Probability Of Correct Recognition Research Articles

Individual age determination based on computed tomography knee analysis using artificial neural networks and computer vision: Preliminary results

BACKGROUND: Currently, studies have focused on the modernization of existing methods of forensic age assessment (bone and skeletal) through the active use of modern methods of medical imaging (e.g., computed tomography) and artificial intelligence for their analysis. This approach enables the creation of new methods for assessing biological age, which is characterized by increased accuracy and reproducibility.
 AIM: To develop and test an algorithm for predicting the biological age of an individual based on computed tomography analysis of the knee joint using artificial neural networks and computer vision.
 MATERIALS AND METHODS: This observational retrospective transverse (one time) study analyzed computed tomography scans (334) of the knee joint performed in the Departments of Radiation Diagnostics of the Priorov Central Institute for Trauma and Orthopedics, Vreden National Medical Center for Traumatology and Orthopedics, between 2018 and 2021. The study enrolled persons of both sexes aged 13–45 years. Cases of developmental abnormalities, knee injuries, signs of general connective tissue pathology were excluded. Research methods include the use of intelligent information technologies (a formalized set of mathematical and software solutions).
 RESULTS: Based on the experiments conducted, an algorithm for assessing age according to the computed tomography scans of the knee joint has been developed. The main components of the developed system are as follows: a preprocessing module, an intelligent computing core, a data analysis module, a three-dimensional reconstruction module, a property extraction module, and a final age assessment module. The essence of the proposed method is the simultaneous use of artificial neural networks and clearly formalized mathematical procedures for calculating the properties of the epiphyseal line. To obtain the results and conduct primary experimental studies that confirmed the feasibility, correctness, and operability of the method, software using the YOLOv5 neural network was developed. The result of the error matrix analysis after training shows a probability of correct recognition of the order of 80%. Verification of experimental studies was performed on 46 cases. At present, the age estimation error is approximately 1 year for children and adolescents.
 CONCLUSIONS: The experimental results have confirmed the adequacy of the age estimates obtained to the actual age of the individual and, consequently, the applicability of the proposed method in forensic medical institutions. The proposed method is currently implemented as a set of software components with subsequent manual integration of automatically calculated data. The plan was to supplement the database of computed tomography images to increase the training sample and the accuracy of age prediction.

THE METHOD OF IDENTIFYING THE TYPE OF UNMANNED OF THE AVIATION COMPLEX BY ITS RADIO SIGNALS

The article proposes a method for determining the type of unmanned aerial complex (UAC) based on its radio signals, which is based on a comparison of the estimated parameters of the UAV radio signals with known values stored in the database. The following characteristics and parameters of the signal are proposed as recognition features: the appearance of the amplitude-frequency spectrum (AFS) and spectrogram of the radio signal; radio line frequency range; accumulated ASF width; ASF width of the frequency element; spread spectrum parameters (hop duration; rate of frequency hopping spectrum spreading (FHSS); number of FHSS frequencies; FHSS frequency grid step); modulation parameters (symbol rate; modulation index); the position of the autocorrelation function (ACF) peaks; constant symbol sequences. Peculiarities of definition and application of the indicated recognition features are analyzed. It is shown that the most informative features are modulation parameters and constant symbol sequences. In order to increase the probability of correct recognition of UAC, in addition to the indicated features, it is suggested to use the power of received radio signals and the direction of the radio emission sources. A combination of two well-known approaches was used to make a decision about the type of UAC: the decision tree method and the minimum metric method. The application of the first made it possible to reduce the computational complexity due to the exclusion of operations for radio signals parameters evaluating that do not correspond to the decisions made at the previous stages. The minimum metric method was chosen taking into account the simplicity of its implementation and ensuring the possibility of expanding the list of known UAC radio signals without significantly changing the developed recognition method. The proposed method consists of six stages, in which recognition features corresponding to the signal spectral characteristics, spread spectrum parameters, modulation parameters, constant symbol sequences, power of the received radio signal and bearing on the UAV are determined in turn. The indicated features are compared with the known ones stored in the UAC radio signal database using the minimum metric method. In case of coincidence of the estimated and known information, a decision is made about the type of UAC, and the remaining stages are not performed. Verification of the developed method was carried out by applying it to the analysis of radio signals of known UAC, in particular, “Orlan-10”, “Supercam S350”, “Lancet”, “Eleron-3SV”.

Estimation psychophysiological state via nonlinear dynamic integral models

The method of experimental research “input-output”of the human oculo-motor system was developed and implemented using innovative eye-tracking technology for recording oculo-motor systemresponses to test visual stimuli. Stimuli are displayed on the monitor screen at different distances from the starting position. This formally corresponds to the action of step signals with different amplitudes at the input of the oculo-motor system. According to the empirical data of the “input-output”studies of the respondent's oculo-motor systemobtained with the aid of the TobiiProTX300 eye tracker, the transient functions of the first and diagonal intersections of the transient functions of the second and third orders of the oculo-motor systemwere determined. Experimental studies of the respondent's oculo-motor systemto identify the state of fatigue were carried out before the beginning (in the morning) and after the working day (in the evening). The obtained multidimensional transient functions are used as a source of primary data in the implementation of intelligent information technology for diagnosis and monitoring of the psychophysiological state of a person. Instrumental algorithmic and software tools for determining diagnostic features based on the identification data of the oculo-motor systemin the form of multidimensional transient functionsin the Python language have been developed.Training samples of data fortwo states of the respondent (“Normal”and “Fatigue”) were formed on the basis of the proposed heuristic features, which are determined using integral and differential transformations of the obtained multidimensional transient functionsof the oculo-motor system. Training samples of data are used to build classifiers of psychophysiological states of an individual using machine learning tools. The informativeness of individual features and all their possible combinations in pairs according to the indicator of the probability of correct recognition was studied using the method of complete search. The research results were obtained by evaluating the quality of recognition of states built by Bayesian classifiers in different spaces of the proposed features. An analysis of the stability of the correct recognitioninformativeness indicator of different feature spaces under the influence of different levels of additive noise on the features was carried out. Two-dimensional feature spaces with the maximum and most stable value of the correct recognitionindicator were found when solving the scientific and practical task of assessing the psychophysiological state (fatigue) of a person (0.9375). Thus, it seems appropriate to use the multidimensional transient functionsobtained from eye-tracking data in diagnostic studies in the fields of neuroscience and experimental psychology

Статистические характеристики сигнала на выходе оптимальной радиолокационной системы распознавания подвижных воздушных объектов

This work can be considered as a continuation of the article, or as separate materials devoted to the recognition of moving air objects based on the theory of testing statistical hypotheses. In the future, when presenting the materials of this article, the terminology proposed in [1] will be used. The relevance of this article is due to the events taking place on the European continent, indicating the need for further development of radar systems for various purposes and the transition from radar to radio vision, which will significantly improve the effectiveness of air defense and airspace control in general. The construction of a radar portrait and further automatic recognition of moving air objects, as an element of artificial intelligence, will eliminate human error, as well as significantly reduce the time it takes to make a decision on the necessary measures to influence the detected aircraft. Studying the freely available scientific works of various scientists devoted to such a direction of research and development of artificial intelligence as the statistical theory of pattern recognition, one can note the absence of such an important element as the function of the dependence of the probability of correct recognition on the quality and noise of the image. Within the framework of this article, a variant of solving this problem in the form of a mathematical model will be proposed. Since in [1] a recognition method is considered that uses the correlation coefficient between the current image and the reference one, and the probability of correct recognition is estimated using the Neumann-Pearson criterion, it is proposed to plot the dependence of the probability of correct recognition on the ratio of the image energy EI to the spectral noise density (R = 2EI/N0R), for a given false recognition probability . In the future, in the presented work, this dependence will be called recognition characteristics. Graphs will also be considered, in this paper referred to as the distribution density of the recognition probability.

METHOD OF DETERMINING PARAMETERS OF MODULATION OF UAV SIGNALS USING ARTIFICIAL NEURAL NETWORKS

The article is devoted to the consideration of the issue of determining the parameters of the modulation of signals of unmanned aerial vehicles using artificial neural networks by recognizing the types of digital modulation and is performed by a system that automatically classifies the type of digital modulation of the received signal. Recognition of digital modulation types is used, which automatically classifies the type of digital modulation of the received signal. The following issues are covered in the article: the analysis of existing approaches in the task of automatic recognition of types of digital modulation is carried out; the analysis and classification of informative features in the task of automatic recognition is carried out the following types of digital modulation: 2-PSK, 4-PSK, 8-PSK, 2-FSK, 8-QAM, 16-QAM, 64-QAM and OFDM. This article uses a neural network approach based on cumulative characteristics. To solve these problems, the methods of calculation and selection of informative cumulative features described in the developed model of the system of automatic recognition of digital modulation types by means of a neural network on cumulative features at a known value of carrier frequency are described. A model of the system of automatic recognition of digital modulation types at a known value is constructed carrier frequency. An algorithm for automatic recognition of digital modulation types has been developed using a multilayer neural network. The influence of noise in the communication channel on the probability of recognizing the types of digital modulation at a known value of the carrier frequency is investigated. It was found that regardless of the type of noise in the communication channel, the law of error distribution in IQ data becomes close to normal. This fact is one important argument for the use of cumulative features in the task of automatic recognition of types of digital modulation. Therefore, the task of automatic recognition of digital modulation types is quite relevant. Further research may be aimed at expanding the range of high-order cumulative features used, due to which it is possible to increase the probability of correct recognition of types of digital modulation, and solving the recognition problem at an unknown value of the frequency and initial phase of the carrier signal.

Diagnosis and monitoring of a person's psychophysiological state based on multidimensional transient functions of the oculo-motor system

An important scientific and practical problem has been solved, which consists in the creation of methods and tools for nonlinear dynamic identification "input-output" of the human oculomotor system (OMS). Identification is carried out based on Volterra models in the form of multidimensional transient functions (n-dimensional integrals from Volterra kernels) using innovative eye tracking technology. As test signals, visual stimuli are used, which are displayed on the monitor screen at different distances from the starting position, which formally corresponds to the action of step signals with different amplitudes at the OMS input. Experimental studies of the OMS of individuals were carried out using the Tobii Pro TX300 eye tracker. The transient functions of the first and diagonal intersections of the transient functions of the second and third order were determined based on the data of oculographic studies. The models differ from the known ones in that they provide the possibility of modeling the OMS in each interval of input signals beyond the radius of convergence of the Volterra series. The obtained multidimensional functions are used as a source of primary data in the implementation of intelligent information technology for the diagnosis and monitoring of human psychophysiological states. A set of heuristic features are proposed, which are determined using integral and differential transformations of the obtained transient functions of the OMS. The informativeness of individual signs and all their possible combinations in pairs according to the indicator of the probability of correct recognition was studied. The research results were obtained using the construction of Bayesian classifiers in different spaces of the proposed features. Figs.: 2. Refs.: 15 titles.
 Keywords: psychophysiological state, diagnosis, monitoring, oculo-motor system, identification, Volterra model, multidimensional transient functions, test visual stimuli, eye-tracking technology

BIOMETRIC AUTHENTICATION OF STUDENTS TO CONTROL THE LEARNING PROCESS IN ONLINE EDUCATION

This article considers the relevant problem of biometric authentication of students in higher educational institutions. The authors present the results of using a turnstile system with a face recognition terminal, with the ability to provide unique biometric data in real time. The study was conducted among students of the D. Serikbayev East Kazakhstan Technical University, Ust-Kamenogorsk, Kazakhstan. The article presents the results of studies of one of the biometric methods of personality recognition. In this method, the process of proving and verifying the identity of the person can be carried out through the presentation by the user of his biometric image. The processing results are sorted and compared with typical images from the database. With its positive decision, the developed software issues the results of biometric authentication of a person who presented himself in front of a digital scanner. The applied value of the results of the work lies in the possibility of using them in the field of education, and various industries to make a decision on providing access to information resources. In the course of the study, a technology was developed to provide biometric authentication processes for university students. Domestic and foreign scientists who have made a significant contribution to the development of methods for processing facial images are noted. A review of biometric methods of recognition is carried out, and tools for electronic authentication and modern information security systems are described. Factors that significantly affect the probability of correct recognition of students’ faces are determined. The analysis of ways to increase the probability of correct recognition of students by the image of the face is carried out.

Research of the efficiency of the automatic system of detection and recognition of objects on radar image using neural networks

The main factors influencing the efficiency of an automatic object recognition system in a radar image obtained with a synthetic aperture radar are considered A clustering algorithm of selecting objects on a radar image has been developed. The results of applying the radar image clustering algorithm, which makes it possible to increase the probability of correct recognition of objects, are presented. An analysis of the effectiveness of the application of the clustering algorithm depending on the threshold detection is given. The structure of the two-stage system of the algorithm of the automatic recognition of objects on a radar image has been developed and presented.

Building a classifier in the personality recognition system based on eye tracking data

Identification systems that use biometric characteristics to solve the problem of access to information systems are becoming more common. The article proposes a new method of biometric identification of computer systems users, based on the determination of the integral Volterra model of the human oculo-motor system (OMS) according to experimental research "input-output" using innovative eye tracking technology. With the help of the Tobii Pro TX300 eye tracker, the data of OMC responses to test visual stimuli were obtained, displayed as bright dots on the computer screen at different distances from the start position in the "horizontal" direction. Based on the data obtained, the transition functions of the first, second and third orders of the OMS for two people were determined. To construct a personality classifier, the informativeness of the proposed heuristic features, determined on the basis of the transition functions in terms of the probability of correct recognition (PCR), is investigated. Pairs of features are established that are resistant to computational errors and have a high PCR value - in the range 0.92 - 0.97. Fig.: 8. Table: 5. Bibliography: 30 items.
 Key words: biometric identification, personality recognition, Volterra model, oculo-motor system, eye tracking technology, informativeness of features, classification.

МАТЕМАТИЧНА МОДЕЛЬ ФУНКЦІОНУВАННЯ ПІДСИСТЕМИ РОЗПІЗНАВАННЯ СИСТЕМИ РОЗВІДКИ

The intelligence system is a multilevel recognition system. One of the top priorities of the intelligence system for intelligence objects is the task of recognizing them. For example, in an electronic reconnaissance system, the type of radio and radar station is a sign of determining the operational and tactical affiliation of a radio network or radio system, while determining the type of radio and radio station is one of the tasks of recognition. The recognition process is defined as one of the types of classification. The principle of recognition as a process of information processing is the division of information into a posteriori and a priori. Recognition is the process of comparing the obtained intelligence data on a certain intelligence object (a posteriori information) with the data, which consists in the description of all intelligence objects (a priori information) with the help of intelligence features, to determine the affiliation of the intelligence object. A posteriori data on the object of reconnaissance is compared with a priori information using recognition algorithms. Recognition is used to establish the operational and tactical affiliation of individual electronic means, networks, communication nodes, command posts, as well as to take into account the impact of enemy resistance. Some processes can be both processing tasks, such as systematization of information, data, features. At the moment, in a number of works, attempts are being made to expand the feature space and improve the scientific and methodological apparatus of recognition. These works are aimed at improving system performance, in particular - the probability of correct recognition. In order to determine how much this value has changed, depending on various factors, methods of mathematical modeling of the system, as well as laboratory and field experiments are used. The use of the latter is associated with certain difficulties, especially when it comes to a complex dynamic system of intelligence on the activities of foreign states in space, the means of measurement of which have a wide geography of location and differ in various features of operation. To assess the main indicators of the recognition system and ensure the possibility of optimal choice of means of measuring the parameters of the reconnaissance objects, it is proposed to develop a model of system operation. A mathematical model of the functioning of the recognition system is proposed. Application of the considered model will allow to solve the problems connected with construction of systems of recognition and definition of necessary quantity of signs.

STUDY NEUR NETWORKS FOR SOFTWARE DEFINED RADIO CONTROL

The scientific article is devoted to the issues of SDR system control. Software Defined Radio is a system designed for software control of information transmission processes in a radio communication channel. Recognition of digital modulation types is used, which automatically classifies the type of digital modulation of the received signal. The following issues are covered in the article: the analysis of existing approaches in the task of automatic recognition of types of digital modulation is carried out; the analysis and classification of informative features in the task of automatic recognition is carried out the following types of digital modulation: 2-PSK, 4-PSK, 8-PSK, 2-FSK, 8-QAM, 16-QAM, 64-QAM and OFDM. This article uses a neural network approach based on cumulative characteristics. To solve these problems, the methods of calculation and selection of informative cumulative features described in the developed model of the system of automatic recognition of digital modulation types by means of a neural network on cumulative features at a known value of carrier frequency are described. A model of the system of automatic recognition of digital modulation types at a known value is constructed carrier frequency. An algorithm for automatic recognition of digital modulation types has been developed using a multilayer neural network. The influence of noise in the communication channel on the probability of recognizing the types of digital modulation at a known value of the carrier frequency is investigated. It was found that regardless of the type of noise in the communication channel, the law of error distribution in IQ data becomes close to normal. This fact is one important argument for the use of cumulative features in the task of automatic recognition of types of digital modulation. Therefore, the task of automatic recognition of digital modulation types is quite relevant. Further research may be aimed at expanding the range of high-order cumulative features used, due to which it is possible to increase the probability of correct recognition of types of digital modulation, and solving the recognition problem at an unknown value of the frequency and initial phase of the carrier signal.

Face recognition using multitasking cascading convolutional networks

The face recognition has many applications in such areas like intelligent security and access control, biometrics, safeguard, verification, attendance accounting, machine vision, etc. Identification of a personality by the face has a number of advantages over other methods: the person does not need to be contacted physically, which is the most acceptable way for mass applications and no complicated or expensive equipment is required. This article discusses the problem of recognition and identification of a person’s face using convolutional neural networks that process frames from a camera in real time or from a recorded video file with the subsequent entry of the identified person into the database. Multitasking Cascade Convolutional Neural Network (MTCNN) has three convolutional networks (P-Net, R-Net, and O-Net) and is capable of outperforming many face detection tests while maintaining real-time performance. The proposed method for human face recognition was developed as a software product, tested and showed the probability of correct recognition in real time 96.02%.

Research of recognition accuracy of dangerous and safe x-ray baggage images using neural network transfer learning

The article considers the use of neural networks to solve the problem of recognizing dangerous and safe objects carried in the luggage of airport passengers. A comparative analysis is performed to define the accuracy achieved on the test sample for different convolutional neural networks. It also explores the influence of various regularizations on the accuracy of a two-class classification. The increased probability of correct recognition is achieved due to augmentation, reset weights and saturation of the network. The method of transfer training is used to increase the efficiency of the recognizer. In this case, a study was carried out for the transfer of various neural networks.

РАСПОЗНАВАНИЕ РЕЧЕВЫХ СООБЩЕНИЙ РАДИООБМЕНА В АВИАЦИИ НА БАЗЕ КОРРЕЛЯЦИОННОГО АНАЛИЗА

The paper considers the problem of speech messages recognition in phraseological radio exchange for tasks of civil aviation. The introduction substantiates the relevance of this problem. The following are research methods based on correlation analysis. Finally, a description of the experiment and the results of the recognition algorithms based on correlation analysis are given. Various variants were recorded for five speech messages and spectral representations of such signals were constructed. Spectral transform can be obtained either using specialized software or based on the Fourier transform of the signal in the time domain. To obtain a more universal reference signal and eliminate the influence of interference, the spectral components of the same speech message recorded several times were averaged. In fact, three spectra of the same speech message were used for averaging. This spectrum averaging over three training components provided a reference sample of phrases or patterns for each phrase, and reduced the influence of additive white Gaussian noise in the reference. Later, on the basis of correlation analysis, the connections between test phrases and all patterns were calculated. On the basis of these connections, a correlation matrix of reference phrases is built. Research has shown that phrases spoken by one person were highly correlated. The analysis showed that the choice of the class (the content of the speech message) when solving the recognition problem corresponding to the value of the correlation coefficient closest to one provides over 90% of correct recognitions on a test sample containing a total of 100 phrases, 20 for each phrase. It should be noted that, when recording test messages, an additive white Gaussian noise was additionally present as a background, reproduced by another audio device. In the case of information analysis without artificially generated noise, the probability of correct recognition for a test sample of 100 phrases, 20 for each phrase, is 100% when using correlation analysis.

Обучение многослойного перцептрона с учителем в задаче распознавания с помощью корреляционного показателя

The article deals with the problem of recognition of handwritten digits using feedforward neural networks (perceptrons) using a correlation indicator. The proposed method is based on the mathematical model of the neural network as an oscillatory system similar to the information transmission system. The article uses theoretical developments of the authors to search for the global extremum of the error function in artificial neural networks. The handwritten digit image is considered as a one-dimensional input discrete signal representing a combination of "perfect digit writing" and noise, which describes the deviation of the input implementation from "perfect writing". The ideal observer criterion (Kotelnikov criterion), which is widely used in information transmission systems and describes the probability of correct recognition of the input signal, is used to form the loss function. In the article is carried out a comparative analysis of the convergence of learning and experimentally obtained sequences on the basis of the correlation indicator and widely used in the tasks of classification of the function CrossEntropyLoss with the use of the optimizer and without it. Based on the experiments carried out, it is concluded that the proposed correlation indicator has an advantage of 2-3 times.

Improvement of the Signature-System Method by Applying Priority Features in the Composition of the Monitored Object Signatures to Ensure Maximum Probability of Correct Recognition

An approach to improving the signature-system method of obtaining and processing information by prioritizing monitoring features in the source (object) signatures with determining their rational number is proposed. The application of the bi-Gaussian mathematical model of radio source signals and priority monitoring features as information support for the signature-system method allows increasing the efficiency of its use when the initial data are partially or completely uncertain.

Identification and parameter estimation algorithm of radar signal subtle features

With the rapid development of electronic technology and its widespread application in modern warfare, the modulation methods of individual radar signal sources are more flexible, the parameters are diverse, and the modern battlefield electromagnetic signal environment is complex, so that radar signal recognition based on traditional conventional parameters Technology cannot meet real needs. Among the many radar signal modulation recognition algorithms that have appeared in recent years, neural network theory and intra-pulse analysis algorithms are widely used, but they each have their own shortcomings, and a single method cannot meet the needs of increasingly complex modulation signal recognition and parameter estimation needs. . Based on the above background, the research content of this paper is to identify the subtle characteristics of radar signals and their parameter estimation algorithms. In this paper, two types of typical modulation signals in intentional modulation of radar signals, namely phase-coded (PSK) signals and frequency-modulated (FM) signals, are proposed. A SVEFD algorithm based on classification from coarse to fine. According to the characteristics of the 3 dB bandwidth of the frequency spectrum of the PSK signal and the FM signal, the coarse classification between the classes is performed first. Finally, through experimental simulations, the results show that the correct recognition probability of the SVEFD algorithm at a lower signal-to-noise ratio is much higher than the m algorithm and also higher than the SVD algorithm. When the signal-to-noise ratio is greater than 1 dB, the average recognition rate reaches 94%, which proves that The effectiveness of the proposed SVEFD algorithm. When the signal-to-noise ratio is .1 dB, in addition to maintaining the correct recognition rate of more than 90% for the LFM and COSTAS signals, the other six types of signals have low correct recognition rates, so the SVEFD algorithm has advantages in comparison.

Structurally topological algorithm for star recognition and near-Earth space’ object detection

When solving a variety of celestial navigation tasks there is a problem of determining parameters of spacecraft motion and onboard primary payload orientation based on the coordinates of registered star images. Furthermore, unwanted objects, like active satellites, natural and artificial space debris, that reduce the probability of correct recognition may get into the field of view of a satellite sensor. This prompts the necessity to filter out such interference from the star field images. However, if the objects under recognition are bodies located in near-Earth space, in this case, the star images themselves will act as interferences. In addition, since the detection and cataloging of these objects from the Earth’s surface is complicated by their small size, the atmospheric effects, as well as other technical difficulties, it is worthwhile to use the existing equipment onboard spacecrafts to solve this task. The existing recognition algorithms for star groups, as well as their classification, are presented in this paper. Moreover, a structurally topological approach for identifying groups of stars based on the properties of enveloping polygons used in constructing topological star patterns is proposed. Specific features in the construction of topological configurations on the analyzed set of points, as well as the principles of dynamic space object detection within their limits are described. Results of the numerical experiments performed using the developed algorithm on the star field maps and model scenes are presented.

Удосконалена методика раціонального розподілу інформації за важливістю і кількістю джерел радіомоніторингу в АСУ військами та зброєю

When designing automated systems for command and control of troops and weapons, an important role is given to the form and volume of information transmitted information. Information in the control loop comes from sensors that make up the monitoring system and continuously monitor the sources and objects of monitoring. The control and communication system of a group object of radio monitoring is a complex hierarchical structure. The number of sources that operate in the interests of providing a control system exceeds the number of available forces and means that carry out radio monitoring. There is a discrepancy in the recognition of objects and sources of radio monitoring: identification of the object of observation, the number of signs. Therefore, it is necessary to solve the multicriteria optimization problem for identifying the operating mode of radio monitoring sources for guaranteed determination of the state of the control system and communication of the group object of radio monitoring, which consists in choosing those sources whose utility is greatest. The proposed methodology for the rational distribution of information in automated control systems by troops and weapons, by selecting the most important sources of radio monitoring and rationalizing the number of gradations of radio monitoring objects to ensure a given probability of correct recognition and guaranteed determination of their condition.

Classification of surface water objects in visible spectrum images

This paper considers the problem of classifying surface water objects, e.g. ships of different classes, in visible spectrum images using convolutional neural networks. A technique for forming a database of images of surface water objects and a special training dataset for creating a classification are presented. A method for forming and training of a convolutional neural network is described. The dependence of the probability of correct recognition on the number and variants of the selection of specific classes of surface water objects is analysed. The results of recognizing different sets of classes are presented.