Applied Geomatics最新文献

To date, various image registration approaches have been conducted to deal with distortions between multimodal image pairs. However, significant existing noise as an unavoidable issue deteriorates many conventional and advanced methods. The critical key is to choose a highly robust local feature detection and description method as the principle for many matching frameworks. However, few studies have concentrated on dealing with the noise issue. For this purpose, this paper evaluates the most well-known and state-of-the-art feature descriptors against artificial sequential noise levels. The employed methods consist of various handcrafted learning-based descriptors. It is further indicated that in addition to the designed structural feature map, multiple criteria, such as spatial arrangement, and the magnitude of the support area, play roles in achieving successful matching, especially in the presence of dramatic noise and complex distortion between multimodal images. Moreover, to filter out the noisy features, the employed local feature detectors are integrated with the uniform competency algorithm. Experimental results demonstrate the overall superiority (20.0% on average) of the MKD (multiple-kernel descriptor) due to advanced designed integrated kernels and polar arrangements.

Nowadays, with the increase in the use of unmanned aerial vehicles (UAVs), small-area aerial photography has become a viable alternative to traditional data surveys, such as topography or satellite imagery analysis, mainly due to its high spatial and temporal resolution. Thus, the objective of this work was to evaluate and compare the survey of the built area of the UTFPR – Dois Vizinhos Campus, Brazil, conducted in the field using total station, with an orthomosaic obtained from a UAV using non-metric camera, with both methods using georeferenced control points in the ground. The analyses showed that there was a high correlation between the areas obtained by these methodologies, with an acceptable error for many purposes, as shown by the Pearson correlation coefficient of 0.9991 and the relative error of 2.23432%, proving to be an effective tool for such surveys. Thus, this work concluded that it is possible to survey the built area from a UAV orthomosaic using a non-metric camera, which required less equipment and allowed to obtain the data in a shorter time when compared to a classical topography survey on the field.

Surface soil type classification is essential to enhance food production in precision farming. However, soil classification is time-consuming, laborious, and costly through the traditional methods. Recently, artificial intelligence-based methods, especially machine learning, have played a vigorous role in soil classification and its mapping. However, machine learning still makes exterior soil type classification and its mapping difficult due to various features and spatio-temporal inconsistencies. Therefore, the present study has tried to determine soil properties and sort accordingly using hyperspectral datasets and machine learning methods. We used field spectra generated by ASD Field Spec 4 device and satellite image. The proposed approach has identified three prominent soil types, Regur soil, Lateritic soil, and sand dunes according to soil taxonomy, with more than 95% success rate using satellite hyperspectral image and machine learning models. Thus, the outcome of the present study can be effectively utilized in healthy agricultural practices to increase global food production. In addition, the suggested strategy can be used in precision agriculture and environmental management.

This study evaluates and compares the accuracy and reliability of multiple freely available digital elevation models (DEMs) including Copernicus Global Land Operations (GLO), Advanced Land Observing Satellite (ALOS), Cartosat, Shuttle Radar Topography Mission (SRTM), and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) for hydrological applications in the Mahi River upper basin in Western India. Through watershed delineation, statistical analysis, error quantification, and 2D hydraulic modeling using HEC-RAS, this research assesses the performance of these DEMs with GLO DEM as the reference. GLO DEM is used as the reference because key findings show it most accurately delineates watershed boundaries and stream networks and has the fewest sinks. ALOS also demonstrates strong performance, with 70.47% watershed boundary similarity to GLO. Cartosat shows reasonable accuracy in watershed delineation with a Jaccard Index (JI) of 68.41% while SRTM and ASTER appear less reliable. Statistical analysis reveals ALOS slightly overestimates while other DEMs underestimate elevations compared to GLO for most of the slope classes. Flood modeling shows GLO produces the smoothest inundation, with ALOS second-best. Overall, GLO and ALOS emerge as the most accurate and reliable options followed by Cartosat among freely available datasets for the study area. The research provides insights into DEM performance to inform selection and improve hydrological applications involving terrain data for the study area.

This study addresses the representation and exchange of geospatial geometric 3D models, which is a common requirement in various applications like outdoor mixed reality, urban planning, and disaster risk management. Over the years, multiple file formats have been developed to cater to diverse needs, offering a wide range of supported features and target areas of application. However, classic exchange formats like the JavaScript Object Notation and the Extensible Markup Language have been predominantly favored as a basis for exchanging geospatial information, leaving out common geometric information exchange formats such as Wavefront’s OBJ, Stanford’s PLY, and OFF. To bridge this gap, our research proposes three novel extensions for the mentioned geometric file formats, with a primary focus on minimizing storage requirements while effectively representing geospatial data and also allowing to store semantic meta-information. The extensions, named GeoOBJ, GeoOFF, and GeoPLY, offer significant reductions in storage needs, ranging from 14 to 823% less compared to standard file formats, while retaining support for an adequate number of semantic features. Through extensive evaluations, we demonstrate the suitability of these proposed extensions for geospatial information representation, showcasing their efficacy in delivering low storage overheads and seamless incorporation of critical semantic features. The findings underscore the potential of GeoOBJ, GeoOFF, and GeoPLY as viable solutions for efficient geospatial data representation, empowering various applications to operate optimally with minimal storage constraints.

In-season wheat growing area identification is of great importance for monitoring crop growth conditions and predicting related yield. In this study, we developed an approach to map wheat crops at a regional scale, using both the Synthetic Aperture Radar (SAR, Sentinel-1, S1) and Copernicus Optical (Sentinel-2, S2) satellite data, to estimate the extent of the wheat growing area. The approach relies on machine learning random forest classification algorithm performed in the Google Earth Engine (GEE) cloud platform. The methodology is based on three experiments, each consisting of the processing of a specific Sentinel time series imageries: a first experiment considering the S1 data solely, a second experiment with the S2 data solely and a third experiment with S1 + S2 data merged. The results showed that the third experiment combining SAR and optical data turned out with the best overall accuracy of 82.36% and a kappa coefficient of 0.77. These results indicate that the integration of Sentinel-1 and Sentinel-2 improved classification accuracy by 1.5 to 6% over the use of Sentinel-2 only. A comprehensive assessment based on survey samples revealed Producer and User accuracies of 84% and 81% respectively; and an F1-score of 0.82. The approach followed in the study provides a basis for mapping seasonal wheat areas that will support planning and policy decisions.

Container cranes are of key importance for maritime cargo transportation. The uninterrupted and all-day operation of these container cranes, which directly affects the efficiency of the port, necessitates the continuous inspection of these massive hoisting steel structures. Due to the large size of cranes, the current manual inspections performed by expert climbers are costly, risky, and time-consuming. This motivates further investigations on automated non-destructive approaches for the remote inspection of fatigue-prone parts of cranes. In this paper, we investigate the effectiveness of color space-based and deep learning-based approaches for separating the foreground crane parts from the whole image. Subsequently, three different ML-based algorithms (k-Nearest Neighbors, Random Forest, and Naive Bayes) are employed to detect the rust and repainting areas from detected foreground parts of the crane body. Qualitative and quantitative comparisons of the results of these approaches were conducted. While quantitative evaluation of pixel-based analysis reveals the superiority of the k-Nearest Neighbors algorithm in our experiments, the potential of Random Forest and Naive Bayes for region-based analysis of the defect is highlighted.

One of the natural threats that arises as a result of temporary surface runoff is flooding, which has a large amount of solid material, a high level of water in the streams, a sudden appearance, and a rapid flow velocity. The Wadi Amlog Basin is characterized by the lack of rain and the prevalence of drought, but it is exposed to sudden rains that lead to surface runoff in its dry valleys in a way that results in threats to infrastructure and spatial development in the coastal region. Within this framework, the purpose of this research is to investigate the possible areas of flood hazard by using GIS techniques based on morphometric assessment parameters to determine the risk level of specified subbasins from a digital elevation model (DEM) using remotely sensed SRTM images. The case study results utilized five evaluation degrees, very low, low, moderate, high, and very high, to interpret the flood danger, in a way that contributes to protecting the places most affected by the dangers of floods in the subbasins in the study area.

The aim of this study is to produce a landslide susceptibility map in Mogods and Hedil using the fuzzy logic method. To increase the objectivity of the approach, the fuzzy membership was calculated using the frequency ratio (FR). Nine factors were considered for landslide control, including slope, aspect, plan curvature, profil curvature, distance from faults, distance from rivers, land use, precipitation, and lithology. The frequency ratio was used to calculate the fuzziness of each factor, and these results were then applied to the fuzzy operators to produce the landslide susceptibility map. The selection of the susceptibility map closest to reality was based on the spatial distribution of landslides in each susceptibility class of each fuzzy operator and on the application of the receiver operating curve (ROC). The results of the area under curve (AUC) analysis show that the GAMMA operator (0.90) provided the most accurate prediction of the landslide susceptibility map, as indicated by the prediction accuracy of the model (0.766). The study area was classified into four classes using Jenks natural fracture classification method: low susceptibility zone, moderate susceptibility zone, high susceptibility zone, and very high susceptibility zone. The use of the fuzzy GAMMA operator for landslide susceptibility mapping gave a very satisfactory result with a reliability rate of 76.6%.

The rich vegetation areas with a variety of biodiversity are designated under categories of protected areas. Protected areas on Earth are the biomes where the elements of nature function together and maintain the life cycle. These protected areas include forest cover, rivers, waterbodies, mangroves, etc. which are the origin of ecology and biodiversity and provide natural resources utilized for human needs. Maintaining protected area is an essential aspect of managing the forest covers and a key strategy for combating the negative effects of biodiversity loss and fragmentation. The research aims to assess the vegetation health in the protected areas with NDVI using remote sensing. The paper also explores the factors for vegetation degradation and related habitat areas. The decline in vegetation quality, related species variety, and effect on their habitat areas are checked with NDVI results. The protected areas are subjected to various anthropogenic pressures, including grazing, forest fire, and wood harvesting. The paper highlights the need for effective management strategies to mitigate the identified challenges and ensure the long-term conservation and sustainability of the protected areas. This will ensure more habitat availability, healthy vegetation, genetic exchange between species populations, and a reduction in human-wildlife conflict. The findings of this paper can inform the development of more effective management strategies to protect and conserve these valuable ecosystems.