A novel image recognition algorithm based on sequential three-way decisions is introduced to speed up the inference in a convolutional neural network. In contrast to the majority of existing studies, ...our approach does not require a special procedure to train a neural network, and thus it can be used with arbitrary architectures including pre-trained convolutional nets. Each image is associated with a sequence of features extracted at different layers of the neural network. Features from earlier layers stand for coarse-grained image representation. Fine-grained representations include embeddings from one of later layers. Confidence scores of classifiers representing the input image at each granularity level are computed in order to populate a set of unlikely classes with low confidence scores. The thresholds for these scores are chosen by using the step-up multiple testing procedure. The categories from this set are not considered at the next levels with finer granularity. The algorithm selecting the granularity levels and thresholds for each level is trained on a small sample. An experimental study for several datasets and neural architectures demonstrated that the proposed approach reduces the running time by up to 40% with a controllable decrease in accuracy.
The paper addresses the issue of insufficient speed of image recognition methods if the number of classes is rather large. We propose the novel algorithm based on sequential three-way decisions and a ...formal description of granular computing. Each image is associated with principal component scores of the high-dimensional features extracted by deep convolution neural network. Low number of principal components stand for the coarse-grained granules, while fine-grained granules include all components. Initially, first principal components of an observed image and all training instances are matched at the coarsest granularity level. Next, negative decisions are defined by using the multiple comparisons theory and asymptotic distribution of the Kullback-Leibler divergence. Namely, the distance factors (ratios of the minimum distance and all other distances) are evaluated. The set of negative decisions is populated by the instances, for which the distance factors exceed a certain threshold. The images from this set are not examined at the next levels with finer granularity. In the experiments unconstrained face recognition and image categorisation are considered using the state-of-the-art deep learning-based feature extractors. We demonstrate that the proposed approach decreases the running time in 1.5–10 times when compared to conventional classifiers and the known multi-class decision-theoretic rough sets.
Impulsive impact of a submerged body Semenov, Y.A.; Savchenko, Y.N.; Savchenko, G.Y.
Journal of fluid mechanics,
05/2021, Letnik:
919
Journal Article
Recenzirano
Odprti dostop
An analytical solution of the impulsive impact of a cylindrical body submerged below a calm water surface is obtained by solving a free boundary problem. The shape of the cross-section of the body is ...arbitrary. The integral hodograph method is applied to derive the complex velocity potential defined in a parameter plane. The boundary-value problem is reduced to a Fredholm integral equation of the first kind in the velocity magnitude on the free surface. The velocity field, the impulsive pressure on the body surface and the added mass are determined in a wide range of depths of submergence for various cross-sectional shapes, such as a flat plate, a circular cylinder and a rectangle.
An impulsively starting motion of a cylindrical body submerged below a calm water surface in an open container of arbitrary shape is considered. This work generalizes the case of an infinite-depth ...liquid studied by Semenov et al. (J. Fluid Mech., vol. 919, 2021, R4). Particular attention is paid to the interaction between the body, the free surface and the container. The integral hodograph method is employed to derive the complex velocity potential in a parameter plane. The boundary-value problem is reduced to a system of integral equations, which is solved numerically. The velocity field, the pressure impulse on the body and the container wall and the added mass just after the impact are determined for a wide range of depths of submergence and container geometries and for various cross-sectional shapes of the body, such as a flat plate, a circle and a rectangle.
—
The article proposes a new algorithm for solving the problem of real-time detection of vowel speech sounds based on (
R
+ 1)-element information and the whitening filter method. An example of ...practical application of the algorithm is described and an assessment of its efficiency is provided. A full-scale experiment is conducted; its results indicate that the proposed algorithm demonstrates a sufficiently high speed and a guaranteed significance level of decisions with minimal performance requirements to the computing equipment.
This paper is devoted to tracking dynamics of psycho-emotional state based on analysis of the user’s facial video and voice. We propose a novel technology with personalized acoustic and visual ...lightweight neural network models that can be launched in real-time on any laptop or even mobile device. At first, two separate user-independent classifiers (feed-forward neural networks) are trained for speech emotion and facial expression recognition in video. The former extracts acoustic features with OpenL3 or OpenSmile frameworks. The latter is based on preliminary extraction of emotional features from each frame with a pre-trained convolutional neural network. Next, both classifiers are fine-tuned using a small number of short emotional videos that should be available for each user. The face of a user is identified during the real-time tracking of emotional state to choose the concrete neural networks. The final decision about current emotion in a short time frame is predicted by blending the outputs of personalized audio and video classifiers. It is experimentally demonstrated for the Russian Acted Multimodal Affective Set that the proposed approach makes it possible to increase the emotion recognition accuracy by 2–15%.
The article considers the problem of personal biometric data “aging” over time. A method has been proposed to overcome this problem by automatically updating the specified data in the biometric ...system storage using the speech signals of registered users obtained during latest requests for their identification and online service. The proposed method uses a scale-invariant indicator of the voice template quality. As a result, it is characterized by guaranteed reliability of the decisions made in the conditions of a wide speech signal dynamic range. It was established that the use of a scale-invariant indicator provides the guaranteed significance level of decisions made by a conventional observer. A full-scale experiment implementing the proposed method has been set up and carried out using an authoring software; practical justification for the effectiveness of the proposed method with real speech data has been given. The results obtained are intended for using in the development of new and modernization of existing systems and technologies for automated quality control and updating of personal biometric data.
ABSTRACT
Dwarf spheroidal galaxies (dSphs) are the most compact dark-matter-dominated objects observed so far. The Pauli exclusion principle limits the number of fermionic dark matter particles that ...can compose a dSph halo. This results in a well-known lower bound on their particle mass. So far, such bounds were obtained from the analysis of individual dSphs. In this paper, we model dark matter halo density profiles via the semi-analytical approach and analyse the data from eight ‘classical’ dSphs assuming the same mass of dark matter fermion in each object. First, we find out that modelling of Carina dSph results in a much worse fitting quality compared to the other seven objects. From the combined analysis of the kinematic data of the remaining seven ‘classical’ dSphs, we obtain a new 2σ lower bound of m ≳ 190 eV on the dark matter fermion mass. In addition, by combining a sub-sample of four dSphs – Draco, Fornax, Leo I, and Sculptor – we conclude that 220 eV fermionic dark matter appears to be preferred over the standard cold dark matter at about the 2σ level. However, this result becomes insignificant if all seven objects are included in the analysis. Future improvement of the obtained bound requires more detailed data, both from ‘classical’ and ultra-faint dSphs.
The present paper discusses the problem of distortions in speech signals transmitted over a communication channel to a biometric system during voice-based remote identification. A possible ...rectification approach involves a preliminary correction of the frequency spectrum of the received signal based on the pre-distortion principle. Taking into account a priori uncertainty, a new information indicator of speech signal distortions is proposed, along with a method for its measurement under conditions of small observation samples. An example of fast practical implementation of the method based on a parametric spectral analysis algorithm is considered. Results of an experimental test of the proposed approach are provided for three different communication channel instantiations. It is shown that the proposed method facilitates the transformation of an initially distorted speech signal into compliance with a registered voice template using an acceptable algorithmic information discrimination criterion. The described approach may be used in existing biometric systems and speaker identification technologies.
The problem of determining a fundamental tone frequency of a speech signal in the presence of white Gaussian noise is examined. A method for measuring this frequency is proposed which takes into ...account the periodic structure of the power spectrum of voiced speech frames and is based on the principle of harmonic energy accumulation in the frequency domain. For this purpose a procedure for equalizing the envelope of the power spectrum is introduced in the algorithm for processing a speech signal using a two-level autoregression model of the observations: within the limits of a single period of the fundamental tone and within an interval of several of these periods. Here adaptation of the order of the autoregression of the lower level to the observed frame is planned. An example of the practical realization of the adaptive method based on the Berg method is examined. The basic advantages of the adaptive method compared to the known analogs are high speed and enhanced noise stability, which are confirmed in a full-scale experiment. A gain in threshold signals of 5-10 dB was obtained through use of the adaptive method.
PDF
