Archaeological sites are increasingly threatened by human activities and natural processes, causing irreversible heritage loss. Unmanned aerial vehicle (UAV) photogrammetry has advanced documentation and monitoring, yet many applications lack accuracy assessment or rely on a single data source. This study applies high-resolution, multi-temporal UAV-photogrammetry to nine archaeological sites in the Barranca Valleys, Peru-the first case study of this scale in the region. Surveys were scheduled to capture geomorphic responses to the forecasted 2023-2024 mega El Ni & ntilde;o, while also documenting pressures such as urbanization, agricultural expansion, and looting. By combining orthomosaics, digital elevation models, image cross-correlation, and horizontal displacement analysis-supported by explicit accuracy evaluation-this research provides a robust framework for assessing site degradation and highlights the added value of multi-source approaches for long-term monitoring.

Unmanned aerial vehicle (UAV) photogrammetry is increasingly used in applications requiring high accuracy, such as determining ground surface changes caused by landslides, mining, or microrelief transformation. While acquisition strategies have been widely studied, the influence of the processing workflow-particularly Bundle Block Adjustment parameter settings-remains insufficiently explored. This study addresses this gap through a systematic, full-factorial evaluation of 768 processing variants applied to ten UAV datasets collected over 1.5 years in a 220 ha study area. Eight key parameters were analysed. The results show substantial variability in final 3D accuracy: the best-performing variant achieved a root mean square error (RMSE) of 16 mm, whereas the weakest reached 303 mm. The most influential factors were the number of ground control points, the application of additional camera calibration corrections, and the use of the Post-Processing Kinematic GNSS method for determining camera projection center coordinates. The study also evaluates how workflow optimization affects the accuracy of displacement, tilt changes, and horizontal strain determination. While random displacement errors remained stable (RMSE of similar to 6-7 mm), systematic errors were significantly reduced by over half in all axes, with vertical median absolute error decreasing from 14 mm to 7 mm in the optimized configuration compared to the baseline previously used by the authors. This study provides the first large-scale, practice-oriented assessment of how processing parameter selection shapes the accuracy of both photogrammetric products and deformation indices determination. The results offer actionable guidance for developing more robust and repeatable UAV photogrammetry workflows tailored to high-precision monitoring.

In recent years, uncrewed aerial vehicle (UAV)-based photogrammetry has developed rapidly and is increasingly used for monitoring and determining displacements. The presentation discusses the author's solutions for the automatic determination of horizontal and vertical displacements of land surface in urban areas, dedicated to very-high-resolution UAV-photogrammetry products. The processing path is based on orthomosaics and digital elevation models and implements normalized cross-correlation for matching multi-temporal images. Its integral part is the process of semi-automatic removal of outliers. As a result of data processing, displacement vectors are determined in a regular grid, which constitute the basis for determining other indices of terrain deformation, such as ground tilts and horizontal deformations. Based on a comparison with reference data, it was estimated that the root mean square error of determining the displacements is 1-2 pixels for the horizontal components and 2-3 pixels for the vertical component. Therefore, the components of ground tilt and horizontal deformation can be determined based on UAV photogrammetry with a root mean square error of 0.3 pixels.

This article introduces an algorithm that uses a U-Net architecture to determine vertical ground surface displacements from unmanned aerial vehicle (UAV)-photogrammetry point clouds, offering an alternative to traditional ground filtering methods. Unlike conventional ground filters that rely on point cloud classification, the proposed approach employs heteroscedastic regression. The U-Net model predicts the conditional expected values of the elevation corrections, aiming to reduce the impact of vegetation on determined ground surface elevations. Concurrently, it estimates the logarithm of the elevation correction variance, allowing for direct quantification of the uncertainty associated with each elevation correction value. The algorithm was evaluated using three metrics: the root mean square error (RMSE) of vertical displacements, the percentage of nodes with determined displacement values, and the percentage of outliers among those values. Performance was assessed using the technique for order of preference by similarity to ideal solution (TOPSIS) method and compared against several ground-filter-based algorithms across four datasets, each including at least two time intervals. In most cases, the U-Net-based approach demonstrated a slight performance advantage over traditional ground filtering techniques. For example, for the U-Net-based algorithm, for one of the test datasets, the RMSE of the determined subsidences was 6.1 cm, the percentage of nodes with determined subsidences was 80.5%, and the percentage of outliers was 0.2%. For the same case, the algorithm based on the next best model (SMRF) allowed an RMSE of 7.7 cm to be obtained; for 77.3% of nodes, the subsidences were determined; and the percentage of outliers was 0.3%.

Measurements for determining terrain surface deformations that are caused by underground mining require high measurement accuracies to be achieved. Uncrewed aerial vehicles (UAVs) have not been widely used for this purpose. The presentation presents the process of selecting optimal bundle block adjustment (BBA) parameters for UAV-acquired data. The analyses were carried out for 25 measurement series on a test field of 2 km2. A total of 59 ground control points (GCP) and check points (CP) were used in the study. The analyzed parameters included: the GCP accuracy: 15mm or 25mm, the GCP number: 9 or 23, the tie point accuracy: 1px or 2px, the impact of the tie point filtration, the impact of the additional corrections of camera calibration (based on the 96-parameter Fourier series) not included in Brown’s model, the accuracy of the coordinates of the projection centers of the images: the values estimated by the GNSS receiver, or 50mm or 100mm. The final result of the study is the identification of BBA parameters that allow the highest accuracy of UAV photogrammetry products to be achieved.

The hazard in mining areas due to deformation is typically assessed through ground surface deformation indices. While unmanned aerial vehicle (UAV)-photogrammetry is effective at determining ground surface displacements, tilt changes, and horizontal strains-critical for infrastructure and building safety-have rarely been determined using such data. This study introduces a comprehensive methodology for deriving ground surface deformation indices from high-resolution UAV-photogrammetry, covering data acquisition, photogrammetric product development, and index determination. Two methods for displacement determination are proposed and compared: one using orthomosaics and digital elevation models with normalized cross-correlation for multitemporal data matching and the other based on point cloud registration. Both methods include outlier removal, with a key step involving the random sample consensus method combined with polynomial surface fitting. The study assesses both methods' accuracy, robustness, and effectiveness in determining ground surface displacements and their derivatives-tilt changes and horizontal strains. UAV-photogrammetry data were acquired with a ground sampling distance of 10 mm in two study areas. Using true or reference values, the root mean square error (RMSE) of displacement determination was estimated at 10 mm for horizontal components and 20 mm for subsidence, regardless of the data type processed. Both methods achieved an RMSE of 0.3 mm/m for tilt change components, but the raster-based method showed higher accuracy for horizontal strain components, with an RMSE of 0.2 mm/m (over a 30-meter baseline). These results suggest that deformation indices determined using UAV-photogrammetry products can be applied to monitor the risk of damaging structures in urbanized mining areas.

Detection of spatial and temporal deformations of landslides, along with the acquisition of precursor information, is crucial for hazard prediction and landslide risk management. Contemporary landslide monitoring systems based on remote sensing techniques (RST) play an important role in risk management and provide important support for Early Warning Systems. Research into the feasibility of using RST for monitoring different types of landslides also includes an analysis of the impact of radar wavelength on the obtained displacement results. The paper compares the time series results of landslide displacements obtained from satellite interferometric imaging in the C-band and L-band. The focus has been particularly on analyzing how the radar wavelength can impact the accuracy of the obtained displacement values and the ability to penetrate dense vegetation, especially under conditions of varying vegetation density. This poses a significant challenge for correct displacement detection. The obtained results are particularly relevant for geographical areas, such as Poland, where a large number of landslides occur in regions covered by dense vegetation. These are the conditions under which scientists encounter the greatest challenges in accurately monitoring these areas using radar systems. The final findings of the research are an important contribution to the development of landslide risk management strategies, crucial for the safety of people and infrastructure.

The main aim of the paper was to determine the rate of morphogenetic processes on a road that had been abandoned and the directions of road development. The research was conducted in the Lejowa Valley in the Tatra Mountains in southern Poland. To determine the rate of changes, we used data from Terrestrial Laser Scanning (TLS) (2019-2020) and unmanned aerial vehicle (UAV)-derived photogrammetric data (2022-2023). The research showed that, between the two study periods (2019-2020 and 2022-2023) the difference in erosion rate was not statistically significant within the roadbed but was significant for the roadbanks. The net change in roadbed erosion rate was from -432 m3 ha- 1 yr- 1 (2019-2020) to -308 m3 ha- 1 yr- 1 (2022-2023). The net change for roadbanks in each period was erosion and equalled -247 m3 ha- 1 yr- 1 (2019-2020) and -146 m3 ha- 1 yr- 1 (2022-2023). Changes across the alluvial fan at the end of the road in 2022-2023 were an accumulation of +260 m3 ha- 1 yr- 1. The development of microrelief within the road strictly depends on the course of deep and lateral erosions, which is modified by the road intercepting of additional runoff from the denudation valley and springs, by flow being redirected by debris such as uprooted trees, or by exposed bedrock.

This report summarises the results of field research conducted in October 2022 as part of the ArTu-DTu – The Archaeological Study of Da’jāniya & Tūwāneh, South Jordan project. The project focused on the sites of Tūwāneh and Da’jāniya, where a research team from the Jagiellonian University and AGH University of Kraków conducted excavations, geophysical surveys and analysis of looting damage. In Tūwāneh, excavations were carried out in the bath complex, documenting the architectural remains and stratigraphy associated with the site’s use and abandonment. Additionally, geophysical prospection was performed in selected areas adjacent to the so-called “caravanserai”, uncovering subsurface structures potentially linked to the complex. The team also systematically recorded Arabic graffiti on the walls of the caravanserai, preserving contemporary marks of local cultural heritage. In Da’jāniya, geophysical surveys aimed to locate remnants of a vicus surrounding the Roman fort, potentially shedding light on civilian settlement patterns near military installations. The report provides insights into the stratigraphy, architectural elements and temporal analysis of looting activity at the site, contributing valuable data on settlement dynamics and heritage conservation challenges in the region.

Looting is a worldwide issue that occurs not only in conflict zones or areas with weak governmental control. Although national and international agencies are addressing the problem, we are far from solving it, due to its complexity and the insufficient allocation of resources. In this article, we examine the temporal and spatial patterns of looting at the single site level (Tuwaneh, southern Jordan) over the past decade. Our analysis utilized orthomosaics created in 2018 and 2019, a systematic surface survey conducted in November 2022, and publicly available satellite imagery (via Google Earth Pro) dating back to August 2013. We identified a total of 723 looting pits, of which 259 were excavated before August 2013 and 140 between August 2013 and November 2022; 324 were inconclusive due to methodological limitations. The findings suggest that looting is a persistent issue in the area, highlighting the importance of implementing effective measures to prevent the loss of archaeological heritage.

This study compared classifiers that differentiate between urbanized and non-urbanized areas based on unmanned aerial vehicle (UAV)-acquired RGB imagery. The tested solutions included numerous vegetation indices (VIs) thresholding and neural networks (NNs). The analysis was conducted for two study areas for which surveys were carried out using different UAVs and cameras. The ground sampling distances for the study areas were 10 mm and 15 mm, respectively. Reference classification was performed manually, obtaining approximately 24 million classified pixels for the first area and approximately 3.8 million for the second. This research study included an analysis of the impact of the season on the threshold values for the tested VIs and the impact of image patch size provided as inputs for the NNs on classification accuracy. The results of the conducted research study indicate a higher classification accuracy using NNs (about 96%) compared with the best of the tested VIs, i.e., Excess Blue (about 87%). Due to the highly imbalanced nature of the used datasets (non-urbanized areas constitute approximately 87% of the total datasets), the Matthews correlation coefficient was also used to assess the correctness of the classification. The analysis based on statistical measures was supplemented with a qualitative assessment of the classification results, which allowed the identification of the most important sources of differences in classification between VIs thresholding and NNs.

Landslides are a widely recognized phenomenon, causing huge economic and human losses worldwide. The detection of spatial and temporal landslide deformation, together with the acquisition of precursor information, is crucial for hazard prediction and landslide risk management. Advanced landslide monitoring systems based on remote sensing techniques (RSTs) play a crucial role in risk management and provide important support for early warning systems (EWSs) at local and regional scales. The purpose of this article is to present a review of the current state of knowledge in the development of RSTs used for identifying landslide precursors, as well as detecting, monitoring, and predicting landslides. Almost 200 articles from 2010 to 2024 were analyzed, in which the authors utilized RSTs to detect potential precursors for early warning of hazards. The applications, challenges, and trends of RSTs, largely dependent on the type of landslide, deformation pattern, hazards posed by the landslide, and the size of the area of interest, were also discussed. Although the article indicates some limitations of the RSTs used so far, integrating different techniques and technological developments offers the opportunity to create reliable EWSs and improve existing ones.

To date, research on the use of unmanned aerial vehicle (UAV) photogrammetry to monitor deformations caused by underground mining has focused on the methods and accuracy of determining displacements. This study presents the results of research on the new application of UAV photogrammetry to determine the value of the coefficient of proportionality (B) between horizontal displacements and the change in the tilt of the terrain surface. The developed methodology for processing photogrammetric products is fully automated and robust. It includes automatic determination of displacement and the values of the B coefficient, as well as outlier removal. The processing results were compared to reference data derived from measurements on observation lines. The analyzed UAV photogrammetric products allowed the displacements to be determined with an accuracy of 2-3 cm for the horizontal component and 3-5 cm for the subsidence. As a result of the analyses, it was also found that the average value of the B coefficient determined using the proposed methodology is consistent with the average reference value. The values of the B coefficient based on UAV photogrammetry are also characterized by a similar coefficient of variation as the reference values. However, the accuracies obtained from the proposed method-ology are not sufficient for studying the spatial distribution of the B coefficient, mainly due to the influence of the relatively low accuracy of the tilt determined from UAV photogrammetry. The improvement of the accuracy of the tilt and B coefficient determined from UAV photogrammetry will be the subject of further research.

The main aim of this paper was to determine the rate of erosion and aggradation within a forest road on different geological structures as well as a comparison of changes between roadbeds and road cutslopes. The research was conducted in the Lejowa Valley in the Tatra Mountains. The measurements of the forest road were performed using a Leica ScanStation C10 terrestrial laser scanner (TLS) with a set of six-inch Leica High Definition Survey (HDS) targets. Based on the multi-temporal digital elevation models (DEMs) two DEM of Differences (DoD) were created for the summer season of 2019 and the 2019-2020 period. The research has shown that there is a significant difference between the rate of relief change for roads constructed atop conglomerates and limestone. The net change for the road within conglomerates was an erosion of 432.37 m(3)ha(-1) yr(-1), while for the road within limestone it was a sediment deposition which amounted to 3.87 m(3)ha(-1) yr(-1). Research has shown that there exists a difference in the rate of erosion for road cutslopes and roadbeds constructed within conglomerates. It was shown that the erosion rate for the roadbeds was 1.7-times larger than that for the studied road cutslopes in the period 2019-2020. Improperly designed and constructed forest roads may be subject to intensive relief transformations, even after they are no longer in use.

The problem of brightness differences between images of the same scene is important to the field of unmanned aerial vehicle (UAV) photogrammetry and affects both the aesthetics and the interpretation of the final product. This problem can be caused by changes in the positions of the camera or sun, as well as weather conditions. This article deals with the problem of varied image brightness caused by the latter Relative radiometric normalisation (RRN) of acquired RGB imagery is used to diminish the effects of this phenomenon and improve the visual quality of the resultant product. The presented algorithm considers the specificity of UAV-acquired data. It utilises image positions and their relationships to group similar images, choose references and perform RRN via histogram matching. The final method is robust and fully automatic. Validation performed on two independent datasets confirms its effectiveness both qualitatively (improving the appearance of the orthomosaic) and quantitatively.

This study presents an approach to the problem of minimizing the impact of low vegetation on the accuracy of a UAV-derived DEM, based on the use of a deep neural network (DNN). It is proposed to use the U-Net network to determine corrections to the height of the raw point cloud so that the processed data reflect the actual earth’s surface. The implemented solution is therefore based on regression, not classification. As a result of the proposed processing method, the expected value of the land surface height is determined for each point of the unified point cloud. In addition, a second U-Net network is trained, enabling the uncertainty of the corrected heights of the land surface to be determined for each point of the unified cloud. The training set includes data from different seasons, which makes the models more resistant and allows for assessment of the impact of the season and more generally the related vegetation status on the model accuracy. The processing results can be used in DEM generation, and also for determining the vertical displacements of the terrain surface associated with underground mining, as well as natural phenomena such as landslides. A key advantage of the proposed processing method is the ability to predict the uncertainty of the results.

This paper presents the results of a study to determine the accuracy of changes in building tilts based on photogrammetry from unmanned aerial vehicles. The changes in building tilts are calculated from the relative horizontal displacements (HzDiff method) and from the changes in heights of selected points (VDiff). For both methods, the accuracy is analyzed using reference data collected by terrestrial laser scanning. The root mean square error of the estimated total changes in building tilts is 2.7 mm/m for the HzDiff method and 3.3 mm/m for the VDiff method. Both methods provide a rather general view of the scale of tilt changes caused by mining, but do not allow for a detailed analysis of the tilt changes of individual objects.

Underground mining operations result in displacements and deformations of the land surface, which may pose a threat to building structures. The scale of horizontal displacements usually does not exceed several decimeters. These displacements should be monitored to assess and minimize their harmful effects. One of the methods of their observation is aerial photogrammetry. There are solutions that allow for the automation of horizontal displacement determination based on photogrammetric products. In most cases, however, they were created and used to process spaceborne or aerial imagery (with resolutions from about a few decimeters to several meters) and to analyze displacements on a much larger scale. This article proposes a workflow for automatic determination of the field of horizontal displacements caused by underground mining with the use of ultra-high resolution orthomosaics. The study included a comparison of the effectiveness of image registration algorithms for matching of multi-temporal orthomosaics. The outlier removal process is an integral part of the proposed workflow. The results showed that the weighted normalized cross correlation algorithm has the greatest potential for determining displacements based on UAV-derived orthomosaics, while the feature detection and matching algorithms turned out to be less effective in this task. After applying the proposed outlier removal path, the obtained accuracy of determining the displacements is at the level of 1–2 pixels, which was tested for two independent study areas. Accuracy was assessed both in comparison to displacements determined manually on the basis of UAV-derived orthomosaics, and in comparison to displacements independently determined using terrestrial laser scanning.

Development of the new methods of surface water observation is crucial in the perspective of increasingly frequent extreme hydrological events related to global warming and increasing demand for water. Orthophotos and digital surface models (DSMs) obtained using UAV photogrammetry can be used to determine the Water Surface Elevation (WSE) of a river. However, this task is difficult due to disturbances of the water surface on DSMs caused by limitations of photogrammetric algorithms. In this study, machine learning was used to extract a WSE value from disturbed photogrammetric data. A brand new dataset has been prepared specifically for this purpose by hydrology and photogrammetry experts. The new method is an important step toward automating water surface level measurements with high spatial and temporal resolution. Such data can be used to validate and calibrate of hydrological, hydraulic and hydrodynamic models making hydrological forecasts more accurate, in particular predicting extreme and dangerous events such as floods or droughts. For our knowledge this is the first approach in which dataset was created for this purpose and deep learning models were used for this task. Additionally, neuroevolution algorithm was set to explore different architectures to find local optimal models and non-gradient search was performed to fine-tune the model parameters. The achieved results have better accuracy compared to manual methods of determining WSE from photogrammetric DSMs.

Unmanned Aerial Vehicles (UAVs) can be an excellent tool for environmental measurements due to their ability to reach inaccessible places and fast data acquisition over large areas. In particular drones may have a potential application in hydrology, as they can be used to create photogrammetric digital elevation models (DEM) of the terrain allowing to obtain high resolution spatial distribution of water level in the river to be fed into hydrological models. Nevertheless, photogrammetric algorithms generate distortions on the DEM at the water bodies. This is due to light penetration below the water surface and the lack of static characteristic points on water surface that can be distinguished by the photogrammetric algorithm. The correction of these disturbances could be achieved by applying deep learning methods. For this purpose, it is necessary to build a training dataset containing DEMs before and after water surfaces denoising. A method has been developed to prepare such a dataset. It is divided into several stages. In the first step a photogrammetric surveys and geodetic water level measurements are performed. The second one includes generation of DEMs and orthomosaics using photogrammetric software. Finally in the last one the interpolation of the measured water levels is done to obtain a plane of the water surface and apply it to the DEMs to correct the distortion. The resulting dataset was used to train deep learning model based on convolutional neural networks. The proposed method has been validated on observation data representing part of Kocinka river catchment located in the central Poland.This research has been partly supported by the Ministry of Science and Higher Education Project “Initiative for Excellence – Research University” and Ministry of Science and Higher Education subsidy, project no. 16.16.220.842-B02 / 16.16.150.545.