Wyniki wyszukiwania - Biblioteka Nauki

1

Quantification of bare soil and its spatio-temporal dynamic using different image classification methods

100%

Fernández M. P. , Lavado Contador J. F. , Gutiérrez Á. G. , Lozano Parra F. J. F. J.

|

tom T. 7, z. 2

33-40

EN

Soil is essential for the development of a wide range of vital functions. Its availability and productivity in the long-term is sometimes seriously threatened by human activities. In a general term, soil degradation is defined as the deterioration of the soil function, causing a reduction of its biological potential and productivity that include multiple processes affecting their physical, chemical and biological properties (Imeson, 1998). In rangelands, physical, and also biological degradation, are mainly related to mismanagement. As an example, excessive trampling by livestock usually produces an increase in the bulk density of the soil and a decrease in porosity, significantly reducing the water holding capacity and infiltration rate, finally affecting the pasture productivity, degree of soil cover and soil vulnerability. (Gamougoun et al., 1984; Mulholland and Fullen, 1991). In this respect, the degree of ground cover is known to be one of the key elements in reflecting the soil condition and protecting soil from degradation. The amount and type of vegetation cover strongly depends on climate, topographic factors, lithology and human factors such as the history of land use. The reduction of vegetative biomass and degree of soil cover in rangelands due to grazing causes a decrease in the interception of rainfall, an increase in the percentage of bare soil and, consequently, less protection against the direct impact of raindrops on the soil surface. Indirectly the reduction of biomass, with the consequent reduction of soil organic matter, leads to the physical degradation of the soil, negatively affecting infiltration capacity and water retention properties, and increasing the amount of surface runoff (Imeson, 1998). In the Mediterranean regions, the existence of long summer drought periods coupled with the intense autumn rains, are likely to cause considerable loss of soil material by rainfall erosion. Therefore, rate of uncovered versus covered soil in rangelands is considered as an indicator of soil quality. Quantification of uncovered or bare soil can be done in different ways and at different spatiotemporal scales. Quantifying bare soil surface and its temporal dynamics by field work can be a highly demanding activity in terms of both time and economic costs for medium-to-large scale areas, such as rangeland farms. Using satellite images overestimates or underestimates bare soil surfaces at farm level. This leads to the need for exploring other methods for the quantification of bare soil surfaces. Among the objectives of this study are: (1) to test a set of available techniques and methods for quantifying bare soil surfaces, (2) to analyse the possibilities of orthophotos for land cover analysis and the detection of land degradation phenomena, (3) to compare the efficiency of different methods of image classification and to improve them in order to monitor changes in vegetation cover and bare soil as a soil health indicator. The study was carried out on the 1024 ha farm "Parapunos de Dona María", located in Monroy municipalty, Cáceres province, SW Spain (Fig. 1). Located about 24 kilometres northeast of the city of Cáceres, bounded to the south by the Almonte River, one of the main tributaries of the Tagus river. The farm can be considered as representative of the rangelands (dehesas) of the peneplain geomorphological domain of Cáceres province (Spain), characterized by the presence of thin soils, mainly developed on schist and subject to typical Mediterranean climate characteristics. Vegetation is characteristic of the dehesa wooded rangelands systems, with the presence of therophitic pasture communities with more or less dense shrubs and scattered trees (Quercus rotundifolia). The predominant land use is livestock grazing of sheep, cows and pigs.

PL

Pomiar obszarów odsłoniętej gleby na pastwiskach jest bardzo ważny, gdyż silny wpływ na stopień erozji gleby mają jej podatność na degradację i stan zdrowia. Szczególnie w ubogich w roślinność obszarach śródziemnomorskich, na początku sezonu deszczowego, duża intensywność wypasu zwierzyny hodowlanej może prowadzić do ogołocenia i degradacji gleby, zwłaszcza w miejscach stale odsłoniętych i tych o mniejszym udziale roślinności trawiastej. Zatem ważnym wydaje się opracowanie metody umożliwiającej identyfikację oraz kwantyfikację obszarów odsłoniętej gleby oraz opis ich dynamiki czasowej. W artykule przedstawiono metodę służącą do kwantyfikacji obszarów odsłoniętej gleby i analizy jej czasoprzestrzennej dynamiki, przy użyciu ortofotomap lotniczych. Ortofotomapy z okresów letniego, wiosennego i zimowego przedstawiają farmę położoną na mało zalesionym obszarze, z dużą ilością terenów wypasu (dehes), typowym dla południowo-zachodniej części Półwyspu Iberyjskiego. Do analizy tych ortofotomap zastosowano algorytmy klasyfikacji obrazu. W badaniu wykorzystano obraz panchromatyczny z zimy 1998, dwa trójkanałowe obrazy (RGB) zarejestrowane latem 2002 i wiosną 2006 wraz z komponentem jasności, wszystkie z pikselem o wymiarze 0.4 m. W celu określenia obszarów odsłoniętej gleby zdjęcia zostały sklasyfikowane przy użyciu klasyfikacji nienadzorowanej, nadzorowanej oraz obiektowej. Porównano wyniki uzyskane różnymi metodami. Stopień dokładności klasyfikacji był bardzo zbliżony dla wszystkich porównywanych metod (powyżej 90%). Wykorzystując najlepszą metodę klasyfikacji, określono dynamikę w obszarach o glebie odsłoniętej i porośniętej jako funkcję czasu, analizując zmiany powstałe na obszarach zarówno porośniętych, jak i nieporośniętych. 2009

2

Image classification for jpeg compression

80%

Tichonov J. , Kurasova O. , Filatovas E.

|

tom Vol. 12, no 2

29--34

EN

We analyse storage problems of digital images in accordance with image quality and image compression efficiency. Storage problems are relevant for Cloud storage and file hosting services, online file storage providers, social networks, etc. In this paper, an approach is proposed to process a group of images with a JPEG algorithm that all the processed images satisfy the minimum threshold of quality with the automatic selection of the quality factor (QF). The experimental investigation reveals advantages of the compression efficiency of the proposed approach over the traditional JPEG algorithm. The proposed approach enables saving storage spaces while maintaining the desirable image quality.

3

Object detection in the police surveillance scenario

70%

Wilkowski A. , Kasprzak W. , Stefańczyk M.

|

2019

|

tom Vol. 18

363–-372

EN

Police and various security services use video analysis when investigating criminal activity. One typical scenario is the selection of object in image sequence and search for similar objects in other images. Algorithms supporting this scenario must reconcile several seemingly contradicting factors: training and detection speed, detection reliability and learning from sparse data. In the system that we propose a combined SVM/Cascade detector is used for both speed and detection reliability. In addition, object tracking and background-foreground separation algorithm together with sample synthesis is used to collect rich training data. Experiments show that the system is effective, useful and suitable for selected tasks of police surveillance.

4

Machine-learning at the service of plastic surgery: a case study evaluating facial attractiveness and emotions using R language

61%

Štěpánek L. , Měšťák J. , Kasal P.

|

2019

|

tom Vol. 18

107--112

EN

Since the plastic surgery should consider that facial impression is always dependent on current facial emotion, it came to be verified how precise classification of facial images into sets of defined facial emotions is.