Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139190
A. Sokolova, A. Savchenko
In this paper, we explore the possibility to improve efficiency of face recognition using information about anomaly input images. Indeed, modern publicly-available datasets typically contain images of mostly middle-aged and Caucasian people, which cause most algorithms to fail on photos of older people or children, rarer ethnicities, poor-quality images, etc. Detection of such anomaly data and its subsequent rejection helps to improve the classification accuracy. We propose a novel algorithm, at the first stage of which a convolutional neural network is used to detect anomalies in input images. This network is trained on a specially created set of rare data. The second stage is the sequential analysis of neural descriptors extracted from input faces to improve the computational efficiency of classification. An experimental study on the VGGFace2 and MS-Celeb-1M datasets using neural network descriptors, including contemporary InsightFace models, demonstrated the effectiveness of the proposed algorithm.
{"title":"Effective face recognition based on anomaly image detection and sequential analysis of neural descriptors","authors":"A. Sokolova, A. Savchenko","doi":"10.1109/ITNT57377.2023.10139190","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139190","url":null,"abstract":"In this paper, we explore the possibility to improve efficiency of face recognition using information about anomaly input images. Indeed, modern publicly-available datasets typically contain images of mostly middle-aged and Caucasian people, which cause most algorithms to fail on photos of older people or children, rarer ethnicities, poor-quality images, etc. Detection of such anomaly data and its subsequent rejection helps to improve the classification accuracy. We propose a novel algorithm, at the first stage of which a convolutional neural network is used to detect anomalies in input images. This network is trained on a specially created set of rare data. The second stage is the sequential analysis of neural descriptors extracted from input faces to improve the computational efficiency of classification. An experimental study on the VGGFace2 and MS-Celeb-1M datasets using neural network descriptors, including contemporary InsightFace models, demonstrated the effectiveness of the proposed algorithm.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"200 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121072235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139151
R. Aleshko, V. Berezovsky, K. Shoshina, Alexander Gordeychik, V. Krekhalev, Alexander Ponomarev
The paper shows that high solar activity and the associated increased geomagnetic disturbance have a significant impact on the functioning of small spacecraft (SSC) in low Earth orbit (LEO). An increase in the number of anomalies in the telemetry data of small satellites with an increase in the absolute value of the geomagnetic activity index Dst is studied. The influence exerted by space weather (SW) factors on the current state of the SSC systems is also studied. An increasing dependence between the spacecraft parameters and changes in geomagnetic activity is shown, which intensifies in the subpolar regions of the orbit.
{"title":"Analysis of the Influence of Space Weather Factors on the Telemetry Parameters of Small Spacecraft in Low Earth Orbit","authors":"R. Aleshko, V. Berezovsky, K. Shoshina, Alexander Gordeychik, V. Krekhalev, Alexander Ponomarev","doi":"10.1109/ITNT57377.2023.10139151","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139151","url":null,"abstract":"The paper shows that high solar activity and the associated increased geomagnetic disturbance have a significant impact on the functioning of small spacecraft (SSC) in low Earth orbit (LEO). An increase in the number of anomalies in the telemetry data of small satellites with an increase in the absolute value of the geomagnetic activity index Dst is studied. The influence exerted by space weather (SW) factors on the current state of the SSC systems is also studied. An increasing dependence between the spacecraft parameters and changes in geomagnetic activity is shown, which intensifies in the subpolar regions of the orbit.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114257563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139310
O. Golovnin, Irina Y. Dubinina, A. Ivaschenko, A. Krivosheev, P. Sitnikov
The paper proposes an approach to identifying problematic issues according to open sources for socially oriented topics. The approach involves the use of texts and metadata of publications in open thematic groups and on public pages of users in social media. A multi-stage processing of publications is performed: data collection, spam removal, filtering by problem area, sentiment analysis, classification by topic, and clustering. The software implementation of the proposed approach is based on the Integrated Monitoring Digital Platform. Experimental testing was carried out on the data of the VK social media. The application of the approach makes it possible to identify acute social problematic issues of the administrative region in the operational mode.
{"title":"Social Climate Analysis based on Open Data","authors":"O. Golovnin, Irina Y. Dubinina, A. Ivaschenko, A. Krivosheev, P. Sitnikov","doi":"10.1109/ITNT57377.2023.10139310","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139310","url":null,"abstract":"The paper proposes an approach to identifying problematic issues according to open sources for socially oriented topics. The approach involves the use of texts and metadata of publications in open thematic groups and on public pages of users in social media. A multi-stage processing of publications is performed: data collection, spam removal, filtering by problem area, sentiment analysis, classification by topic, and clustering. The software implementation of the proposed approach is based on the Integrated Monitoring Digital Platform. Experimental testing was carried out on the data of the VK social media. The application of the approach makes it possible to identify acute social problematic issues of the administrative region in the operational mode.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"71 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130738545","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139185
A.E. Rusetskaya
In this paper, a comparative study of the effect of high-order polynomial phase apodization of an optical system on increasing the depth of focus is carried out. On the basis of numerical modeling, the dependencies of the conservation of the point scattering function on the parameters of the apodizing function during defocusing are constructed.
{"title":"Comparative study of power-law apodizing functions when encoding the wavefront in order to increase the depth of focus","authors":"A.E. Rusetskaya","doi":"10.1109/ITNT57377.2023.10139185","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139185","url":null,"abstract":"In this paper, a comparative study of the effect of high-order polynomial phase apodization of an optical system on increasing the depth of focus is carried out. On the basis of numerical modeling, the dependencies of the conservation of the point scattering function on the parameters of the apodizing function during defocusing are constructed.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125600075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139122
Daniil Kozlov, V. Myasnikov
The article proposes an ensemble method for reinforcement learning algorithms. The proposed approach is on average more efficient than each of the algorithms in the ensemble separately. The article discusses the implementation of the method, which includes an ensemble of REDQ and SAC algorithms. The output from the ensemble is the output of the algorithm selected following the output of the DQN acting as the control algorithm. It is possible to ensemble other algorithms in a different quantity. Reinforcement learning is a promising area in machine learning. An important unsolved problem of reinforcement learning is the generalization of complex problems, and their solution using meta-algorithms. The proposed method can be used in complex problems consisting of many subtasks, effective solutions for which can be offered by various algorithms from the ensemble.
{"title":"Ensemble Method for Reinforcement Learning Algorithms Based on Hierarchy","authors":"Daniil Kozlov, V. Myasnikov","doi":"10.1109/ITNT57377.2023.10139122","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139122","url":null,"abstract":"The article proposes an ensemble method for reinforcement learning algorithms. The proposed approach is on average more efficient than each of the algorithms in the ensemble separately. The article discusses the implementation of the method, which includes an ensemble of REDQ and SAC algorithms. The output from the ensemble is the output of the algorithm selected following the output of the DQN acting as the control algorithm. It is possible to ensemble other algorithms in a different quantity. Reinforcement learning is a promising area in machine learning. An important unsolved problem of reinforcement learning is the generalization of complex problems, and their solution using meta-algorithms. The proposed method can be used in complex problems consisting of many subtasks, effective solutions for which can be offered by various algorithms from the ensemble.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126542152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139147
V. Sokolov
In this article I summarize the results of thirty-five years of work of the Image Processing Systems Institute of the Russian Academy of Sciences. I describe the scientific achievements of the team and the changes that have taken place over the past five years since my previous article. I analyze the organizational and scientific success of the team and the effectiveness of the decisions taken by the management and staff of the Institute in overcoming the problems faced by Russian scientists.
{"title":"Image Processing Systems Institute of the RAS: Responses to Current Challenges","authors":"V. Sokolov","doi":"10.1109/ITNT57377.2023.10139147","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139147","url":null,"abstract":"In this article I summarize the results of thirty-five years of work of the Image Processing Systems Institute of the Russian Academy of Sciences. I describe the scientific achievements of the team and the changes that have taken place over the past five years since my previous article. I analyze the organizational and scientific success of the team and the effectiveness of the decisions taken by the management and staff of the Institute in overcoming the problems faced by Russian scientists.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117347434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139230
V. Sergeev, I. Frolov, O. Radaev
An analysis and estimation of measuring errors the 3dB frequency of the LED electroluminescence by the standard method at low excitation levels are presented. It is noted that when measuring the variable component of the LED emission power at low currents (up to 100 μA), errors predominate due to the nonlinearity of the LED watt-ampere characteristic and the influence of additive noise of the measuring circuits. It is shown that there is a value of the alternating current amplitude at which the total measurement error of the optical signal at low currents takes on a minimum value.
{"title":"Measurement of the LED electroluminescence 3dB frequency at low currents","authors":"V. Sergeev, I. Frolov, O. Radaev","doi":"10.1109/ITNT57377.2023.10139230","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139230","url":null,"abstract":"An analysis and estimation of measuring errors the 3dB frequency of the LED electroluminescence by the standard method at low excitation levels are presented. It is noted that when measuring the variable component of the LED emission power at low currents (up to 100 μA), errors predominate due to the nonlinearity of the LED watt-ampere characteristic and the influence of additive noise of the measuring circuits. It is shown that there is a value of the alternating current amplitude at which the total measurement error of the optical signal at low currents takes on a minimum value.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123871186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10138953
D. Asatryan, Mariam Haroutunian, Grigor Sazhumyan, A. Kupriyanov, R. Paringer, D. Kirsh
This article is devoted to the study of global binarization methods based on the optimization of various quality criteria for the results obtained. Two types of quality criteria are considered, based on the properties of the histogram of the tested image and evaluating the similarity of the latter with the binarized image. The first type includes the criteria used in the well-known OTSU method, the PUN entropy test, as well as the new CHP test. The second type includes the SSIM image similarity test, the W2 gradient test, and the correlated test COR. It is substantiated that for a fixed quality criterion, the binarization procedure is reduced only to searching for a threshold from a certain set of values that maximizes this criterion. It is shown that the criterion PSNR of the closeness of images is unsuitable in this procedure; because the optimal threshold value obtained by this criterion does not depend on the tested image and equal to 127.
{"title":"Comparative Quality Analysis of Image Global Binarization Procedures","authors":"D. Asatryan, Mariam Haroutunian, Grigor Sazhumyan, A. Kupriyanov, R. Paringer, D. Kirsh","doi":"10.1109/ITNT57377.2023.10138953","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10138953","url":null,"abstract":"This article is devoted to the study of global binarization methods based on the optimization of various quality criteria for the results obtained. Two types of quality criteria are considered, based on the properties of the histogram of the tested image and evaluating the similarity of the latter with the binarized image. The first type includes the criteria used in the well-known OTSU method, the PUN entropy test, as well as the new CHP test. The second type includes the SSIM image similarity test, the W2 gradient test, and the correlated test COR. It is substantiated that for a fixed quality criterion, the binarization procedure is reduced only to searching for a threshold from a certain set of values that maximizes this criterion. It is shown that the criterion PSNR of the closeness of images is unsuitable in this procedure; because the optimal threshold value obtained by this criterion does not depend on the tested image and equal to 127.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122666383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10138949
V. Pechenin, E. Pechenina, M. Bolotov
A model has been developed to predict the angular reversals of the blades in the impeller that occur during assembly. The calculated angles characterize the tightness in the connections of the blade end flanges. The model inputs data on geometry deviations from part inspection operations. The model uses the random forest method, and the model was trained on a set of numerical experiments performed in the ANSYS environment. According to experimental findings, the ANSYS model's error does not exceed 10 angular minutes, and the regression model's overall error does not exceed 20 angular minutes.
{"title":"Predictive model for compressor impeller tightness","authors":"V. Pechenin, E. Pechenina, M. Bolotov","doi":"10.1109/ITNT57377.2023.10138949","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10138949","url":null,"abstract":"A model has been developed to predict the angular reversals of the blades in the impeller that occur during assembly. The calculated angles characterize the tightness in the connections of the blade end flanges. The model inputs data on geometry deviations from part inspection operations. The model uses the random forest method, and the model was trained on a set of numerical experiments performed in the ANSYS environment. According to experimental findings, the ANSYS model's error does not exceed 10 angular minutes, and the regression model's overall error does not exceed 20 angular minutes.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122678595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-17DOI: 10.1109/ITNT57377.2023.10139047
M. A. Chubar, M. Gashnikov
The structure of existing neural network video compression methods in most cases includes predictive encoding, which uses a subtraction operation between the predicted and current frames to remove redundancy. To increase efficiency, an approach based on deep contextual video compression is used. In addition to the difference frame, this approach relies heavily on specialized algorithms for extracting additional information characterizing the difference of closely spaced frames. The use of context in this case makes it possible to achieve a better quality of reconstruction of video sequences, in particular for complex textures with a large number of high frequencies. This implies that the proposed method can potentially lead to significant savings in storage and transmission costs while maintaining high-quality video output. This article presents the results of computational experiments to evaluate the effectiveness of the investigated method of deep contextual video compression on real video sequences. Experimental findings demonstrate the advantages of the considered technique in PSNRR/bpp coordinates when compared to the performance of three common video codecs: H.264, H.265, and VP9.
{"title":"Deep Contextual Video Compression Based on Machine Learning","authors":"M. A. Chubar, M. Gashnikov","doi":"10.1109/ITNT57377.2023.10139047","DOIUrl":"https://doi.org/10.1109/ITNT57377.2023.10139047","url":null,"abstract":"The structure of existing neural network video compression methods in most cases includes predictive encoding, which uses a subtraction operation between the predicted and current frames to remove redundancy. To increase efficiency, an approach based on deep contextual video compression is used. In addition to the difference frame, this approach relies heavily on specialized algorithms for extracting additional information characterizing the difference of closely spaced frames. The use of context in this case makes it possible to achieve a better quality of reconstruction of video sequences, in particular for complex textures with a large number of high frequencies. This implies that the proposed method can potentially lead to significant savings in storage and transmission costs while maintaining high-quality video output. This article presents the results of computational experiments to evaluate the effectiveness of the investigated method of deep contextual video compression on real video sequences. Experimental findings demonstrate the advantages of the considered technique in PSNRR/bpp coordinates when compared to the performance of three common video codecs: H.264, H.265, and VP9.","PeriodicalId":296438,"journal":{"name":"2023 IX International Conference on Information Technology and Nanotechnology (ITNT)","volume":"111 3S 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116361838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：