V
Vladimir Khryashchev
Researcher at Yaroslavl State University
Publications - 51
Citations - 293
Vladimir Khryashchev is an academic researcher from Yaroslavl State University. The author has contributed to research in topics: Convolutional neural network & Face detection. The author has an hindex of 8, co-authored 47 publications receiving 209 citations.
Papers
More filters
Book ChapterDOI
Unscented RGB-D SLAM in Indoor Environment
TL;DR: An improvement of the classical FastSLAM algorithm has been obtained by replacing the method of landmarks’ observations filtering with unscented Kalman filters and an improved resampling algorithm for the particle filtering through the adaptive thresholding based on the data of the effective particle number evolution is presented.
Proceedings ArticleDOI
Research of the speech signal reconstruction at empirical mode decomposition
TL;DR: A high quality of the signal reconstruction at its full restoration, and also advantages of the complementary decomposition in comparison with the customary decomposition are demonstrated.
Proceedings ArticleDOI
Analysis of Pathologies on Endoscopic Images of the Stomach Using SSD and RetinaNet Neural Network Architecture
TL;DR: In this article, neural network algorithms for detecting pathologies on endoscopic images of the stomach have been developed and investigated on an NVIDIA DGX-I supercomputer using a database collected in cooperation with the Yaroslavl Oncological Hospital.
Proceedings ArticleDOI
Improving the eyes localization algorithm for face recognition systems
TL;DR: An eye center localization algorithm based on multi-block local binary patterns with impact on face recognition rate is described and performance is compared to other popular approaches on standard BioID and FERET image databases.
NeuralNetworkAdaptive Switching Median Filter forImage Denoising
TL;DR: A new neuralnetwork adaptive switching median (NASM)filter is proposed to removealt-and-pepper impulse noise from highly corrupted image by combining advantages of the known progressive medianfilter with pulse detection scheme.