging detection models exceeded 90%, and the multispectral lodging detection model was proved to be more accurate than the corresponding RGB model. Our results confirm that UAV-based remote sensing can play an important role in assessing rice lodging.

This article aims to develop and evaluate two methodologies for improving the spatial resolution without significant loss of the spectral resolution of a multi-spectral Ikonos image using an image acquired with an UAV. In the first method from multi-spectral image and image acquired with the UAV get the intensity of the multispectral image (Im) and the intensity of UAV (Iv), using the RGB-IHS transformation respectively. Then applying fusion to the intensity components of Im and Iv using Wavelet transform together with ARSIS' technique to genere a new intensity (n_Imv). This n_Imv together with hue and saturation obtained from Ikonos image are transformed inversely to generate a new multispectral image (N-MULT-I). In the second process a new panchromatic is created through of average of three RGB bands of UAV's image (fpan), then fusion is applied using wavelet transform to Im and fpan to generate a new intensity (n_Ifpm). Using n_Ifpm and hue and saturation components of IKonos image a new image multispectral is generated (N-MULT-fpan). Finally, results are shown and there are evaluated with Correlation Coefficient, ERGAS, RASE and Qu. Finding that better results are obtained with the second process.

Vertical profiles of O3, NO2 and PM in a major fine chemical industry park in the Yangtze River Delta of China detected by a sensor package on an unmanned aerial vehicle

Our work aims to monitor wheat crop using a variety-based approach by taking into consideration four different phenological stages of wheat crop development. In addition to highlighting the contribution of Red-Edge vegetation indices in mapping wheat dry matter and nitrogen content dynamics, as well as using Random Forest regressor in the estimation of wheat yield, dry matter and nitrogen uptake relying on UAV (Unmanned Aerial Vehicle) multispectral imagery. The study was conducted on an experimental platform with 12 wheat varieties located in Sidi Slimane (Morocco). Several flight missions were conducted using eBee UAV with MultiSpec4C camera according to phenological growth stages of wheat. The proposed methodology is subdivided into two approaches, the first aims to find the most suitable vegetation index for wheat’s biophysical parameters estimation and the second to establish a global model regardless of the varieties to estimate the biophysical parameters of wheat: Dry matter and nitrogen uptake. The two approaches were conducted according to six main steps: (1) UAV flight missions and in-situ data acquisition during four phenological stages of wheat development, (2) Processing of UAV multispectral images which enabled us to elaborate the vegetation indices maps (RTVI, MTVI2, NDVI, NDRE, GNDVI, GNDRE, SR-RE et SR-NIR), (3) Automatic extraction of plots by Object-based image analysis approach and creating a spatial database combining the spectral information and wheat’s biophysical parameters, (4) Monitoring wheat growth by generating dry biomass and wheat’s nitrogen uptake model using exponential, polynomial and linear regression for each variety this step resumes the varietal approach, (5) Engendering a global model employing both linear regression and Random Forest technique, (6) Wheat yield estimation. The proposed method has allowed to predict from 1 up to 21% difference between actual and estimated yield when using both RTVI index and Random Forest technique as well as mapping wheat’s dry biomass and nitrogen uptake along with the nitrogen nutrition index (NNI) and therefore facilitate a careful monitoring of the health and the growth of wheat crop. Nevertheless, some wheat varieties have shown a significant difference in yield between 2.6 and 3.3 t/ha.

Carbon dioxide (CO2) is one of the greenhouse gases that causes global warming with the highest concentration in the atmosphere. Mangrove forests can absorb CO2 three times higher than terrestrial forests and tropical rainforests. Moreover, mangrove forests can be a source of Indonesian income in the form of a blue economy, therefore an accurate method is needed to investigates mangrove carbon stock. Utilization of remote sensing data with the results of the above-ground carbon (AGC) detection model of mangrove forests based on multispectral imaging and vegetation index, can be a solution to get fast, cheap, and accurate information related to AGC estimation. This study aimed to investigates the best model for estimating the AGC of mangroves using Sentinel-2 imagery in Benoa Bay, Bali Province. The random forest (RF) method was used to classified the difference between mangrove and non-mangrove with the treatment of several parameters. Furthermore, a semi-empirical approach was used to assessed and map the AGC of mangroves. Allometric equations were used to calculated and produced AGC per species. Moreover, the model was built with linear regression equations for one variable x, and multiple regression equations for more than one x variable. Root Mean Square Error (RMSE) was used to assess the validation of the model results. The results of the mangrove forests area detected in the research location around 1134.92 ha, with an Overall Accuracy (OA) of 0.984 and a kappa coefficient of 0.961. This study highlights that the best model was the combination of IRECI and TRVI vegetation indices (RMSE: 11.09 Mg/ha) for a model based on red edge bands. Meanwhile, the best results from the model that does not use the red edge band were the combination of TRVI and DVI vegetation indices (RMSE: 13.63 Mg/ha). The use of red edge and NIR bands is highly recommended in building the AGC model of mangrove forests because they can increase the accuracy value. Thus, the results of this study are highly recommended in estimating the AGC of mangrove forests, because it has been proven to be able to increase the accuracy value of previous studies using optical images.

Forest floor dead fuel moisture content (DFMC) is an important factor in the occurrence of forest fires, and predicting DFMC is important for accurate fire risk forecasting. Large areas of forest surface DFMC are difficult to predict via manual methods. In this paper, we propose an unmanned aerial vehicle (UAV)-based forest surface DFMC prediction method, in which a UAV is equipped with a multispectral camera to collect multispectral images of dead combustible material on the forest surface over a large area, combined with a deep-learning algorithm to achieve the large-scale prediction of DFMC on the forest surface. From 9 March to 23 March 2023, 5945 multispectral images and 480 sets of dead combustible samples were collected from an urban forestry demonstration site in Harbin, China, using an M300 RTK UAV with an MS600Pro multispectral camera. The multispectral images were segmented by a K-means clustering algorithm to obtain multispectral images containing only dead combustibles on the ground surface. The segmented multispectral images were then trained with the actual moisture content measured by the weighing method through the ConvNeXt deep-learning model, with 3985 images as the training set, 504 images as the validation set, and 498 images as the test set. The results showed that the MAE and RMSE of the test set are 1.54% and 5.45%, respectively, and the accuracy is 92.26% with high precision, achieving the accurate prediction of DFMC over a large range. The proposed new method for predicting DFMC via UAV multispectral cameras is expected to solve the real-time large-range accurate prediction of the moisture content of dead combustible material on the forest surface during the spring fire-prevention period in northeast China, thus providing technical support for improving the accuracy of forest fire risk-level forecasting and forest fire spread trend prediction.

Imaging from an unmanned aerial vehicle: agricultural surveillance and decision support

The availability, cost, and applicability of Unmanned Aerial Vehicles (UAVs) are rapidly altering the way we perceive the landscape, can access sites, and model the landscape as it unfolds through remote sensing technologies. It falls to present and future researchers to take advantage of this and use the technology to improve our understanding of our surroundings. One way that geomorphologists would be able to take advantage of the improvements in technology is to use the UAV to map soil processes, in particular erosion. The current methods and technology for mapping gully erosion do not look at the smaller features but rather the landscape as a whole. The purpose of this research was to assess the prospects of UAV technology mapping gully erosion. Specifically, the study investigates the application of the UAV imagery to other technologies or methods in mapping gully erosion activities. This was achieved by testing the multiple available modelling software, by mapping objects of known shapes and volumes to determine which software produced results comparable to that of the known shapes. Tests were conducted to determine if this software and UAV combination would have the ability to map and 3D render soil erosion, thus allowing researchers to verify if this technique could be implemented into mapping gully systems. From these tests, it was found the UAV was able to improve the resolution to that of the available methods. Obtaining a resolution of 0.03m would allow geomorphologists to map and situate the soil erosional landscapes and map the test objects. The software package AgiSoft MetaShape had the highest accuracy and reliability in mapping various objects making it ideal in the mapping of smaller erosional features. Multiple erosional landscapes were mapped using a UAV and the attached RGB sensor and a second UAV with a multi-spectral sensor and AgiSoft MetaShape software. Both sensors mapped plant health, capturing plant health with band widths related to each sensor. Ideally the multi-spectral sensors were seen to be more versatile, however the RGB sensor can be useful. The combination of both UAVs created a 3D-rendered model, and the resulting data was useful in determining the potential future areas of erosion and existing areas of high erosion risk. The multi-sensor camera allowed the user to determine micro-features that could potentially become erosional areas in the future as it identified areas with a concentration in water flow. In conclusion, the UAV is a useful spatial tool to monitor, measure and manage soil erosion and gully mapping. Iqoqa. Ukusetshenziswa kweUAV kwenza izinguquko ngokushesha endleleni okubukwa ngayo izwe (landscape) kanti kukubacwaningi ukuthi sithathe ithuba lokusebenzisa lensizakusebenza ngendlela ezoba wusizo ukuze sibe nokuqonda okungcono kolwazi olukhona. Lobuchwepheshe bungasetshenziswa ukubona ukuqopha ukunyakaza komhlabathi ngoguguleka kwemihosha . Inhloso kube wukuhlola ukusetshenziswa kwe UAV ukubona ukuthi iyakwazi yini ukukhiqiza izithombe ezithathwe ngobuchwepheshe basemoyeni obusezingeni eliphezulu. Inhloso yenzekie ngokuthi kwenziwe ukuhlola okunhlobonhlobo kwimodelling software. Lokhu kwenziwe ngezimpawu zebalazwe zomumo owaziwayo kanye nezinga lomsindo, ukuqonda ukuthi ilona luphi uhlelokusebenza olusebenzayo ngezinga lomsindo oseduze . Ukuhlolwa kwenziwe ukubona ukuthi lenhlanganisela yohlelokusebenza ne UAV ingakwazi yini ukuveza ukuguguleka kwenhlabathi igugulwa ngebalalazwe nemodli ye3D kanye nokuqinisekisa ukuthi lobu buchwepheshe bungasetshenziswa ebalazweni lwemihosha kanye nemodeli yekhompyutha eyaziwa nge 3D computer models of gullies. Kulokhu kuhlola kutholakale ukuthi iUAV iyakwazi ukuthuthukisa ukuhleleka kohlelo okukhona. Ukuthola ukuhleleka okungango 0.03 m okungavumela igeomorphologist ukungena ebalazweni kuphinde kuveze ukuguguleka komhlabathi. Izinjongo eziwusayizi owaziwayo zifakwe ebalazweni, okuzale iAgiSoft MetaShape okube sezingeni eliphezulu lokuqagula kanye nokwethembeka ekufakeni ebalazweni izinto ezahlukahlukene . Imiphumela yemininingo ibe wusizo ukuqonda izindawo ezingaguguleka ngengomuso kanye nezindawo ezikhona ezisethubeni lobungozi bokuguguleka .Ikhamera eyimulti sensor yenze umcwaningi ukuthi abone izimpawu ezaziwa ngemicro features ezingaba sethubeni lokuba sendaweni engaguguleka ngengomuso njengoba zikhombisa ukuba nenkinga yobuningi kwamanzi agelezayo. Ukuphetha, i-UAV okwamanje iyinsizakusebenza ewusizo engasetshenziswa ocwaningweni lwegeomorphologists futhi ishibhile kunezinye izinsizakusebenza ezifana nemanned vehicle noma isatellite.

Unmanned aerial vehicle (UAV) imagery is considered the best approach to acquire high-resolution data over a large area in a short time, such as in the case of coastal wetlands comprising small-sized patches of perennial plants. However, to the best of our knowledge, no previous studies have used UAV imagery to quantitatively estimate blue carbon in coastal wetlands. Therefore, this study aimed to estimate the area-wide carbon stocks accumulated in halophyte habitats through the case studies of the southern coastal wetlands in South Korea using UAV imagery. Accordingly, object-based image analysis was used to identify the spatial distribution of major halophytes based on the spectral reflectance derived from multispectral UAV images. Elemental analysis (EA) was performed to estimate the content of organic carbon accumulated in individual halophytes, and the halophyte communities identified in the case study area were confirmed to have accumulated 20.80 MgC of blue carbon. These results demonstrate that applying EA to UAV images can be an innovative method to solve the spatial uncertainty encountered in the monitoring of blue carbon in existing field-based surveys and traditional remote sensing.

Abstract. Early stress detection is critical for proactive field management and terminal yield prediction, and can aid policy making for improved food security in the context of climate change and population growth. Field surveys for crop monitoring are destructive, labor-intensive, time-consuming and not ideal for large-scale spatial and temporal monitoring. Recent technological advances in Unmanned Aerial Vehicle (UAV) and high-resolution satellite imaging with frequent revisit time have proliferated the applications of this emerging new technology in precision agriculture to address food security challenges from regional to global scales. In this paper, we present a concept of UAV and satellite virtual constellation to demonstrate the power of multi-scale imaging for crop monitoring. Low-cost sensors integrated on a UAV were used to collect RGB, multispectral, and thermal images during the growing season in a test site established near Columbia, Missouri, USA. WorldView-3 multispectral data were pan-sharpened, atmospherically corrected to reflectance and combined with UAV data for temporal monitoring of early stress. UAV thermal and multispectral data were calibrated to canopy temperature and reflectance following a rigorous georeferencing and ortho-correction. The results show that early stress can be effectively detected using multi-temporal and multi-scale UAV and satellite observation; the limitations of satellite remote sensing data in field-level crop monitoring can be overcome by using low altitude UAV observations addressing not just mixed pixel issues but also filling the temporal gap in satellite data availability enabling capture of early stress. The concept developed in this paper also provides a framework for accurate and robust estimation of plant traits and grain yield and delivers valuable insight for high spatial precision in high-throughput phenotyping and farm field management.

In recent years, unmanned aerial vehicles (UAVs) have been actively applied in the agricultural sector. Several UAVs equipped with multispectral cameras have become available on the consumer market. Multispectral data are informative and practical for evaluating the greenness and growth status of vegetation as well as agricultural crops. The precise monitoring of rice paddy, especially in the Asian region, is crucial for optimizing profitability, sustainability, and protection of agro-ecological services. This paper reports and discusses our findings from experiments conducted to test four different commercially available multispectral cameras (Micesense RedEdge-M, Sentera Single NDVI, Mapir Survey3, and Bizworks Yubaflex), which can be mounted on a UAV in monitoring rice paddy. The survey has conducted in the typical paddy field area located in the alluvial plain in Tottori Prefecture, Japan. Six different vegetation indices (NDVI, BNDVI, GNDVI, VARI, NDRE and MCARI) captured by UAVs were also compared and evaluated monitoring contribution at three different rice cropping phases. The results showed that the spatial distribution of NDVI collected by each camera is almost similar in paddy fields, but the absolute values of NDVI differed significantly from each other. Among them, the Sentera camera showed the most reasonable NDVI values of each growing phase, indicating 0.49 in the early reproductive phase, 0.62 in the late reproductive stage, and 0.38 in the ripening phase. On the other hand, compared to the most commonly used NDVI, VARI which can be calculated from only visible RGB bands, can be used as an easy and effective index for rice paddy monitoring.

The experiments were conducted in a fully-productive olive orchard (cv. Frantoio) at the experimental farm of University of Pisa at Venturina (Italy) in 2015 to assess the ability of an unmanned aerial vehicle (UAV) equipped with RGB-NIR cameras to estimate leaf area index (LAI), tree height, canopy diameter and canopy volume of olive trees that were either irrigated or rainfed. Irrigated trees received water 4–5 days a week (1348 m3 ha-1), whereas the rainfed ones received a single irrigation of 19 m3 ha-1 to relieve the extreme stress. The flight altitude was 70 m above ground level (AGL), except for one flight (50 m AGL). The Normalized Difference Vegetation Index (NDVI) was calculated by means of the map algebra technique. Canopy volume, canopy height and diameter were obtained from the digital surface model (DSM) obtained through automatic aerial triangulation, bundle block adjustment and camera calibration methods. The NDVI estimated on the day of the year (DOY) 130 was linearly correlated with both LAI and leaf chlorophyll measured on the same date (R2 = 0.78 and 0.80, respectively). The correlation between the on ground measured canopy volumes and the ones by the UAV-RGB camera techniques yielded an R2 of 0.71–0.86. The monthly canopy volume increment estimated from UAV surveys between (DOY) 130 and 244 was highly correlated with the daily water stress integral of rainfed trees (R2 = 0.99). The effect of water stress on the seasonal pattern of canopy growth was detected by these techniques in correspondence of the maximum level of stress experienced by the rainfed trees. The highest level of accuracy (RMSE = 0.16 m) in canopy height estimation was obtained when the flight altitude was 50 m AGL, yielding an R2 value of 0.87 and an almost 1:1 ratio of measured versus estimated canopy height.

High-resolution images obtained with multispectral cameras mounted on unmanned aerial vehicles (UAVs) offer significant potential for agricultural monitoring activities. Chlorophyll content is one of the important parameters in monitoring the growing periods of sunflower plants. In this study, the potential of UAV technology to monitor spatial and temporal changes of sunflower (Helianthus annuus) chlorophyll content was tested. Chlorophyll and vegetation indices calculated using multispectral images taken from the UAV were compared with field chlorophyll measurements using a Spad meter. Significant (p<0.01) high correlations were obtained between the calculated indices and chlorophyll values. The results revealed that the chlorophyll content of sunflower plants can reliably be modeled at p<0.001 significance level using a multivariate linear regression model. The model obtained could successfully explain 28% 𝑅𝐴𝑑𝑗 2 = 0.28 of the variation in chlorophyll content of sunflowers. The results showed that the models developed between Spad measurement and UAVimages were successful in explaining the variation at a high level of importance. However, although statistically significant, the indices produced in the study could only explained 28% of the variation in chlorophyll content of sunflower plants. Keywords: Sunflower, UAV, Spad meter, Estimation, Vegetation indices

Climatic variability and extreme weather events impact agricultural production, especially in sub-Saharan smallholder cropping systems, which are commonly rainfed. Hence, the development of early warning systems regarding moisture availability can facilitate planning, mitigate losses and optimise yields through moisture augmentation. Precision agricultural practices, facilitated by unmanned aerial vehicles (UAVs) with very high-resolution cameras, are useful for monitoring farm-scale dynamics at near-real-time and have become an important agricultural management tool. Considering these developments, we evaluated the utility of optical and thermal infrared UAV imagery, in combination with a random forest machine-learning algorithm, to estimate the maize foliar temperature and stomatal conductance as indicators of potential crop water stress and moisture content over the entire phenological cycle. The results illustrated that the thermal infrared waveband was the most influential variable during vegetative growth stages, whereas the red-edge and near-infrared derived vegetation indices were fundamental during the reproductive growth stages for both temperature and stomatal conductance. The results also suggested mild water stress during vegetative growth stages and after a hailstorm during the mid-reproductive stage. Furthermore, the random forest model optimally estimated the maize crop temperature and stomatal conductance over the various phenological stages. Specifically, maize foliar temperature was best predicted during the mid-vegetative growth stage and stomatal conductance was best predicted during the early reproductive growth stage. Resultant maps of the modelled maize growth stages captured the spatial heterogeneity of maize foliar temperature and stomatal conductance within the maize field. Overall, the findings of the study demonstrated that the use of UAV optical and thermal imagery, in concert with prediction-based machine learning, is a useful tool, available to smallholder farmers to help them make informed management decisions that include the optimal implementation of irrigation schedules.