Vol 14, No 3 (2018)
- Year: 2018
- Articles: 9
- URL: https://vestnikugrasu.org/byusu/issue/view/651
Articles
Using geoinformation technologies for the study of small thermokarst lakes fields in the western siberia arctic based on satellite images of Kanopus-v
Abstract
Remote studies were performed with use of high-resolution images Kanopus-V at 30 test sites, located fairly evenly on the territory of the Arctic zone of Western Siberia. Data were obtained on the number of lakes and their areas in each test site. The number of lakes varied in different test sites from 135 to 1620, and the area of lakes is from 40 m2 to 400 hectares. Changes in the characteristics of the fields of small thermokarst lakes are studied depending on the landscape zoning. Changes in the density of lakes, their total area in the test areas and limnicity in the Arctic and northern subarctic landscapes are on average quite poorly expressed, but when they move to the southern subarctic zone, their values show a significant increase, which may indicate a difference in the intensity of thermokarst processes in different landscape areas.
Satellite remote sensing approaches and field measurements to tracking coastline changes of the Sambia peninsula
Abstract
The purpose of this work is to assess the accuracy of measuring the coastal boundary of the sea on optical images for the study of the short-term dynamics of the coastal zone of the Sambian Peninsula and its development trends using high-resolution and ultra-high resolution satellite images. The methodological issues of studying the dynamics of the coastal zone of the Sambian Peninsula using ground data and optical images of high and ultra-high resolution are considered. The estimation of the measurement error of the coastline on satellite images by comparison with the data of ground-based measurements of the coordinates of the coastline was carried out. It has been established that remote measurement of the coastal boundary of the sea using optical images is possible with an error not exceeding a few meters. It is shown that Change Detection is used to detect spatio-temporal changes in the coastal zone in multi-time images. Analysis of the quantitative assessment of changes in the area of erosion and accumulative sites on the coast of the Sambian Peninsula over the study period 2010-2017. Showed the prevalence of abrasion processes over accumulative, which is also confirmed by ground-based observations in the survey of the coastal zone after the storm in the period 2012-2017. It has been established that in most of the peninsula, over 5 years, the tendency for the coastline to recede towards the coast from 6 to 36 m on average prevails.
The modelling study of area measurement accuracy using remote sensing techniques
Abstract
Based on the simulation, a relationship was established between the spatial resolution of the satellite image and the relative error in determining the area of the object being measured. The formula is proposed for calculating the relative error of measuring the area of an object using remote sensing techniques. A method for constructing random flat geometric figures with a given shape factor is proposed.
Information extraction using neural language models for the case of online job listings analysis
Abstract
In this article we discuss the approach to information extraction (IE) using neural language models. We provide a detailed overview of modern IE methods: both supervised and unsupervised. The proposed method allows to achieve a high quality solution to the problem of analyzing the relevant labor market requirements without the need for a time-consuming labelling procedure. In this experiment, professional standards act as a knowledge base of the labor domain. Comparing the descriptions of work actions and requirements from professional standards with the elements of job listings, we extract four entity types. The approach is based on the classification of vector representations of texts, generated using various neural language models: averaged word2vec, SIF-weighted averaged word2vec, TF-IDF-weighted averaged word2vec, paragraph2vec. Experimentally, the best quality was shown by the averaged word2vec (CBOW) model.
The algorithm of simultaneous navigation and mapping for mobile robot based on the iterative algorithm of the nearest pixels and descriptor calculated in a circular moving window
Abstract
New combined algorithm for simultaneous navigation and map construction is developed using visual characteristics and depth information to compare images, register 3D-point clouds, and build global sequential 3D-maps of the surrounding space. The performance and computational complexity of the proposed RGB-D SLAM algorithm are presented and discussed with reference and real data. The results can be applied in real-time tracking of objects, in non-cooperative remote observation, and semantic mapping of mobile robot navigation problems.
Inverse problems for mathematical models with the pointwise overdetermination
Abstract
In the article we examine well-posedness questions in the Sobolev spaces of an inverse source problem in the case of a quasilinear parabolic system of the second order. These problem arise when describing heat and mass transfer, diffusion, filtration, and in many other fields. The main part of the operator is linear. The unknowns occur in the nonlinear right-hand side. In particular, this class of problems includes the coefficient inverse problems on determinations of the lower order coefficients in a parabolic equation or a system. The overdetermination conditions are the values of a solution at some collection of points lying inside the spacial domain. The Dirichlet and oblique derivative problems under consideration. The problems are studied in a bounded domain with smooth boundary. However, the results can be generalized to the case of unbounded domains as well for which the corresponding solvability theorems hold. The conditions ensuring local (in time) well-posedness of the problem in the Sobolev classes are exposed. The conditions on the data are minimal. The results are sharp. The problem is reduced to an operator equation whose solvability is proven with the use of a priori bounds and the fixed point theorem. A solution possesses all generalize derivatives occurring in the system which belong to the space with and some additional necessary smoothness in some neighborhood about the overdetermination points.
Restoration of the function of the source and environmental parameters in heat and mass transfer systems with incomplete data of overdetermination
Abstract
The question of well-posedness of the problem of recovering a source function and parameters of an environment in the heat-and-mass transfer systems with incomplete data of overdetermination is considered. The overdetermination conditions are values of a part of the vector of a solution in interior points of a domain.
Wavelet analysis of the El Niño – La Niña phenomenon dynamics and its forecasting
Abstract
Having calculated the frequency content of a solar constant, solar activity from the time series in (1610-2012), the El Niño curve in both (1470-1984) and (1950-2075), it has been found that the frequency content of an El Niño - La Niña curve is induced by frequency contents of solar variables. The frequency contents of the variables have been calculated by developing their wavelet phase-frequency responses. Instantaneous phase differences of the solar variables curves CO2(t), global surface air temperature , El Niño in the two time intervals, in (1891-1950) and (1950-2009), have been calculated; linear approximations with coefficients of instantaneous phase differences between variables in these time intervals have been obtained. Based on relational approximation coefficient analysis of the two time intervals, it has been identified that rising surface air temperature and El Niño alike had been markedly influenced by solar variables variations during the first time interval, with the El Niño rise being affected by that of the surface air temperature amid the global climate change in 1950-2009. The predicted El Niño curves have been obtained from the 2015/16 to 2050 time period by the trained data curve in 1950-2015/16 in two versions as the sum of predicted wavelet approximating and detailing components of the original signal according to the Mallat rule. The accuracy of the predictive El Niño curve values is » 83%. On the obtained curves, coordinates of local maximum and minimum are nearly matching. Wavelet phase-frequency response imaging of one curve reflects an impact on El Niño - La Niña variations of the Earth's solar and climatic variables in the past and the future alike.
Problems of modeling the spatial structure of thermokarst lake fields in the permafrost based on satellite imagery
Abstract
The problems of modeling the spatial structure of the fields of thermokarst lakes in the permafrost zone of Western Siberia based on the geo-simulation approach are considered. The basic principles for the development of a geo-simulation model for the structure of thermokarst lakes fields are outlined. Experimental data on the properties of lake fields obtained on the basis of the integration of high- and medium-resolution satellite images are presented. Data analysis showed that the empirical histogram of the distribution of lakes by size is approximated by the lognormal distribution law by the Pearson criterion with a probability of 0.99. This law of distribution, in contrast to the power law of size distribution of lakes that is commonly used in such studies, also takes into account small lakes, considered as intense sources of methane emission into the atmosphere from thermokarst lakes. An example of modeling the field of thermokarst lakes with a lognormal law of their size-distribution is given.