Orthophoto Map Research Articles

Roof’s Potential and Suitability for PV Systems Based on LiDAR: A Case Study of Komárno, Slovakia

The case study focuses on evaluating the suitability of roof surfaces in terms of their solar potential based on their geometric parameters. The selected processing methodology detects segments of roof surfaces from the LiDAR base, supplemented with spatial information (orthophoto map, real estate cadastre (REC)—footprint, basic database for the geographic information system (ZBGIS)—classification of buildings—current use). The approach based on spatial analyses takes into account the limit conditions for determining the impact of solar radiation resulting from the roof area, slope, aspect, and hillshade. Considering to the available subsidy scheme for family houses in the conditions of the Slovak Republic, a narrower sample of 35 family houses was selected from the total number of typologically represented buildings (194). A 3D model of the building created by combining REC and LiDAR substrates shows the roof surface without overlap, while another 3D model made of LiDAR substrates alone represents the actual dimension of the roof surface. The results presented for each selected building show good agreement with each other, and their visualizations were obtained using two GIS environment approaches. In the area of family houses, up to 94% of the roof areas of buildings registered in the REC meet the conditions for the installation of a PV system with an output of 2.6/3.3 kW.

Automatic 3D building reconstruction from multi-view aerial images with deep learning

The study presented in this paper introduced a new fully automatic three-dimensional building reconstruction method that can generate first level of detail (LoD 1) building models from multi-view aerial images without any assistance from other data. The accuracy and completeness of our reconstructed models have approached that of manually delineated models to a large extent. The presented method consists of three parts: (1) efficient dense matching and Earth surface reconstruction, (2) reliable building footprint extraction and polygon regularization, and (3) highly accurate height inference of building roofs and bases. First, our novel deep learning-based multi-view matching method, composed of a convolutional neural network, gated recurrent convolutions, and a multi-scale pyramid matching structure, is used to reconstruct the digital surface model (DSM) and digital orthophoto map (DOM) efficiently without generating epipolarly rectified images. Second, our three-stage 2D building extraction method is introduced to deliver reliable and accurate building contours. Deep-learning based segmentation, assisted with DSM, is used to segment buildings from backgrounds; and the generated building maps are fused with a terrain classification algorithm to reach better segmentation results. A polygon regularization algorithm and a level set algorithm are thereafter employed to transfer the binary segmentation maps to structured vector-form building polygons. Third, a novel method is introduced to infer the height of building roofs and bases using adaptive local terrain filtering and neighborhood buffer analysis. We tested our method on a large experimental area that covered 2284 aerial images and 782 various types of buildings. Our results as far as correctness and completeness exceeded the results of other similar methods in a between-method comparison by at least 15% for individual 3D building models with many of them comparable to manual delineation results.

POST-EARTHQUAKE 3D BUILDING MODEL (LOD2) GENERATION FROM UAS IMAGERY: THE CASE OF VRISA TRADITIONAL SETTLEMENT, LESVOS, GREECE

Abstract. In recent years 3D building modelling techniques are commonly used in various domains such as navigation, urban planning and disaster management, mostly confined to visualization purposes. The 3D building models are produced at various Levels of Detail (LOD) in the CityGML standard, that not only visualize complex urban environment but also allows queries and analysis. The aim of this paper is to present the methodology and the results of the comparison among two scenarios of LOD2 building models, which have been generated by the derivate UAS data acquired from two flight campaigns in different altitudes. The study was applied in Vrisa traditional settlement, Lesvos island, Greece, which was affected by a devastating earthquake of Mw = 6.3 on 12th June 2017. Specifically, the two scenarios were created by the results that were derived from two different flight campaigns which were: i) on 12th January 2020 with a flying altitude of 100 m and ii) on 4th February 2020 with a flying altitude of 40 m, both with a nadir camera position. The LOD2 buildings were generated in a part of Vrisa settlement consisted of 80 buildings using the footprints of the buildings, Digital Surface Models (DSMs), a Digital Elevation Model (DEM) and orthophoto maps of the area. Afterwards, a comparison was implemented between the LOD2 buildings of the two different scenarios, with their volumes and their heights. Subsequently, the heights of the LOD2 buildings were compared with the heights of the respective terrestrial laser scanner (TLS) models. Additionally, the roofs of the LOD2 buildings were evaluated through visual inspections. The results showed that the 65 of 80 LOD2 buildings were generated accurately in terms of their heights and roof types for the first scenario and 64 for the second respectively. Finally, the comparison of the results proved that the generation of post-earthquake LOD2 buildings can be achieved with the appropriate UAS data acquired at a flying altitude of 100 m and they are not affected significantly by a lower one altitude.

Physical and Mechanical Characteristics of Lunar Soil at the Chang'E‐4 Landing Site

AbstractChang'E‐4, with the Yutu‐2 rover, is the first lunar probe to successfully land and conduct a tour on the far side of the Moon from early 2019. We analyze the physical and mechanical characteristics of lunar soil through the in situ terrain data collected by the panoramic camera onboard the Yutu‐2 rover. With the slip ratio and wheel sinkage obtained by the derived Digital Orthophoto Map (DOM) and Digital Elevation Model (DEM), the mechanical parameters of lunar soil are derived from the slip‐sinkage model. These mechanical parameters and wheel size of the rover are used to obtain the pressure‐sinkage curves, which can estimate the lunar soil strength. The experimental results indicate that the soil strength at the Chang'E‐4 landing site is much higher than that at the Chang'E‐3 landing site. The discrepancies in lunar soil strength between the two landing sites may be related to the local surface topography and degree of space weathering.

Ecological niche models improve home range estimations

AbstractHome ranges in animals can be estimated by different methods like minimum convex polygons, characteristic hulls or kernels while correlative ecological niche models (ENMs) are commonly employed for forecasting species' ranges. However, ENMs can also model the distribution of individuals if environmental very high spatial resolution data are available. Indeed, remote sensing (RS) can provide images with pixel sizes of few centimetres. Here, we modelled the distribution of individual lizards (Podarcis bocagei) combining aerial‐like photographs recorded with a compact camera and a matrix of temperature/humidity data‐loggers to obtain several environmental layers with very high spatial resolution. We recorded lizards’ positions in a 20 × 20 m mesocosm with a high precision GPS device (~10 cm of error), multiple times per day throughout the whole period of daily activity. We built an orthophoto map (pixels of 20 cm2) from camera pictures, a digital surface model, and a land‐cover supervised classification map. We recreated climate‐like variables by combining data‐logger data. For each individual, we calculated the distance to males and females, excluding the focal lizard. We computed individual realized niche models with Bioclim, GAM, GLM, Maxent and random forest. Models attained a very high evaluation score in most cases. The most contributing variables were related to microclimate (isothermality, minimum temperature and humidity) and distance to conspecifics. Our very high spatial resolution models provided robust information on how space is used by each lizard. Correlative models can identify the most suitable areas inside the home range, similar to core areas estimated from kernel algorithms, but allowed better statistical inference. Overall, RS tools generated high‐quality environmental data, and when combined with ENMs, improved the robustness of the predictions on spatial patterns of small terrestrial animals.

Multi-scale morphologic investigation of craters in the Chang'e-4 landing area

China's Chang'e-4 (CE-4) rover landed in Von Kármán crater on January 3, 2019, and has acquired a large amount of data of the landing site. Craters are an important landform in lunar surface and the analysis of craters at the landing site with high-resolution images is of great significance for both scientific study and engineering safety. In this study, multi-sources data, including Lunar Reconnaissance Orbiter Camera Narrow Angle Camera (NAC) images, NAC digital elevation model (DEM), Chang'e-2 (CE-2) digital orthophoto map (DOM), SLDEM2015 and stereo images from panoramic cameras onboard the CE-4 rover, are involved to reveal the crater morphologic features at the landing site. The diameter, depth, spatial density and depth/diameter ratio (d/D) of the mapped craters have been investigated. Randomness analysis and the size-frequency distributions of craters in two scales are conducted to thoroughly elaborate the crater distribution in the landing area. The multi-scale crater investigation establishes a foundation for subsequent in-depth studies, e.g., crater evolution and degradation of lunar farside and CE-4 landing area dating.

A big landslide on the Jinsha River, Tibet, China: geometric characteristics, causes, and future stability

On October 10, 2018, a big landslide occurred on the right bank of the Jinsha River in Baige Village, Polo Township, Jiangda County, Tibet (hereafter called the Baige landslide), which blocked the Jinsha River, forming a barrier lake. Afterward, the landslide dam broke, producing a flood. On November 3, the rear wall of this landslide failed again, also blocking the Jinsha River and creating a bigger barrier lake. Then, by local people, a discharge channel was excavated on the top of the landslide dam, making lake water cross over the discharge channel. As the water flow gradually increased, the landslide dam broke again, producing a more severe flood, resulting in huge economic losses downstream. The purpose of this study is to understand the cause of this landslide and predict the future stability of its head scarp, providing some support for the control scheme in the later stage. A digital orthophoto map (DOM) and a digital elevation model (DEM) of the landslide were created using an unmanned aerial vehicle. Then based on the DOM and DEM, the geometric characteristics of the landslide were described. Multi-phase Planet 5 images were used to infer the development process of the landslide. Finally, the cause of the landslide was analyzed based on the rainfall data and the limit equilibrium calculation. The results show that the Baige landslide was a self-weight creeping event, and its development and trigger were independent of the rainfall. Before the landslide, the slope experienced five stages of evolution: steady deformation, slow deformation, rapid deformation, steady deformation, and rapid deformation. The limit equilibrium calculation indicates that the stability coefficient of the middle section of the head scarp is the lowest, thus which should be cut down as a priority. This study provides a typical example of a self-weight creep type landslide, and an important reference for prediction and prevention of similar large landslides in the Tibetan Plateau, southwestern China.

UAS DATA ACQUISITION PROTOCOL FOR MARINE HABITAT MAPPING: AN ACCURACY ASSESSMENT STUDY

Abstract. Marine habitat mapping is essential for updating existing information, preserving, and protecting the marine environment. Unmanned Aerial Systems (UAS) are an important tool for monitoring and mapping coastal and marine environment because of their ability to provide very high-resolution aerial imagery.Environmental conditions have a critical role in marine mapping using UAS. This is due to the limitations of UAS surveys in coastal areas, i.e. the environmental conditions prevailing in the area. The limitations of weather and oceanographic conditions affecting the quality of marine data led to the creation of a UAS protocol for the acquisition of reliable marine information. The produced UAS Data Acquisition Protocol consists of three main categories: (i) Morphology of the study area, (ii) Environmental conditions, (iii) Flight parameters. These categories include the parameters that must be considered for marine habitat mapping.The aim of the present study is the accuracy assessment of the UAS protocol for marine habitat mapping through experimental flights. For the accuracy assessment of the UAS protocol, flights on different dates and environmental conditions were conducted, over a study area. The flight altitude was the same for all the missions, so the results were comparable. The high-resolution orthophoto maps derived from each date of the experiment were classified. The classification maps show several differences in the shape and size of the marine habitats which are directly dependent on the conditions that the habitats were mapped. A change detection comparison was conducted in pairs to examine the exact changes between the classified maps.The results emphasize the importance of the environmental conditions prevailing in an area during the mapping of marine habitats. The present study proves that the optimal flight conditions that are proposed of the UAS Data Acquisition protocol, respond to the real-world conditions and are important to be considered for an accurate and reliable mapping of the marine environment.

TOPOGRAPHIC MAPPING AND ANALYSIS BASED ON 3D RECONSTRUCTION MODEL OF SIMULATED ASTEROID

Abstract. Asteroid exploration is of great scientific significance in understanding the evolution of the solar system. Unlike lunar missions, asteroids are far away from the earth, the communication between the spacecraft and the ground station has a large delay. Moreover, the small size, irregular shape and uneven mass distribution of asteroids further increase the difficulty of landing missions. This paper presented a topographic mapping and analysis method based on three-dimensional (3D) reconstruction model of a simulated asteroid. First, 3D reconstruction of a simulated asteroid was achieved using Structure from Motion (SfM) technique. Then Digital Elevation Model (DEM) and Digital Orthophoto Map (DOM) were produced based on the 3D point cloud data. By analysing the DEM data, we finally obtained the obstacles distribution map. In order to verify the proposed method, we used a simulated asteroid to conduct experiments. And the experimental results demonstrated the effectiveness of the developed method. The topographic mapping and analysis method presented in this study can be applied to landing/sampling area selection in further asteroid exploration research.

AUTOMATIC BENGGANG RECOGNITION BASED ON LATENT SEMANTIC FUSION OF UHR DOM AND DSM FEATURES

Abstract. Benggang are characterized by deep-cut slopes with various shapes and depressions on the vast weathered crust slopes in southern China. The gully heads have been continuously collapsed and eroded to form a chair-like erosion landforms. It develops rapidly, and leads to large amounts of erosion, with the hazards of damaging land resources, destroying basic farmland, and deteriorating ecological environment. To study and manage Benggang, the primary task is to discover it. Traditional methods based on local in-situ investigations, which are not only labour-consuming but also inefficient. These methods are difficult to meet the needs of large-scale investigations of Benggang. This paper proposes a method for automatic Benggang recognition based on Ultra-High Resolution (UHR) DOM (Digital Orthophoto Map) and DSM (Digital Surface Model) obtained from UAV (Unmanned Aerial Vehicle) survey. This method adopts a Bag of Visual-Topographical Words (BoV-TW) model. The local features extracted from DOM and DSM are represented based on BoV-TW, and fused by Latent Dirichlet Allocation (LDA). Finally Support Vector Machine (SVM) is adopted as a supervised classifier to achieve high-precision automatic Benggang recognition. Experimental results prove that the total accuracy of our method can be maintained at about 95%, with recall and precision above 80% (the highest are 97.22% and 94.44%, respectively), which are significantly higher than the methods of using only DOM local features and using only BoV-TW.

The Use of Drones in Disaster Aerial Needs Reconnaissance and Damage Assessment – Three-Dimensional Modeling and Orthophoto Map Study

The aim of this research is to provide disaster managers with the results of testing three-dimensional modeling and orthophoto mapping, so as to add value to aerial assessments of flood-related needs and damages. The relevant testing of solutions concerning the real needs of disaster managers is an essential part of the pre-disaster phase. As such, providing evidence-based results of the solutions’ performance is critical with regard to purchasing them and their successful implementation for disaster management purposes. Since disaster response is mostly realized in complex and dynamic, rather than repetitive, environments, it requires pertinent testing methods. A quasi-experimental approach, applied in a form of a full-scale trial meets disaster manager’s requirements as well as addressing limitations resulting from the disaster environment’s characteristics. Three-dimensional modeling and orthophoto mapping have already proven their potential in many professional fields; however, they have not yet been broadly tested for disaster response purposes. Therefore, the objective here is to verify the technologies regarding their applicability in aerial reconnaissance in sudden-onset disasters. The hypothesis assumes that they will improve the efficiency (e.g., time) and effectiveness (e.g., accuracy of revealed data) of this process. The research verifies that the technologies have a potential to facilitate disaster managers with more precise damage assessment; however, their effectivity was less than expected in terms of needs reconnaissance. Secondly, the overall assessment process is heavily burdened by data processing time, however, the technologies allow a reduction of analytical work.

Fast Georeferenced Aerial Image Stitching With Absolute Rotation Averaging and Planar- Restricted Pose Graph

Accurate digital orthophoto map generation from high-resolution aerial images is important in various applications. Compared with the existing commercial software and the current state-of-the-art mosaicing systems, a novel fast georeferenced orthophoto mosaicing framework is proposed in this study. The framework can adapt to the challenging requirements of high-accuracy orthoimage generations with relatively fast speed, even if the overlap rate is low. We provide appearance and spatial correlation-constrained fast low-overlap neighbor candidate query and matching. On the basis of GPS information, we introduce an absolute position and rotation-averaging strategy for global pose initialization, which is essential for the high convergence and efficiency of nonconvex pose optimization of every image. We also propose a planar-restricted global pose graph optimization method. The optimization is extremely efficient and robust considering that point clouds are parameterized to planes. Finally, we apply a matching graph-based exposure compensation and region reduction algorithm for large-scale and high-resolution image fusion with high efficiency and novel precision. Experimental results demonstrate that our method can achieve the state-of-the-art performance while maintaining high precision and robustness.

A novel index to detect green-tide using UAV-based RGB imagery

Unmanned aerial vehicles (UAV) equipped with high-resolution camera have been increasingly applied in environment monitoring as an important complement to traditional satellite remote sensing. An accurate extraction of marine green-tide regions still faces many technological challenges, such as the absence of centimeter-level orthophoto maps and a dedicated green-tide index based on red-green-blue (RGB) bands. In this study, a new green-tide index, namely, the red-green-blue floating algae index (RGB-FAI) using RGB images captured by ship-borne UAV, is developed for green-tide detection in the Yellow Sea, China. Specifically, RGB-FAI is defined to measure the green-reflectance height by using the red and blue bands as the baselines. Our results show that the RGB-FAI performs well in the detection of green-tide and the accuracy is satisfactory (kappa = 0.95). It is worthy to note that RGB-FAI has the highest extraction accuracy among these competing indices for green-tide in the declining phase under a hazy atmospheric condition. In addition, by combining the bi-temporal UAV images with RGB-FAI, the drift velocity of green-tide has also been estimated as 0.26 m/s in a 17.1° east by north during aerial photography. In conclusion, the proposed RGB-FAI is effective for green-tide detection and has more potential usage in marine environment monitoring.

Magnitude of soil erosion in small catchments with different land use patterns under an extreme rainstorm event over the Northern Loess Plateau, China

Characteristics of soil erosion under extreme rainfall conditions are indicators of the ecological restoration measures as well as the rationality of land use patterns in the region. To this end, 12 dam-controlled catchments were selected for investigation after an extreme rainstorm event in the northern Shaanxi Province on 25–26 July 2017 (known as the “7.26” rainstorm). Soil erosion intensities in the 12 catchments were obtained by excavations of sedimentation profiles and sedimentation area measurements. Digital orthophoto maps and digital terrain models by unmanned aerial vehicle were generated to characterise land use types and areas, slope gradients, and the distance from the land to the boundary of dam-land along the flow accumulation path. Stepwise regression analysis was used to determine the main factors affecting catchment erosion intensity. Results showed that the average sedimentation thickness and overall soil erosion intensity across the 12 catchments ranged from 0.16 to 1.67 m and 10295–49227 t km−2, respectively. Soil erosion caused by the extreme rainstorm was 10–50 times of the soil loss tolerance on the Loess Plateau (1000 t km−2 year−1). Stepwise regression analysis revealed that the more similar the shape of a catchment to a circle, the closer the distance between the slope-cropland and dam-land, the larger the erosion modulus within the catchment. Furthermore, the presence of a cement road along the valley shoulder greatly reduced the modulus of soil erosion. These findings indicate that the existing ecological conditions in the dam-controlled catchments were not able to effectively resist erosion under the extreme rainfall conditions of this event. Thus, optimising the distribution of land use types within catchments should be a focus for soil erosion control.

Post-seismic monitoring of cliff mass wasting using an unmanned aerial vehicle and field data at Egremni, Lefkada Island, Greece

On 17th November 2015, a strong earthquake affected Lefkada Island, Greece, leading to extensive translational debris slides along the cliff overhanging Egremni beach and the re-activation of previously deposited slope material. The cliff is crossed by the Cephalonia Fault Zone, with mass-wasting processes inducing dry flows occurring in association with long-term tectonic activity. Large-scale co-seismic and post-seismic disturbance of the forest on top of the cliff has also resulted from the seismicity. Three years after the 2015 event, the cliff had retreated due to active translational debris slides and the forest disturbance, which represents a safety concern for the tourist beach below. We aimed to qualitatively assess the transient evolution of debris slides at Egremni and clarify the level of risk posed to the beach below. Annual monitoring was performed between 2015 and 2018 including extensive field measurements and unmanned aerial vehicle (UAV) flight campaigns. Based on the high-resolution images acquired by the UAV, we created orthophoto maps and digital surface models (DSMs) of the cliff using Structure-from-Motion (SfM) photogrammetry. The cliff slope models indicate that the overhanging cliff is progressively entering a stage of decline and approaching equilibrium due to climatic forcing alongside minor seismicity. The DSMs were combined with field survey data to further investigate the role of erosion and deposition across the cliff. Our results indicate that the cliff dynamics at Egremni are primarily controlled by earthquake- and storm-induced debris slides, with rock material moving within the cliff formation via rock falls, and cliff inclination decreasing as a result of primitive slope ‘self-healing’ process. Mass-wasting processes remained widespread across the western part of the Lefkada coast three years after the 2015 earthquake sequence. For safety reasons, we suggest that pre-emptive inspection of the cliffs should be carried annually, after humid and winter periods, and following strong earthquakes using a combination of field surveys and UAV campaigns.

Constructing the landscape security pattern of urban stormwater in Nanjing, China

Stormwater management must address three direct threats to urban ecological security during the urbanisation of China: waterlogging, water pollution and shortages of water resources. Based on the theory of landscape security patterns, macro-level stormwater landscape security patterns (SLSPs) associated with regional flood security, runoff pollution control security structures and water resource utilisation can be set up to solve urban stormwater and environmental problems. The objective of this study was to construct the SLSP of the Tiebei area in Nanjing, east China. A high-precision digital terrain model and a digital orthophoto map were obtained using portable, unmanned aerial vehicle remote sensing technology and Pix4Dmapper software. These data were integrated in an ArcGIS platform and used to calculate runoff flow direction and flow accumulation, measure runoff pollution levels and determine priority areas for water resources protection. Combined with regional planning of the Tiebei area, the SLSP was effectively and efficiently constructed. This study also examined current management and use of urban stormwater and addressed the important bases for ecological flood prevention, regional planning and site design of the urban landscape.

Drone Application for Generating a High Precision Orthophoto to Support Village Boundary and Land Use Mapping in Indonesia

Many existing village maps in Indonesia are available without spatial references, including in Pandanrejo village in Kaligesing Sub-district, Purworejo District, Central Java. To date, the available village map of the Pandanrejo is sketches map which potentially overlaps with the area of adjacent villages. In this study, we used a commercial drone DJI Phantom 4 for generating a high precision orthophoto to create the exact boundaries and land use information of the Pandanrejo village. The drone captured the aerial images from an altitude of approximately 100 m and automatically take an image every 3 seconds giving an overlap of each image around 70% of each pair of neighboring images. By this mission plan, the drone collected of about 1,312 aerial photographs. In order to obtain better geo-referencing results, eleven Ground Control Points (GCPs) distributed over the study area were obtained by GNSS geodetic tools. Furthermore, the Structure from Motion (SfM) algorithm was used to generate point clouds, digital surface models, and digital orthophoto model. These processes were carried out by using Agisoft PhotoScan software completing 10 steps in the workflow starting from adding drone images to the software to building orthomosaic as a basis for a digital orthophoto model. We then analyzed the errors of the resulted orthophoto map by comparing between the ratio of village boundary length to the total village area from drone mapping and terrestrial mapping using GNSS geodetic. The results showed that the drone mapping has successfully produced the village boundary map of the Pandanrejo village (i.e. a boundary length 7,473.2 m and an area 2.18 km2) that is comparable with the result of terrestrial area mapping corrected by GNSS geodetic. In addition, the drone has also successfully mapped the village land use with a total accuracy of 100% accurate. With these results, we suggest drone technology as an alternative method besides a traditional terrestrial method to create a definite village boundary and land use map to support the implementation of Indonesia Law No. 6 of 2014.

A catalogue of impact craters larger than 200 m and surface age analysis in the Chang'e-5 landing area

Chang'e-5 (CE-5) is the first sample-return mission of China's lunar exploration and will be launched in 2020. The planned landing area is near Mons Rümker in Oceanus Procellarum. It is important to have a detailed geological analysis of the landing area prior to the lunar sample collection. The crater size-frequency distribution curve, combined with radioisotope dating of components of the returned sample by CE-5, may allow a refinement of the lunar chronology function. Based on a digital orthophoto map with a pixel size of 1.5 m generated from more than 700 Lunar Reconnaissance Orbiter Camera Narrow Angle Camera images of the CE-5 landing area, all impact craters larger than 200 m in diameter were extracted. Using the resultant crater catalogue of 32,277 craters, a systematic analysis of the landing area including the crater distribution, morphology characteristics, and geologic model ages is conducted. The results show that craters smaller than 1,000 m are denser in the east and the west parts of the region, while those larger than 1,000 m tend to be more concentrated in the central part, particularly within the range of 61.5°W–64.5°W. Within nine geologic units divided using the spectral characteristics, the geologic unit of Em1 (Eratosthenian-aged mare, Em) near the Mons Rümker is the youngest with a model age of 2.02−0.16+0.16 Ga. The three Rümker plateau (IR) units are the oldest parts of the landing area with the largest model age of 3.49−0.12+0.073 Ga. The crater catalogue and the surface model age analysis results can contribute to the CE-5 mission as well as further scientific research.

High-resolution orthophoto map and digital surface models of the largest Argentine Islands (the Antarctic) from unmanned aerial vehicle photogrammetry

ABSTRACT This study presents the first high-resolution orthophoto maps and digital surface models (DSMs) of the largest Argentine Islands, West Antarctica. Aerial surveys with small unmanned aerial vehicle (UAV) were performed in Austral summer, 2018, taking 10,041 aerial photographs. Accuracy requirements were ensured using ground control points (GCPs). A resolution of 3.4 and 6.8 cm/px of orthomosaics and DSMs is reached on average, and the RMS reprojection error is 0.22 m on average. We report the morphometric parameters of surveyed islands and discuss issues related to accuracy and the usage of UAVs in polar conditions. This study demonstrates that small and low cost UAVs can be successfully used in harsh polar conditions to obtain accurate orthomosaics and DSMs of mainly glaciated terrain. We provide all generated materials in full resolution available in a scientific data repository that could be used for the monitoring of ice cap changes, vegetation cover, and wildlife populations.

Automatic Seamline Determination for Urban Image Mosaicking Based on Road Probability Map from the D-LinkNet Neural Network.

Image mosaicking which is a process of constructing multiple orthoimages into a single seamless composite orthoimage, is one of the key steps for the production of large-scale digital orthophoto maps (DOM). Seamline determination is one of the most difficult technologies in the automatic mosaicking of orthoimages. The seamlines that follow the centerlines of roads where no significant differences exist are beneficial to improve the quality of image mosaicking. Based on this idea, this paper proposes a novel method of seamline determination based on road probability map from the D-LinkNet neural network for urban image mosaicking. This method optimizes the seamlines at both the semantic and pixel level as follows. First, the road probability map is obtained with the D-LinkNet neural network and related post processing. Second, the preferred road areas (PRAs) are determined by binarizing the road probability map of the overlapping area in the left and right image. The PRAs are the priority areas in which the seamlines cross. Finally, the final seamlines are determined by Dijkstra’s shortest path algorithm implemented with binary min-heap at the pixel level. The experimental results of three group data sets show the advantages of the proposed method. Compared with two previous methods, the seamlines obtained by the proposed method pass through the less obvious objects and mainly follow the roads. In terms of the computational efficiency, the proposed method also has a high efficiency.