The INTAS project for the elaboration of automated technology of land cover classification: the scientific problems, main results and prospects

Brodsky, Lukas; Bushuev, E.I.; Voloshin, V.I.; Kozlova, A.O.; Parshina, O.I.; Popov, Mykhailo; Sablina, V.I.; Sakhatsky, A.I.; Sirotenko, A.V.; Soukup, Tomáš; Stankevich, Sergey А.; Tarariko, A.G.

doi:10.15407/knit2009.02.036

Cited by 8 publications

(11 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is reasonable to use both the images of the multiarea space shots and NDVI and textural indices prepared in the stage of ERS preliminary processing as the initial informative layers for the solution of the taskthe object oriented classi cation of the agricultural lands mask [20]. Having shaped the hierarchy of classes and established solution-making rules and threshold values for each class, classi cation is accomplished at the levels, which, for example, identify sowing of winter cultures and natural vegetation, at that false single objects are moved away.…”

Section: Resultsmentioning

confidence: 99%

Untitled

2014

Agric. sci. pract.

View full text Add to dashboard Cite

The contemporary development of the remote space survey syst ems and elements of geoinformation t echnolo gies o ffers the fundamentally new possibilities of control, forecasting and interpretation of the data obtained from agroecological monitoring. Aim. To describe natural and climatic conditions of the various zones within the territory of Ukraine, its agricultural acquirement and risk of the soils' erosive degradation manifestations in the meaning of climate changes. To determine the factors in uencing upon the spectral characteristics of the eroded soils for their identi cation, deciphering, and also the cultivated lands and land tenure systems degradatio n monitoring according to satellite data. Methods. The logical model of water erosion determination and identi cation according to the data of the Earth remote sensing (ERS) of high spatial resolution is developed on the basis of classi cation in basic deciphering signs and the procedure of molding of the training samples forming. The materials of the Landsat 8, SPOT, ASTER and RapidEye space surveys, map materials and data of full-scale ground observations on the test objects were used for identi cation of the processes of sheet and linear erosion. The soil erosion was determined according to two approaches. The rst one is based on the plowed soil and the second -on soil covered with plants. The soil erosion class was determined according to the spectral characteristics and humus content, while gully rateby reference to gullies' length and square. Results. The humus content in so il was proposed to be determined according to the spatial distribution of spectral characteristics within the limits of uniform regions and corresponding mathematical-statistical models. The opportunities of linear and sheet erosion classi cation according to the ERS data, and also their use in the system of monitoring and evaluating the ecological state of agrolandscapes and land tenure systems are shown. C onclus ions. The space mo nitoring data of the soils erosive degradation and agrolandscapes in whole provide the opportunity of more effective use of soil resources due to the strategic determination of degradation processes with the subsequent planning and workout the measures for th e optimization of the erosive dangerous agrolandscapes structure, and also introduction of the ground water-guarding systems of soil management.

show abstract

Section: Resultsmentioning

confidence: 99%

Untitled

2014

Agric. sci. pract.

View full text Add to dashboard Cite

show abstract

“…SWOT and Gap-analysis of agricultural policy were applied to implement the Rio Conventions within the framework of UNDP/GEF (United Nations Development Programme / Global Environmental Facility) "Integration of the Rio Conventions into the national policy of Ukraine" [1], the investigation involved the use of spatial imagery of different spatial resolution, methodological and regulatory provision of the Institute of Agroecology and Environmental Economics of the National Academy of Agrarian Sciences (IAEE NAAS) on establishing the network of testing agrarian grounds (TAG), the set of indices for vegetation, erosion degradation of soils and landscape diversity [2][3][4]. The investigation of the impact of climate changes on vegetation was conducted on the territory of natural-climatic zones of Polissia, Forest-Steppe and Steppe which were geographically represented by Chernihiv, Poltava and Zaporizhzhia regions respectively.…”

Section: Methodsmentioning

confidence: 99%

Satellite agroecological monitoring within the system of sustainable environmental management

Tarariko¹,

Ilienko²,

Kuchma³

et al. 2019

Agric. sci. pract.

View full text Add to dashboard Cite

Satellite data are a relevant part of information, required for sustainable environmental management, assessment of the impact of economic activity of ecosystems, determination of risks, related to global climate changes, desertifi cation processes, loss of landscape and biotic diversity. Aim. To substantiate the reasonability and prove the effi ciency of using satellite data in the agroecologic monitoring system regarding the impact of climate changes on vegetation, processes of soil erosion degradation, and assessment of landscape diversity. Methods. The study was conducted in the territory of Ukraine. It involved the application of SWOT and Gap-analysis methodology, materials of NOAA satellite observa- tions, Sentinel, different spatial resolution, methodological and regulatory provision of the Institute of Agroecology and Environmental Economics of the National Academy of Agrarian Sciences regarding satellite monitoring of the structure of agrolandscapes, norms of establishing a network of testing agrarian grounds, list of vegetation state indicators, in par- ticular, “Remote sensing of the Earth from space. Land data about controlling the condition of plantings and performance of agricultural crops. General requirements: DSTU 7307:2013”, “Remote sensing of the Earth from space. Ground in- spection of plantings. Classifi er of objects and functions: SUC 01.1-37-907:2011”, “Methodological recommendations on establishing the network of testing agrarian grounds in the system of monitoring of plantings using the materials of cosmic information”. The investigation on the impact of climate changes on vegetation state was conducted on the territory of three natural-climatic zones which were geographically represented by Chernihiv, Poltava and Zaporizhzhia regions re- spectively. The determination of the threat of erosion degradation of arable lands and landscape diversity was performed on the territory of two administrative districts with high level of ploughness of agrolandscapes, intense agrarian produc- tion and manifestation of erosion degradation of lands. Results. Inadequacy of the traditional system of agroecological monitoring was determined. It was proven that it was reasonable to have comprehensive application of satellite data regarding climate warming within the natural climatic zones and its impact on vegetation according to the normalized dif- ference vegetation index (NDVI), erosion degradation of soils and landscape diversity. According to satellite data of the National Oceanic and Atmospheric Administration (NOAA), the correlation analysis was performed on the connection between the dynamics of the sum of effective temperatures and the sum of NDVI values for the vegetation period. There was positive impact of climate warming on vegetation state according to NDVI index in the zone of Polissia and Forest- Steppe. The correlation coeffi cients were R = 0.64 and R = 0.77 respectively. In the Steppe zone the correlation coeffi cient dropped down to R = 0.35 which demonstrated the elevated risk of droughts. Conclusions. Satellite data of Sentinel-1 were used to determine critical zones of erosion degradation of arable lands, requiring preservation and their inclusion to the natural fi elds, which had a positive impact on the optimization of agrolandscape diversity.

show abstract

“…where H(C, B 1 , ..., B m ) is a cumulative joint entropy of fuzzy attributes, which describes the equation [14] H(B 1 ,...,B m ) = log 2 m×n -log 2 …”

Section: Fuzzy Decision Tree Modelmentioning

confidence: 99%

“…The CORINE Land Cover (CLC) implementation over whole territory is a mandatory requirement for all countries of the European Union. In reality, land cover accuracy rarely obtains the commonly recommended 85% target [2].…”

Section: Introductionmentioning

confidence: 99%

Fuzzy decision tree model adaptation to multi- and hyperspectral imagery supervised classification

Stankevich

Levashenko

Zaitseva

2013

The International Conference on Digital Technologies 2013

View full text Add to dashboard Cite

Now the land cover classification system is very important for various remote sensing applications and many sectors of economy. Therefore, development of algorithms for multi-and hyperspectral imagery classification is an urgent task. In this paper we present a new efficient algorithm for multi-and hyperspectral imagery classification based on fuzzy decision tree approach. We use the multispectral imagery spectral bands as fuzzy data source attributes and cumulative mutual information between them and the resulting fuzzy classification as decision tree inducing criterion. Proposed algorithm provides classification accuracy than traditional ones and significant data dimensionality reduction by means of informative spectral bands selection.

show abstract

The INTAS project for the elaboration of automated technology of land cover classification: the scientific problems, main results and prospects

Cited by 8 publications

References 1 publication

Untitled

Untitled

Satellite agroecological monitoring within the system of sustainable environmental management

Fuzzy decision tree model adaptation to multi- and hyperspectral imagery supervised classification

Contact Info

Product

Resources

About