UAV Images Research Articles

Glacier, permafrost and thermokarst interactions in Alpine terrain: Insights from seven decades of reconstructed dynamics of the Chauvet glacial and periglacial system (Southern French Alps)

AbstractThis study analyses the long‐term dynamics in the Chauvet glacial and periglacial system (southern French Alps) over seven decades (1948–2020), where several lake outburst floods have been documented since 1930. To accurately describe and explain the complex dynamics of this site, our multidisciplinary approach combines (1) photogrammetry of historical aerial photographs and modern high‐resolution satellite and UAV images, (2) geophysical surveys and (3) geomorphological mapping. We provide evidence for spatial and functional interactions between glacial and periglacial features, especially in the lower sector where different landforms with variable ice‐ and debris‐content and specific dynamics are interplayed. We found the highest thinning rates on ice‐rich terrain located in the central part of the valley bottom, which, together with bedrock morphology, most probably determine the location of the thermokarst. We also documented an overall acceleration of the creeping of the landforms after the 2000s, with a flow direction largely oriented towards the thermokarst depression. The outburst water flowed through a conduit whose successive opening and closure seem to mainly depend on the rate of lateral convergence of left‐ and right‐hand landforms and on the rate of ice melting (and roof collapse) along the conduit walls. Today, the site of Chauvet still represents a potential hazard for the region due to the large water storage capacity (up to 180 000 ± 450 m3) and the development of a predominantly bucket shape in the thermokarst sector.

CHARACTERIZATION OF SOME AROMATIC AND MEDICINAL PLANTS SPECIES BASED ON UAV IMAGES

The study used imaging analysis to characterize several species of aromatic and medicinal plants. UAV images were taken on five aromatic and medicinal plants species, cultivated by SCDA Lovrin, Timis county, Romania. The biological material was represented by Lavandulua x intermedia (Lavandin), P1; Hyssopus officinalis L., P2, P6; Origanum vulgare L., P3; Thymus serpyllum L., P4; Salvia officinalis L., P5 (P1 to P6 species code in the article). The images were analyzed (four repetitions) and the values of the RGB color parameters were obtained. The ANOVA test confirmed the presence of variance and the safety of the data (F>F crit, p<0.001; Alpha=0.001). The correlation analysis highlighted positive and negative correlations of different intensity levels between RGB parameters and calculated ratios (R/G, R/B, G/R, G/B, B/R, B/G). The variation of G in relation to R was described by a polynomial equation of degree 3 (R2=0.956), and the variation of B in relation to R was described by a polynomial equation of degree 2 (R2=0.736). High variability was recorded in the case of parameter B (CVB=23.44534), plant P3 (Origanum vulgare L.), and low variability was recorded in the case of color parameter G (CVG=2.66030), plant P2 (Hyssopus officinalis L.). Based on PCA, the distribution diagram of plant species in relation to RGB color parameters and calculated ratios, as biplot, was obtained. PC1 explained 66.037% of variance, and PC2 explained 33.095% of variance in relation to RGB, and PC1 explained 58.923% of variance, and PC2 explained 36.46% of variance in relation to the calculated ratios. The cluster analysis facilitated the grouping of plant species (P1 to P6) based on similarity in relation to the RGB color parameter values (Coph.corr. = 0.959).

Multi-temporal NDVI analysis using UAV images of tree crowns in a northern Mexican pine-oak forest

Multi-temporal NDVI analysis using UAV images of tree crowns in a northern Mexican pine-oak forest

Optimization of desert lake information extraction from remote sensing images using cellular automata

Desert lakes are important wetland resources in the blown-sand area of western China and play a significant role in maintaining the regional ecological environment. However, large-scale coal mining in recent years has considerably impacted the deposition condition of several lakes. Rapid and accurate extraction of lake information based on satellite images is crucial for developing protective measures against desertification. However, the spatial resolution of these images often leads to mixed pixels near water boundaries, affecting extraction precision. Traditional pixel unmixing methods mainly obtain water coverage information in a mixed pixel, making it difficult to accurately describe the spatial distribution. In this paper, the cellular automata (CA) model was adopted in order to realize lake information extraction at a sub-pixel level. A mining area in Shenmu City, Shaanxi Province, China is selected as the research region, using the image of Sentinel-2 as the data source and the high spatial resolution UAV image as the reference. First, water coverage of mixed pixels in the Sentinel-2 image was calculated with the dimidiate pixel model and the fully constrained least squares (FCLS) method. Second, the mixed pixels were subdivided to form the cellular space at a sub-pixel level and the transition rules are constructed based on the water coverage information and spatial correlation. Lastly, the process was implemented using Python and IDL, with the ArcGIS and ENVI software being used for validation. The experiments show that the CA model can improve the sub-pixel positioning accuracy for lake bodies in mixed pixel image and improve classification accuracy. The FCLS-CA model has a higher accuracy and is able to identify most water bodies in the study area, and is therefore suitable for desert lake monitoring in mining areas.

Integration of an Innovative Atmospheric Forecasting Simulator and Remote Sensing Data into a Geographical Information System in the Frame of Agriculture 4.0 Concept

In a world in continuous evolution and in which human needs grow exponentially according to the increasing world population, the advent of new technologies plays a fundamental role in all fields of industry, especially in agriculture. Optimizing times, automating machines, and guaranteeing product quality are key objectives in the field of Agriculture 4.0, which integrates various innovative technologies to meet the needs of producers and consumers while guaranteeing respect for the environment and the planet’s resources. In this context, our research aims to propose an integrated system using data coming from an innovative experimental atmospheric and forecasting simulator (capable of predicting some characteristic climate variables subsequently validated with local sensors), combined with indices deriving from Remote Sensing and UAV images (treated with the data fusion method), that can give fundamental information related to Agriculture 4.0 with particular reference to the subsequent phases of system automation. These data, in fact, can be collected in an open-source GIS capable of displaying areas that need irrigation and fertilization and, moreover, establishing the path of an automated drone for the monitoring of the crops and the route of a self-driving tractor for the irrigation of the areas of interest.

UAV-Based Computer Vision System for Orchard Apple Tree Detection and Health Assessment

Accurate and efficient orchard tree inventories are essential for acquiring up-to-date information, which is necessary for effective treatments and crop insurance purposes. Surveying orchard trees, including tasks such as counting, locating, and assessing health status, plays a vital role in predicting production volumes and facilitating orchard management. However, traditional manual inventories are known to be labor-intensive, expensive, and prone to errors. Motivated by recent advancements in UAV imagery and computer vision methods, we propose a UAV-based computer vision framework for individual tree detection and health assessment. Our proposed approach follows a two-stage process. Firstly, we propose a tree detection model by employing a hard negative mining strategy using RGB UAV images. Subsequently, we address the health classification problem by leveraging multi-band imagery-derived vegetation indices. The proposed framework achieves an F1-score of 86.24% for tree detection and an overall accuracy of 97.52% for tree health assessment. Our study demonstrates the robustness of the proposed framework in accurately assessing orchard tree health from UAV images. Moreover, the proposed approach holds potential for application in various other plantation settings, enabling plant detection and health assessment using UAV imagery.

Analysis of forest tree dieback using UltraCam and UAV imagery

ABSTRACT In recent years, increasing tree diebacks and mortality in some forests, particularly in forest parks, created a need amongst forest managers to find effective methods to gather information about the rate of dieback and mortality and their reasons. High-quality air and space-born remote sensing data has established as an alternative to field surveys for certain inventory tasks. This study used high-quality UltraCam-Xp and UAV drone images from 2016 and 2021 to map tree dieback and mortality in Daland Forest Park, Golestan Province, Iran. High-quality ortho mosaics and Digital Surface Models (DSMs) were generated from UltraCam (2016) and UAV (2021) images. The images were then classified through object-based classification by Nearest Neighbor (NN), Support Vector Machine (SVM), and Bayes algorithms using various input data sets including spectral bands, Canopy Height Model (CHM), vegetation indices, and texture analysis features. Our results indicate that the Bayes algorithm is more precise in mapping tree dieback for the two time steps compared to other algorithms. The best tree dieback map on UltraCam images was obtained using the spectral bands with CHM, texture analysis features, and vegetation indices. This combination resulted in an overall accuracy of 91.20% and a Kappa coefficient of 0.88. It was also found that combining the UAV main bands with CHM and texture features did produce a high-accuracy map with an overall accuracy of 88.46% and a Kappa coefficient of 0.84. Change detection analysis of tree dieback showed that between 2016 and 2021, the number of healthy trees decreased, and the number of gaps and open areas increased in the study area. We conclude that UltraCam and UAV photographs can serve to identify and map tree dieback and dead trees with good accuracies and can hence support forest health monitoring.

The effect of DEM resolution on topographic wetness index calculation and visualization: An insight to the hidden danger unraveled in Bozkurt in August, 2021

Topographic Wetness Index, also known as the compound topographic index, (TWI) is a topographic indicator that calculates the potential of where water is likely to accumulate during excessive precipitation cycles resulting from abrupt atmospheric anomalies. High index values represent serious potential of water accumulation due to low slope, and the opposite for high slope. As expected from the term, slope, Digital Elevation Model (DEM) datasets play an important role in the calculation of TWI. DEMs are produced utilizing tachometry, GPS benchmarking, UAV, aerial or satellite image capture and LIDAR capabilities. However, no matter how it is generated, a DEM is as good as the actual ground sampling algorithm, on which the final resolution is based. Using six different DEM resolutions coming from three global and one national source presented in three different setting coverages, upper feeder basin of Bozkurt sub-province, Kastamonu, was analyzed emphasizing the urbanized part of the sub-province, which was devastated during the August 11th, 2021 flood. Coarser resolution missed the overall precision while the finer resolution captured it nicely. On the flip side, finer resolution excessively fragmented the questioned area while the coarser resolution formed a unity coinciding with the destructed area recorded during the event.

An Image Recognition Method Based on Dynamic System Synchronization

Abstract: At present, image recognition technology first classifies images and outputs category information through the neural network. The next step involves the search. Before retrieval, the feature database needs to be established, followed by one-to-one correspondence. This method is tedious, time-consuming and has low accuracy. In computer vision research, researchers have proposed various image recognition methods to be applied in various fields and made many research achievements. However, at present, the accuracy, stability and time efficiency cannot meet the needs of practical work. In terms of UAV image recognition, high accuracy and low consumption are required. Previous methods require huge databases, which increases the consumption of UAVs. Taking aerial transmission and line images as the research object, this paper proposes a method of image recognition based on chaotic synchronization. Firstly, the image is used as a function to construct a dynamic system, and the function structure and parameters are adjusted to realize chaos synchronization. In this process, different types of images are identified. At the same time, we research this dynamic system characteristics and realize the mechanism of image recognition. Compared with other methods, the self-built aerial image data set for bird's nest identification, iron frame identification and insulator identification has the characteristics of a high identification rate and less calculation time. It is preliminarily proven that the method of synchronous image recognition is practical, and also worthy of further research, verification and analysis.

TOWARDS UAV ASSISTED MONITORING OF AQUATIC VEGETATION WITHIN LARGE RIVERS – THE MIDDLE DANUBE (SERBIA)

UAV technologies provide a time- and cost-efficient framework for a variety of environmental monitoring domains. It also increases data resolution and provides new insights into observed objects and phenomena, especially within the difficult-to-access and complex for monitoring aquatic habitats. The objective of this study was to develop UAV-based acquisition and GIS-based image processing guidelines for aquatic macrophyte detection and monitoring in large temperate rivers. According to the European standard CEN EN -14184:2014, the assessment of aquatic macrophytes should be performed using the transect approach. Large rivers, such as the Danube, represent an exception and should be evaluated using 1km transects. Therefore, seven transects of the Middle Danube in Serbia were simultaneously surveyed using traditional field methods and novel UAV technology. UAV images were acquired using RGB and multispectral cameras carried by a fixed-wing drone. The images were processed and orthomosaics were classified using Object Based Image Analysis (OBIA), to create digital GIS maps of the river transects. During the traditional monitoring approach, the relative abundance of 22 macrophyte species was recorded along the transects. Using the UAV technology and OBIA approach eight macrophyte classes were distinguished based on dominant macrophyte taxa or plant life form traits. Aquatic macrophytes were 'almost perfectly' distinguished from the orthomosaics, achieving a high classification accuracy of 96 % / 88 % / 0.84 for RGB and 94 % / 97 % / 0.95 Producers /Users accuracy/Kappa index for the multispectral approach. Individual macrophyte classes accuracy varied between 0.5 and 1 Kappa and were generally higher for the multispectral imagery approach. Although the resolution of the taxonomic data is lower, UAV monitoring provided the necessary spatial context of macrophytes distribution and absolute area occupied by macrophytes. It also provided information on the diversity and distribution of habitats along the river. Therefore, the UAV-assisted monitoring approach described in this study can be effectively integrated into macrophyte monitoring during large river expeditions such as the JDS.

산불피해강도에 따른 이산화탄소 배출량 추정

In this study, carbon dioxide emission from the 2021 Andong forest fire was analyzed with burn severity, which was classified into high(crown fire), moderate(heat damage), and low(surface fire). Drone videos of the wildland fire area were taken using a multispectral sensor based on UAV and classified by grouping the differences in derived NDVI values between satellite images before the fire and UAV images immediately after the fire. Fuel load consumption and carbon dioxide emission from forest fires were estimated through field investigation of each burn severity site. According to the results of this study, the total burned areas were calculated to be 350ha, of which H(crown fire), M(heat damage), and L(surface fire) were to be 85ha, 125ha, and 141ha, respectively. 15.40ton/ha fire carbon dioxide in H(crown fire), 4.92ton/ha in M(heat damage), 4.69ton/ha in L(surface fire) were emitted from the fire. The total amount of carbon dioxide emitted by the Andong forest fire in 2021 was estimated to be 2,570 tons.

Flood Analysis by Utilizing UAV Images: The Case of Çan (Kocabaş) Stream

Geçmişten günümüze meydana gelen, maddi ve manevi sorunlara yol açan doğal afetler, doğal ve beşeri faktörlerin birleşik etkileriyle ortaya çıkmaktadır. Doğal afetlere neden olan doğal faktörler, gerçekleşen olayın büyüklüğü, yerleşim alanlarına olan uzaklığı ve yerleşim alanlarının etkileme miktarıdır. Doğal afetlere neden olan beşeri faktörler ise; az gelişmişlik, nüfus artışının dengesiz olması, bilinçsiz yerleşim ve eğitimsizlik gibi nedenlerdir. Doğal afetlere deprem, toprak kayması, kuraklık, sel ve taşkınlar gibi pek çok örnek verilebilmektedir. Doğal afetlerden biri olan ve bu çalışmanın da konusunu oluşturan taşkınlar genel anlamıyla, bir akarsuyun, doğal veya yapay olarak, mevcuttaki yatağından taşarak yatak sınırları dışında akışını devam ettirmesi sonucunda çevreye zarar vermesi olayı olarak ifade edilebilmektedir. Taşkınlar ülkemizi en çok etkileyen doğal afetlerden biridir. Buna göre taşkınlar can ve mal kayıpları açısından ülkemizde ikinci sırada yer alırken, meteorolojik afetler içerisinde ise ilk sırada yer almaktadır. Bu çalışmada Çan ve Biga ilçelerini kapsayan, Biga (Çan) Çayı Havzasının, Çan ilçe merkezinden geçen bölümünün eEbee X model sabit kanatlı İnsansız Hava Aracı (İHA) aracılığıyla görüntüleri elde edilmiş, elde edilen bu görüntüler çeşitli fotogrametrik değerlendirme programları kullanılarak 3 boyutlu model haline getirilmiş ve taşkın analizi için gerekli veriler elde edilmiştir. Son olarak elde edilen bu çıktı ürünler, ilgili kurumlardan temin edilen verilerle birlikte kullanılarak Çan ilçesine ait 100 ve 500 yıllık taşkın haritaları koordinatlı bir şekilde üretilmiştir.

Image Mapping Accuracy Evaluation Using UAV with Standalone, Differential (RTK), and PPP GNSS Positioning Techniques in an Abandoned Mine Site.

Global navigation satellite systems (GNSSs) provide a common positioning method that utilizes satellite signals to determine the spatial location of a receiver. However, there are several error factors in standalone GNSS positioning due to instrumental, procedural, and environmental factors that arise during the signal transmission process, and the final positioning error can be up to several meters or greater in length. Thus, real-time kinematic (RTK) correction and post-mission precise point positioning (PPP) processing technologies are proposed to improve accuracy and accomplish precise position measurements. To evaluate the geolocation accuracy of mosaicked UAV images of an abandoned mine site, we compared each orthomosaic image and digital elevation model obtained using standalone GNSS positioning, differential (RTK) GNSS positioning, and post-mission PPP processing techniques. In the three types of error evaluation measure (i.e., relative camera location error, ground control points-based absolute image mapping error, and volumetric difference of mine tailings), we found that the RTK GNSS positioning method obtained the best performance in terms of the relative camera location error and the absolute image mapping error evaluations, and the PPP post-processing correction effectively reduced the error (69.5% of the average total relative camera location error and 59.3% of the average total absolute image mapping error) relative to the standalone GNSS positioning method. Although differential (RTK) GNSS positioning is widely used in positioning applications that require very high accuracy, post-mission PPP processing can also be used in various fields in which it is either not feasible to operate expensive equipment to receive RTK GNSS signals or network RTK services are unavailable.

A Ground Point Fitting Method for Winter Wheat Height Estimation Using UAV-Based SfM Point Cloud Data

Height is a key factor in monitoring the growth status and rate of crops. Compared with large-scale satellite remote sensing images and high-cost LiDAR point cloud, the point cloud generated by the Structure from Motion (SfM) algorithm based on UAV images can quickly estimate crop height in the target area at a lower cost. However, crop leaves gradually start to cover the ground from the beginning of the stem elongation stage, making more and more ground points below the canopy disappear in the data. The terrain undulations and outliers will seriously affect the height estimation accuracy. This paper proposed a ground point fitting method to estimate the height of winter wheat based on the UAV SfM point cloud. A canopy slice filter was designed to reduce the interference of middle canopy points and outliers. Random Sample Consensus (RANSAC) was applied to obtain the ground points from the valid filtered point cloud. Then, the missing ground points were fitted according to the known ground points. Furthermore, we achieved crop height monitoring at the stem elongation stage with an R2 of 0.90. The relative root mean squared error (RRMSE) of height estimation was 5.9%, and the relative mean absolute error (RMAE) was 4.6% at the stem elongation stage. This paper proposed the canopy slice filter and fitting missing ground points. It was concluded that the canopy slice filter successfully optimized the extraction of ground points and removed outliers. Fitting the missing ground points simulated the terrain undulations effectively and improved the accuracy.

Knowledge discovery of suspicious objects using hybrid approach with video clips and UAV images in distributed environments: a novel approach

Knowledge discovery of suspicious objects using hybrid approach with video clips and UAV images in distributed environments: a novel approach

SCA-YOLO: a new small object detection model for UAV images

SCA-YOLO: a new small object detection model for UAV images

MGFNet: A Progressive Multi-Granularity Learning Strategy-Based Insulator Defect Recognition Algorithm for UAV Images

Due to the low efficiency and safety of a manual insulator inspection, research on intelligent insulator inspections has gained wide attention. However, most existing defect recognition methods extract abstract features of the entire image directly by convolutional neural networks (CNNs), which lack multi-granularity feature information, rendering the network insensitive to small defects. To address this problem, we propose a multi-granularity fusion network (MGFNet) to diagnose the health status of the insulator. An MGFNet includes a traversal clipping module (TC), progressive multi-granularity learning strategy (PMGL), and region relationship attention module (RRA). A TC effectively resolves the issue of distortion in insulator images and can provide a more detailed diagnosis for the local areas of insulators. A PMGL acquires the multi-granularity features of insulators and combines them to produce more resilient features. An RRA utilizes non-local interactions to better learn the difference between normal features and defect features. To eliminate the interference of the UAV images’ background, an MGFNet can be flexibly combined with object detection algorithms to form a two-stage object detection algorithm, which can accurately identify insulator defects in UAV images. The experimental results show that an MGFNet achieves 91.27% accuracy, outperforming other advanced methods. Furthermore, the successful deployment on a drone platform has enabled the real-time diagnosis of insulators, further confirming the practical applications value of an MGFNet.

New methodological approach to characterize dryland´s ecohydrological functionality on the basis of Balance between Connectivity and potential Water Retention Capacity (BalanCR)

Abstract Drylands are ecohydrologically-coupled ecosystems whose functioning depends on the interplay between hydrological connectivity between runoff source areas and the capacity of vegetation to retain water fluxes and associated resources. In this study we present a new easily applicable methodology for the ecohydrological characterization of dryland ecosystem functioning grounded in the balance between these two strongly interrelated processes using easily obtainable remote sensing data (e.g. UAV and SENTINEL-2 images), the BalanCR method (Balance between Connectivity and potential Water Retention Capacity). This methodology was first tested on synthetic hillslopes representing different configurations of the patchy and heterogenic distribution of vegetation in drylands. The analysis of these synthetic vegetation spatial patterns involving different vegetation patch densities, sizes, and fractional coverage values showed that BalanCR properly characterizes the expected ecohydrological interactions between potential conditions of runoff connectivity and water retention by plants operating in drylands. In a second step, we applied the BalanCR method on four semiarid hillslopes along an altitudinal aridity gradient covered by Mediterranean alpha steppes at very detailed spatial resolution (0.2 m) and at medium resolution (10 m). The obtained results were validated based on soil moisture data and vegetation greening and clearly recognized the four study sites as functional ecosystems, with very low water resource losses, and a pattern of increasing water redistribution processes as vegetation coverage declines. However, the sensitivity of methodology depends on the resolution of the input data (vegetation map and Digital Elevation Model; DEM), and the expected positive effect of small vegetation structures (vegetation patches smaller than the pixel size) on water redistribution is underestimated. Even in this case, the functionality and connectivity of the analyzed sites is correctly characterized as ecosystems showed similar values of both components for the methodology BalanC (hydrological connectivity component) and BalanR (potential water retention capacity component) than those obtained at very detailed scale, with a similar pattern of water allocation values in response to increased aridity. Thus, the proposed metric represents a promising tool for the proper evaluation of dryland conditions and to incorporate hillslope processes in climate change models, which is one of the main gaps to better understand the drylands response upon ongoing climate change.

Assessing the utilization of UAV technology for cadastre update applications in Turkey based on cost-benefit analysis

The aim of this study is to examine the use of UAVs from various perspectives in cadastral update studies and present the advantages and disadvantages of this method. The cadastral data obtained through direct fieldwork, the stereo model created by UAV images, and the orthophoto developed from UAV images were compared within the scope of the cadastre update study. Based on the data obtained from the UAV, it was observed that the detail points were identified more accurately and in a shorter time, saving 66.0% of the time and 75.7% of the cost.

A Scale Conversion Model Based on Deep Learning of UAV Images

As a critical component of many remote sensing satellites and model validation, pixel-scale surface quantitative parameters are often affected by scale effects in the acquisition process, resulting in deviations in the accuracy of image scale parameters. Consequently, various successive scale conversion methods have been proposed to correct the errors caused by scale effects. In this study, we propose ResTransformer, a deep learning model for scale conversion of surface reflectance using UAV images, which fully extracts and fuses the features of UAV images in the sample area and sample points and establishes a high-dimensional nonlinear spatial correlation between sample points and sample area in the target sample area, so that the scale conversion of surface reflectance at the pixel-scale can be completed quickly and accurately. We collected and created a dataset of 500k samples to verify the accuracy and robustness of the model with other traditional scale conversion methods. The results show that the ResTransformer deep learning model works best, providing average MRE, average MRSE, and correlation coefficient R values of 0.6440%, 0.7460, and 0.99911, respectively, and the baseline improvements compared with the Simple Average method are 92.48%, 92.45%, and 16.59%, respectively. The ResTransformer model also shows the highest robustness and universality and can adapt to surface pixel-scale conversion scenarios with different sizes, heterogeneous sample areas, and arbitrary sampling methods. This method provides a promising, highly accurate, and robust method for converting pixel-scale surface reflectance scale.