Development and use of feature models of anthropogenic objects in thematic processing of space imagery

The article describes the problem of automated recognition of the anthropogenic elements of landscape. The recognition is based on the aerospace data in the optical range of the spectrum and a feature model of an object, consisting of the geometric and reflectance characteristics. Using this model, we formed training samples for a convolutional neural network. There is a real example of the practical implementation of the model in identification of the aviation objects.

Sharpness indicator of aerial- and space images obtained for mapping purposes

The main characteristics of aerial- and space imagery, obtained for mapping purposes, are the distinct details transferring and sharpness, which determine their visual properties. The sharpness of aerial- and space images affects the accuracy of measurements made on them, as well as the quality products obtained. The ability of the image to transfer fine details is separately determined by its actual spatial resolution. It is shown that the actual resolution does not completely represent the structural-and-geometric characteristics of the images’ fine quality. Calculations showing that neither actual spatial resolution nor the gradation characteristic “local contrast” make it possible to evaluate its sharpness are presented. The author proposes a method of numerically image sharpness assessing, based on the gradient characteristic of the edge profile curve. The advantage of this method in comparison with alternative ones is shown. The general form of the image sharpness criteria calculating formula for any radiometric resolution is given. The proposed method enables obtaining a normed value, varying from 0 to 1, which provides obvious interpretability of the result. The permissible value of the sharpness index of aerial and space imagery obtained for mapping is determined. Examples of images obtained by an aerial survey complex based on a light drone, as well as those obtained by a remote sensing spacecraft, satisfying the established acceptable sharpness criteria, are given. The correspondence of obtained numerical estimates to the visual perception of the image sharpness is shown.

Comparative analysis of satellite images obtained by spacecraft “Canopus-V” and “Aist-2D”

A comparative analysis of the remote sensing data, obtained by spacecrafts “Canopus-V” and “Aist-2D”, in terms of determining their actual spatial resolution – the smallest linear size of the terrain object that can be transmitted by the image. Within the researches of studied space imagery structural transfer properties the calculations of their general and local contrast for different types of underlying surface were performed. In particular, it was shown, that despite the smaller ground sampling interval (GSI), images, obtained by “Aist-2D”, are comparable in their structural transfer characteristics to the images of “Canopus-V” because of the greater blurring. However, “Canopus-V” images have a disadvantage – the artifacts due to direct solar reflection from smooth (mirror) surfaces, which “Aist-2D” images do not have. Basing on the obtained values of structural transfer characteristics and the requirements of the Instructions on photogrammetric work for digital topographic maps and plans creation, the mapping scale, corresponding the attainable accuracy of measurements, made by images, obtained by spacecrafts “Canopus-V” and “Aist-2D”, is determined. The conclusions about the comparability of considered space imagery structural transfer characteristics, the advantages and disadvantages in terms of their photographic quality, as well as the possibility of their use for mapping purposes are formulated.

МОНІТОРИНГ СІЛЬСЬКОГОСПОДАРСЬКИХ КУЛЬТУР ІЗ ЗАСТОСУВАННЯМ КОСМІЧНИХ ЗНІМКІВ SENTINEL-2

The subject of the study in the article is the growing need for the use of spatial information for efficient agricultural production, due to the growing tendency of Earth remote sensing data accessibility, which, due to the spatial and temporal resolution improvement, can be used in the land cover analysis and other related jobs. The goal is to review the obtaining process of satellite multispectral space imagery from Sentinel-2 and to consider the possibility of their use for monitoring crops during the entire vegetation phase. The tasks: to study the modern needs of agricultural producers in the field of analysis of land cover occupied by agricultural crops; the analysis of the European Space Agency programs and the global land program Copernicus, which uses spatial information from Sentinel-2 for use in the agricultural sector; estimation of the constellation characteristics of Sentinel-2, imaging equipment and remote sensing data processing results by ground services received from Internet services; the use of Sentinel-2 multispectral space imagery for monitoring crops during the entire vegetation phase. The following results were obtained. After analyzing agricultural producers needs and the European Space Agency program, the feasibility of using multispectral space images taken by the Multispectral Instrument installed on satellites Sentinel-2 was established. Free access to the space imagery database is provided through the Copernicus Open Access Hub Internet Service. For the researched territory, Poltava region, Chutov district, the village of Vilkhovatka, various time space images were obtained and the normalized difference vegetation index (NDVI) was calculated. Histogram analysis of the obtained vegetation index values distribution within a single field (corn to grain) allows to reveal a quantitative and qualitative change in biomass, indicating a change in the vegetative phase. Conclusions. The approach described in this paper allows to conduct monitoring of the cropping state during the vegetation phase using both qualitative – visual analysis and quantitative – NDVI index, criteria. The change in the values of the normalized difference vegetation index can reveal a change in the biomass state. However, for calculating the NDVI index, data from near-infrared and red channels is needed, which complicates the acquisition of the original image. Therefore, in order to obtain the quantitative criteria in subsequent jobs, it is expedient to consider the possibility of using fractal dimension, which will reduce the amount of input data required for calculations.