Geomatics (Norway)

Today, autonomous mobile robots are widely used in all walks of life. Autonomous navigation, as a basic capability of robots, has become a research hotspot. Classical navigation techniques, which rely on pre-built maps, struggle to cope with complex and dynamic environments. With the development of artificial intelligence, learning-based navigation technology have emerged. Instead of relying on pre-built maps, the agent perceives the environment and make decisions through visual observation, enabling end-to-end navigation. A key challenge is to enhance the generalization ability of the agent in unfamiliar environments. To tackle this challenge, it is necessary to endow the agent with spatial intelligence. Spatial intelligence refers to the ability of the agent to transform visual observations into insights, insights into understanding, and understanding into actions. To endow the agent with spatial intelligence, relevant research uses scene graph to represent the environment. We refer to this method as scene graph-based object goal navigation. In this paper, we concentrate on scene graph, offering formal description, computational framework of object goal navigation. We provide a comprehensive summary of the methods for constructing and applying scene graph. Additionally, we present experimental evidence that highlights the critical role of scene graph in improving navigation success. This paper also delineates promising research directions, all aimed at sharpening the focus on scene graph. Overall, this paper shows how scene graph endows the agent with spatial intelligence, aiming to promote the importance of scene graph in the field of intelligent navigation.

Soil heavy metal (HM) pollution constitutes a global environmental issue. Measuring the soil magnetic susceptibility (MS) is straightforward and expeditious. However, the correlation of HMs with MS in agricultural soils remains ambiguous. This study aimed to determine the correlation between HMs and soil MS on Black Soil farmland in Northeast China. Soil samples were selected from three typical black soil farmland slopes in the 0–20 cm topsoil. Based on measurable magnetic and chemical indexes, the pattern of soil MS along the slope and the environmental implications of HMs were evaluated. The results show that MS was higher on the lowerslope of farmland hillslopes corresponding to deposition position; the most prominent HMs in the studied farmland are Pb and Ni; the predominant source of Pb and Ni is agricultural machinery, while the source analysis revealed that the primary source of Cu is chemical fertilizers; and the correlation analysis showed a negative correlation between MS and Cu, Cr, Pb, and Zn, and positive correlation with Ni, and Mn. These results prove the relationship between magnetism and heavy metal elements, which providing the possibility of using soil magnetism to estimate soil pollution. Further data are needed to verify the exact correlativity of MS and HMs in soil with different land uses.

Abstract This study evaluates the performance of the Soil and Water Assessment Tool (SWAT) hydrological model in simulating daily river flow and quantifying flow components within the Isser catchment, upstream of the Koudiat Acerdoune dam in Algeria. Model parameters were optimized using the SWAT-CUP optimization method and the SUFI-2 algorithm. The model was calibrated (2002-2012) and validated (2013-2018) against observed daily streamflow data, following a three-year warm-up period (1999-2001). Sensitivity analysis identified key parameters influencing streamflow, including groundwater delay, baseflow processes, channel hydraulic conductivity, and surface roughness. The SWAT model demonstrated strong performance, achieving satisfactory to very good performance metrics (e.g., R2 Cal/Val = 0.86/0.82; NSE Cal/Val = 0.81/0.74), indicating its ability to accurately simulate daily streamflow in the Isser watershed. Water balance analysis revealed that actual evapotranspiration accounted for 51% of precipitation. Shallow aquifer recharge constituted a major loss at 28%, while total water yield represented the remaining 21%. Furthermore, the model estimated that surface runoff accounted for approximately 47.4% of the total water yield, lateral flow for 11.6%, while baseflow contributed the remaining 41%. These findings demonstrate the SWAT model's capability to effectively represent the hydrological processes governing the Isser watershed. The study underscores the importance of considering both surface and subsurface water dynamics to guide sustainable water resource planning in semi-arid regions.

Accurate land use and land cover (LULC) classification is essential for sustainable resource management and understanding landscape changes due to climate variations. High-quality datasets and robust classification methods are necessary for effective LULC classification. This study evaluates the LULC pattern of Pokhara Metropolitan City using Landsat 5, 7, and 8 satellite imagery over 30 years (1991-2021) through supervised classification techniques, employing random forest (RF) and support vector machine (SVM) algorithms to compare their accuracy. The study further utilizes the artificial neural network (ANN) model in the MOLUSCE plugin in QGIS to predict future LULC patterns for 2031 and 2050. Results indicate a significant urbanization trend, with agricultural land decreasing from 54.74% in 1991 to 32.02% in 2021, while built-up areas and forest cover increased from 2.93% to 10.96% and from 38.87% to 53.56%, respectively. Future predictions show continued trends, with agricultural land decreasing to 26.62% and built-up areas increasing to 12.08% by 2031. The findings provide crucial spatial information for urban planners, meteorologists, and policymakers to develop local and regional strategies for sustainable development.

In this article, we analyze the relationship between class sizes and examination results in higher education. Data from the University of Stavanger, collected in the Common Student System (FS) in the period from 2011 to 2021, shows that there is a negative correlation between the number of students enrolled on courses and examination results. An increase in the number of students taking a course reduces the average scores. We also show that the difference in examination results for different class sizes decreases when the course level increases.

Abstract Aims Land surface emissivity (LSE) is an important variable in soil studies. Although there are various remote sensing methods to estimate LSE, accurately predicting LSE still remains a major challenge. Typically, the correlation between LSE and the Visible Near Infrared bands is employed for LSE estimation. However, some studies have raised some concerns about this correlation, especially in bare soil areas. Therefore, it is necessary to conduct further investigation to determine if there exists a nonlinear relationship between the LSE and other spectral bands, which was not detected by simple linear correlation/regression. Methods In this study, firstly, a deep Auto-encoder (DAE) network has been used to investigate the correlation between LSE and other spectral bands. Subsequently, we have applied a Conditional Generative Adversarial Network (CGAN) to estimate the LSE. The proposed CGAN was trained using the Landsat and ECOSTRESS satellite datasets. The performance of the developed network was then compared with NDVI-based method on satellite/simulated-based bare soil pixels. Results For satellite data, the RMSE (Root Mean Squared Error) and correlation coefficient (R) between the estimated LSE using proposed CGAN and ECOSTRESS LSE are 0.005 and 0.97, respectively. For the simulated data, the RMSE and R between the estimated LSE and the simulated one are 0.01 and 0.92, respectively. Conclusion The results of the DAE show considerable relationship between the LSE and Short-Wave Infrared bands which not be seen using simple linear correlation.

Abstract Vegetation is affected by the complex urban environment. Exploring vegetation change trends from an urban-rural gradient perspective can help understand the response of vegetation to urbanization and provide scientific basis for predicting future vegetation change. Based on Landsat data processed in the Google Earth Engine (GEE) environment, Enhanced Vegetation Index (EVI) change in the Guanzhong Plain urban agglomeration from 2000 to 2020 was calculated, and EVI change was analyzed from the perspective of the urban-rural gradient and different city sizes. Then, the impacts of biogeochemical driving factors and land cover changes to the EVI trend were investigated. The results showed that: (1) The EVI trends of the Guanzhong Plain urban agglomeration had a "V"-shaped characteristic along with the urban-rural gradient. The EVI trends for the urban cores (UC), urban new towns (UNT), urban fringes (UF), rural fringes (RF), and rural backgrounds (RB) were 0.0459 ± 0.0080/10a, 0.0347 ± 0.0093/10a, -0.0090 ± 0.0121/10a, 0.0738 ± 0.0075/10a and 0.0817 ± 0.0082/10a, respectively. (2) From the perspective of different city sizes, EVI in the UC, UNT, and UF increased as city size increased. EVI in the UC, UNT, and UF respectively increased from 0.0333 ± 0.0058/10a, 0.0281 ± 0.0077/10a, and − 0.0189 ± 0.0082/10a in small towns to 0.0520 ± 0.0091/10a, 0.0402 ± 0.0120/10a, and − 0.0114 ± 0.0042/10a in megacities. (3) EVI change in the UC was mainly affected by urban biogeochemical driving factors (UBD), and its contribution rate was 51.9%. EVI change in UNT was mainly affected by UBD, background biogeochemical drivers (BBD) and urban expansion or densification (UED), which had contribution rates of 37.2%, 30.7%, and − 29.6%, respectively. EVI change in the UF was mainly affected by UED, and its contribution rate was − 51.9%. EVI change in RF and RB were mainly affected by BBD, with contribution rates of 70.2% and 94.0%, respectively. (4) The contribution rates of UBD in the UC, UNT, and UF generally in-creased as city size increased. The above results indicate that biogeochemical drivers and land cover changes jointly regulate EVI change trends in the urban-rural gradient of the Guanzhong Plain urban agglomeration, which can help assess the impact of complex environmental changes on vegetation changes in urban regions.

Building and completing the land price database contributes to completing the land database, promoting digital transformation, improving the effectiveness and efficiency of the state management of land, while increasing transparency in deciding land prices, increasing interconnectivity and modernizing public land services. However, nationwide land price data has not yet been formed, studied. Reserach projects of the Ministry of Natural Resources and Environment are still mainly focusing on solving land valuation problems and building a land database. In the world, the construction of an online land price information system has been implemented on the basis of WebGIS technology and achieved many good results. This article introduces the process and experimental results of building a land price database in Dong Cuong commune, Yen Lac district, Vinh Phuc province. Tools used in this study include ArcGIS Online, ArcGIS API, Microsoft SQL SERVER; Spatial land price data is built from cadastral spatial data after standardization according to Circular No. 05/2017/TT-BTNMT of the Ministry of Natural Resources and Environment. The experimental results of building a Land Information System identify 4 groups of users: Guests, Members, Data Management and System Administrator. The land price information system is designed in the form of a Web-app, installed on the website with the address: https://giadat.gis-humg.com.

With the rise of the Internet of Things (IOT), mobile devices and Location-Based Social Network (LBSN), an abundant amount of trajectory data has made research on location prediction more popular. The check-in data shared through LBSN hides information related to life patterns, obtaining this information is helpful for location prediction. However, the trajectory data recorded by mobile devices is different from check-in data that has semantic information. In order to obtain the user’s semantic, relevant studies match the stay point to the nearest Point of Interest (POI), but location error may lead to wrong semantic matching. Therefore, we propose A Self-Attention for Next Location Prediction based on Semantic Mining to predict the next location. When calculating the semantic feature of a stay point, the first step is to search for the k nearest POI, then use the reciprocal of the distance from the stay point to the k nearest POI and the number of categories as weights. Finally, we use the probability to express semantic without losing other important semantic information. Furthermore, this research combined with sequential pattern mining can bring richer semantic features. In order to better perceive the trajectory, temporal features learn periodicity of time series by sine function. In terms of location features, we build a directed weighted graph and regard the frequency of users visiting locations as the weight, so the location features are rich in contextual information. We then adopt the Self-Attention model to capture long-term dependencies in long trajectory sequences. Experiments in Geolife show that the semantic matching of this study improved by 45.78% in TOP@1 compared with the closest distance search for POI. Compared with the baseline, the model proposed in this study improved by 2.5% in TOP@1.

Reconstruction of the trajectories of moving objects using context-based dynamic time warping similarity measure method