Section 2 - Track 3

On this page, talks in the format of posters are posted. They can be viewed during the whole time of the Conference. To send your question to the authors, please follow to Miro and write your question in the comments. You can also use the feedback form on our website. Authors will answer questions using Miro comments or via email. To get manual go to Instructions (eng / rus).

Vladislav Butorov and Marina Chicheva
Research of lossy image compression algorithm based on fractal discrete cosine transform
Lossy image compression algorithm based on fractal discrete cosine transform is proposed in this paper. Created algorithm is compared to algorithm based on two-dimensional discrete cosine transform. It is shown experimentally that the described algorithm brings less distortion concerning block structure in comparison with square blocks of two-dimensional discrete cosine transform. It is remarked that visual quality characteristics of both algorithms vary poorly for several values ranges of entropy of compressed image.

Nikita Demin, Nataly Ilyasova, Aleksandr Shirokanev and Evgeniy Zamyckij
Segmentation of OCT images for localizing of diabetic macular edema
A method for localizing the region of diabetic macular edema in the fundus images is proposed based on the analysis of optical coherence tomography (OCT) data. The Canny method is used to find the boundary between the vitreous body and the retina in OCT images. The segmentation method, based on the Kruskal algorithm for constructing the minimum spanning tree of a weighted connected undirected graph, is used to select the retina to the pigment layer in the image. Using the obtained segmentation results, a map of the thickness of the retina of the eye and its deviation from the norm were constructed. In the course of the research, the optimal parameter values were selected in the Canny and graph segmentation algorithms, allowing to achieve a 5% error in the allocation of the region of interest. SIFT, SURF, and AKAZE methods were considered for overlaying calculated maps of the thickness of the retina of the eye and its deviation from the norm on the fundus image. In cases where, along with OCT data, a picture is provided from the fundus of the camera of the OCT apparatus, using the SURF method, it is possible to precisely matching with the fundus image.

Olga Belova, Natalia Vlasova, Ludmila Kavelenova, Eugene Korchikov, Victor Fedoseev, Tatiana Chap and Anna Denisova
Monitoring of the recreation effects on land cover with the use of an unmanned aerial vehicle on the example of the Strelnaya mountain in Samara region
The Mount Strelnaya is one of the objects most exposed to recreational stress in the Zhigulevsky State Nature Reserve. Until recently, monitoring studies of the recreational pressure carried out by university staff have been limited to either inspecting plant communities from the metal decking installed on the trace, or descending to the surface of the slope and move along it with the risk of injuring vulnerable vegetation cover. The use of unmanned aerial vehicles (UAVs) opens up new prospects for quick and efficient identification of points affected by recreational exposure, including those remote from the floor, without contacting the slope surface. The first experience of integrating ground-based and UAV-based monitoring was carried out in 2019. The Mount Strelnaya was surveyed in spring and autumn using UAVs. The obtained images were reclassified using the method of reference vectors with radial base functions into "trace" and "non-trace" classes. As a result, it was possible to automatically select trampled slope sections with a high fraction of accuracy. The preliminary results of the work are presented

Vladimir Grishanov, Igor Malov, Georgiy Pleshakov and Seda Gevorkyan
Parameterization of fluorescent images of external tissues of the body for diagnostic purposes
The study of endogenous fluorescence of external tissues of the body is a powerful diagnostic tool. A promising class of diagnostic fluorimeters are fluorimeters that form color images of the surface area excited by radiation of the corresponding spectral composition. In addition, such fluorimeters register images in white light. As a result, two color frames become the object of processing in order to extract diagnostic information. Each frame has three color components. On the basis of experiments carried out on a fluorimeter designed to evaluate the content of glycation end products in the skin (peak wavelength of fluorescence excitation 365 nm), the representative parameter was the ratio of the arithmetic mean value (AMV) of the pixel in the green component of the fluorescent image to the AMV of the pixel in the green component of the image in white light. The use of this ratio revealed age-related and seasonal changes in skin autofluorescence. For the eye sclera at a wavelength of 405 nm fluorescence excitation, the lowest variability was demonstrated by the ratio of the AMV of the pixel in the green component to the AMV of the pixel in the red component of the fluorescent image frame.

Alexey Ruchay, Konstantin Dorofeev and Vsevolod Kalschikov
Accuracy analysis of 3D object reconstruction using mesh filtering
We analyze the accuracy of 3D object reconstruction using mesh filtering applied to data from a RGB-D sensor. Various methods of mesh filtering are tested and compared with respect to the reconstruction accuracy using real data. In order to improve the accuracy of 3D object reconstruction, an efficient method of mesh filtering is designed. The presented results show an improvement in the accuracy of 3D object reconstruction using the proposed mesh filtering algorithm.

Maksim Baranov and Tristan Malleville
Determination of the structures contours parameters in biological films for the development of the cuneiform dehydration method
It is well known, that there are some correlations between diseases and geometrical parameters of structures in dehydrated films of biological liquids, in particular in blood serum films. It is necessary to use digital image processing methods for exploring structures and to correlate of their parameters with pathologies. In this paper we describe the creation of algorithm and computer program for analysis of geometrical parameters of structures in dehydrated blood serum films.

Nina Vinogradova, Andrey Sosnovsky and Stepan Egorov
Analysis of the accuracy of determining the vegetation edges according to the Landsat remote sensing data over the territory of the Sverdlovsk region
The work is devoted to the study of the most commonly used vegetation indices in relation to the territory of the Sverdlovskaya region according to Landsat-7 images. For the image fragments, vegetation maps were constructed using various indices. The accuracy was evaluated of the vegetation map according to digital topographic maps based on the criteria of false alarm errors and missing errors, as well as the total number. The indices having the smallest errors are found. Recommendations on the use of indices of vegetative regions covered by coniferous and mixed forests.

Andrey Kuznetsov and Artem Lanin
Splicing detection based on improved FISH descriptors
Fake images are becoming more common in the modern world. Influencing our opinion about a person, they may cause considerable damage. To detect such images, automatic detectors are needed. This article presents a method for automatic detecting splicing using computer vision, based on a comparison of the illumination parameters of faces in a single image.

Vladimir Panishchev and Sergey Poltoratskiy
Hardware-oriented algorithm for extracting periodic sequence of digital signals
The problems of applying pulse sequence processing algorithms at the hardware level to specialized hardware for image processing and recognition systems, monitoring, analysis, control and diagnostics of complex technical objects and man-machine systems are considered. A hardware-oriented algorithm for extracting periodic sequences in control and digital signal processing devices is described. To improve the detection of deterministic sequences the algorithm analyzes multiple values of the period of pulses arrival by weighted processing.

Lubov Shiripova, Olga Strukova and Evgeny Myasnikov
Study of classification techniques for PCA-based human action recognition
The work presents the results of research of the classification techniques for human action recognition based on PCA and width vectors using a video recorded in the optical range. The method used in this paper consists in the detection of a moving person on a video sequence with size normalization, formation a set of subsequences and feature vectors. The classification of the human action is carried out using support vector machine with different kernels, classifier K-Nearest Neighbors and random forest classifier. The obtained results allowed us to pick up the most effective parameters for the classifiers.

Dmitry Murashov, Yury Obukhov, Ivan Kershner and Mikhail Sinkin
Algorithm for identifying artefact events based on the analysis of video EEG data for monitoring patients with craniocerebral injuries
One of the problems solved by analyzing the data of long-term video EEG monitoring is the differentiation of epileptic and artifact events. For this, not only multichannel EEG signals are used, but also video data analysis, since traditional methods based on the analysis of EEG wavelet spectrograms cannot reliably distinguish an epileptic seizure from a chewing artifact. In this paper, we propose an algorithm for detecting artefact events based on a joint analysis of the level of the optical flow and the ridges of wavelet spectrograms. The preliminary results of the analysis of real clinical data are given. The results show the possibility in principle of reliable distinguishing non-epileptic events from epileptic seizures.

Ilya Kolobov, Alexander Korobeynikov and Alexander Lozhkin
The microcircuit images analysis based on convolutional neural network
The paper deals with the analysis of microcircuit images using an artificial neural network. It was used a convolutional network encoder-decoder (Encoder-Decoder Convolutional Neural Network, ED-CNN) based on the U-net architecture to determine the boundaries of transistor elements: the task of segmentation of microcircuit images. As a result, was obtained an average binary accuracy of 90% for determining the ownership of pixels and 16% loss for binary cross entropy on the test sample. The proposed system effectively solves the problem of segmentation of microcircuit images based on standard blocks of modern convolutional neural networks.

Back to Top