Pub Date : 2024-04-10DOI: 10.1134/s1054661824010176
Hasmik Sahakyan
Abstract
In this paper we consider discrete tomography problems with an additional requirement of non-repeatability of rows of the binary matrix to be reconstructed; as well as discrete tomography problems with given pairwise projections. Representing the problems in the hypergraph model and pointing out their equivalence to the basic definitions, we state the following results: (i) nonconvexity of the set of hypergraphic sequences of simple hypergraphs with (n) vertices and (m) hyperedges in the (n)-dimensional (m + 1)-valued lattice, (ii) characterization of monotone Boolean functions associated with degree sequences of 3-/(n–3)-uniform hypergraphs, (iii) formulation of discrete tomography problems with paired projections, their connection to hypergraph degree sequence problem with generalized degrees, a solution for a particular case.
{"title":"Some Discrete Tomography Problems in Hypergraph Model Interpretation","authors":"Hasmik Sahakyan","doi":"10.1134/s1054661824010176","DOIUrl":"https://doi.org/10.1134/s1054661824010176","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>In this paper we consider discrete tomography problems with an additional requirement of non-repeatability of rows of the binary matrix to be reconstructed; as well as discrete tomography problems with given pairwise projections. Representing the problems in the hypergraph model and pointing out their equivalence to the basic definitions, we state the following results: (i) nonconvexity of the set of hypergraphic sequences of simple hypergraphs with <span>(n)</span> vertices and <span>(m)</span> hyperedges in the <span>(n)</span>-dimensional <span>(m + 1)</span>-valued lattice, (ii) characterization of monotone Boolean functions associated with degree sequences of 3-/(<i>n</i>–3)-uniform hypergraphs, (iii) formulation of discrete tomography problems with paired projections, their connection to hypergraph degree sequence problem with generalized degrees, a solution for a particular case.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"70 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590651","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 : 2024-04-10DOI: 10.1134/s1054661824010097
Iskandar Karapetyan, Karen Karapetyan
Abstract
A cap in a projective or affine geometry over a finite field is a set of points no three of which are collinear. We give several new constructions for complete caps in affine geometry (AGleft( {n,3} right)) over the field ({{F}_{3}} = left{ {0,1,2} right}) implying some well-known results.
{"title":"Complete Caps in Affine Geometry AG(n, 3)","authors":"Iskandar Karapetyan, Karen Karapetyan","doi":"10.1134/s1054661824010097","DOIUrl":"https://doi.org/10.1134/s1054661824010097","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>A cap in a projective or affine geometry over a finite field is a set of points no three of which are collinear. We give several new constructions for complete caps in affine geometry <span>(AGleft( {n,3} right))</span> over the field <span>({{F}_{3}} = left{ {0,1,2} right})</span> implying some well-known results.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"117 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590706","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}
Hyperspectral image classification has become an important issue in remote sensing due to the significant amount of spectral information in HSI. The costly and time-consuming annotation task of HSIs makes the number of labeled samples is limited. To address the above problem, we propose an enhanced semi-supervised support vector machine algorithm for spectral-spatial HSI classification. To fully capture the spectral and spatial information of HSI, we use local binary pattern to obtain spatial feature. The captured spatial features are concatenated with the spectral features to yield the hybrid spectral-spatial features. Self-training mechanism is then adopted to gradually select confident unlabeled samples with their pseudo-labels and add them to the labeled set. To further improve the classification performance of the semi-supervised support vector machine, we choose a cuckoo search algorithm based on the chaotic catfish effect to find its optimal combination of parameters. The experimental results on two publicly available HSI datasets show that the proposed model achieves excellent classification accuracy for each category in hyperspectral images, and also has superior overall accuracy compared with other comparative algorithms. Adequate experiments and analysis illustrate the promising potential and prospect of our proposed model for HSI classification.
{"title":"An Enhanced Semi-Supervised Support Vector Machine Algorithm for Spectral-Spatial Hyperspectral Image Classification","authors":"Ziping He, Kewen Xia, Jiangnan Zhang, Sijie Wang, Zhixian Yin","doi":"10.1134/s1054661824010085","DOIUrl":"https://doi.org/10.1134/s1054661824010085","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>Hyperspectral image classification has become an important issue in remote sensing due to the significant amount of spectral information in HSI. The costly and time-consuming annotation task of HSIs makes the number of labeled samples is limited. To address the above problem, we propose an enhanced semi-supervised support vector machine algorithm for spectral-spatial HSI classification. To fully capture the spectral and spatial information of HSI, we use local binary pattern to obtain spatial feature. The captured spatial features are concatenated with the spectral features to yield the hybrid spectral-spatial features. Self-training mechanism is then adopted to gradually select confident unlabeled samples with their pseudo-labels and add them to the labeled set. To further improve the classification performance of the semi-supervised support vector machine, we choose a cuckoo search algorithm based on the chaotic catfish effect to find its optimal combination of parameters. The experimental results on two publicly available HSI datasets show that the proposed model achieves excellent classification accuracy for each category in hyperspectral images, and also has superior overall accuracy compared with other comparative algorithms. Adequate experiments and analysis illustrate the promising potential and prospect of our proposed model for HSI classification.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"18 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590542","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 : 2024-04-10DOI: 10.1134/s1054661824010206
Hakob A. Tamazyan, Anahit A. Chubaryan
Abstract
A determinative sequent system DS for the classical propositional calculus is introduced on the base of well-known Tseitin’s transformation. It is proved that the system DS is polynomially equivalent to the propositional resolution system R and propositional cut-free sequent system PK–. Then we define the system SDS (DS with a substitution rule) and the systems SkDS (DS with restricted substitution rules, where the number of connectives in substituted formulas is bounded by k). It is proved that for every k ≥ 0 the system Sk+1DS has an exponential speed-up over the system SkDS in the tree form, and the system SDS is polynomially equivalent to the Frege systems.
{"title":"A Hierarchy of Determinative Sequent Systems with Different Substitution Rules","authors":"Hakob A. Tamazyan, Anahit A. Chubaryan","doi":"10.1134/s1054661824010206","DOIUrl":"https://doi.org/10.1134/s1054661824010206","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>A determinative sequent system DS for the classical propositional calculus is introduced on the base of well-known Tseitin’s transformation. It is proved that the system DS is polynomially equivalent to the propositional resolution system R and propositional cut-free sequent system PK<sup>–</sup>. Then we define the system SDS (DS with a substitution rule) and the systems S<sub><i>k</i></sub>DS (DS with restricted substitution rules, where the number of connectives in substituted formulas is bounded by <i>k</i>). It is proved that for every <i>k</i> ≥ 0 the system S<sub><i>k</i>+1</sub>DS has an exponential speed-up over the system S<sub><i>k</i></sub>DS in the tree form, and the system SDS is polynomially equivalent to the Frege systems.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"39 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590543","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 : 2024-04-10DOI: 10.1134/s1054661824010218
Sergo Tsiramua, Hamlet Meladze, Tinatin Davitashvili
Abstract
Multifunctional elements are a special class of elements, the reliability model of which differs from the classical two-pole “on-off” model. A multifunctional element can have partly false states in addition to nonfalse and false states. The multifunctionality of the elements leads to the formation of flexible, adaptable systems with reconfigurable structure, in which, in case of partial failure of the element, it is possible to continue the successful functioning of the system by redistributing the functions between the elements. In this paper, the properties of multifunctional elements (MFE) and systems, assembled on their basis, methods of structural analysis, logical-probabilistic reliability models and issues of optimal reconfiguration of systems based on MFE are discussed.
{"title":"Logical-Probabilistic Modeling and Structural Analysis of Reconfigurable Systems Composed of Multifunctional Elements","authors":"Sergo Tsiramua, Hamlet Meladze, Tinatin Davitashvili","doi":"10.1134/s1054661824010218","DOIUrl":"https://doi.org/10.1134/s1054661824010218","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>Multifunctional elements are a special class of elements, the reliability model of which differs from the classical two-pole “on-off” model. A multifunctional element can have partly false states in addition to nonfalse and false states. The multifunctionality of the elements leads to the formation of flexible, adaptable systems with reconfigurable structure, in which, in case of partial failure of the element, it is possible to continue the successful functioning of the system by redistributing the functions between the elements. In this paper, the properties of multifunctional elements (MFE) and systems, assembled on their basis, methods of structural analysis, logical-probabilistic reliability models and issues of optimal reconfiguration of systems based on MFE are discussed.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"263 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590535","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 : 2024-04-10DOI: 10.1134/s1054661824010073
Mariam Haroutunian
Abstract
Investigation of communication over a wiretap channel is one of the problems of information–theoretic security. The aim in the general wiretap channel model is to maximize the rate of the reliable communication from the source to the legitimate receiver, while keeping the confidential information as secret as possible from the wiretapper (eavesdropper). Here we consider the compound wiretap channel model, when the channels to the legitimate receiver and to the wiretapper depends on the number of possible states. We investigate the (E)-capacity–equivocation region which is the closure of the set of all achievable rate-reliability and equivocation pairs, where the rate-reliability function represents the optimal dependence of rate on the error probability exponent (reliability). Here the outer bound of this region is constructed in the case, when the states are not known to all terminals.
{"title":"Outer Bound for E-Capacity–Equivocation Region of Compound Wiretap Channel","authors":"Mariam Haroutunian","doi":"10.1134/s1054661824010073","DOIUrl":"https://doi.org/10.1134/s1054661824010073","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>Investigation of communication over a wiretap channel is one of the problems of information–theoretic security. The aim in the general wiretap channel model is to maximize the rate of the reliable communication from the source to the legitimate receiver, while keeping the confidential information as secret as possible from the wiretapper (eavesdropper). Here we consider the compound wiretap channel model, when the channels to the legitimate receiver and to the wiretapper depends on the number of possible states. We investigate the <span>(E)</span>-capacity–equivocation region which is the closure of the set of all achievable rate-reliability and equivocation pairs, where the rate-reliability function represents the optimal dependence of rate on the error probability exponent (reliability). Here the outer bound of this region is constructed in the case, when the states are not known to all terminals.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"77 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140602804","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 : 2024-04-10DOI: 10.1134/s1054661824010036
M. Aslanyan
Abstract
Human pose estimation (PE, tracking body pose on-the-go) is a computer vision-based technology that identifies and controls specific points on the human body. These points represent our joints and special points over the body determining the sizes, distances, angle of flexion, and type of the motion. Knowing this in a specific exercise is the basis of work for rehabilitation and physiotherapy, fitness and self-coaching, augmented reality, animation and gaming, robot management, surveillance and human activity analysis. Implementing such capabilities may use special suits or sensor arrays to achieve the best result, but massive use of PE is related to devices that many users own: namely smartphones, smartwatches, and earbuds. The body pose estimation system starts with capturing the initial data. In dealing with motion detection, it is necessary to analyze a sequence of images rather than a still photo. Different software modules are responsible for tracking 2D key points, creating a body representation, and converting it into a 3D space. Action recognition on the other hand is a way to analyze the sequence of estimated pose data with the aim to categorize sequence under the classes. It is widely used various fields. One of the widely known use cases is analysis and detection of potential attacks of illegal action using video from the surveillance cameras. Another use case involves analysis of the sequence of pose with the aim of creating a virtual coaching environment. Specifically, our research will target this challenging issue and aim to create this environment for mobile devices. We will describe some of the solutions that are suitable for effectively pose estimation and action recognition on mobile devices. We will show how lightweight models based on convolution neural networks can be used to efficiently solve pose estimation issue and address action recognition problem with the dynamic time warping algorithm.
摘要 人体姿态估计（PE，随身追踪人体姿态）是一种基于计算机视觉的技术，可识别和控制人体上的特定点。这些点代表我们的关节和身体上的特殊点，决定着运动的大小、距离、弯曲角度和类型。在特定运动中了解这些点是康复和理疗、健身和自我教练、增强现实、动画和游戏、机器人管理、监控和人体活动分析等工作的基础。要实现这些功能，可能需要使用特殊的服装或传感器阵列来达到最佳效果，但 PE 的大量使用与许多用户拥有的设备有关：即智能手机、智能手表和耳塞。身体姿态估计系统从捕捉初始数据开始。在处理运动检测时，有必要分析一系列图像而不是静态照片。不同的软件模块负责跟踪二维关键点、创建身体表征并将其转换为三维空间。另一方面，动作识别是一种分析估计姿势数据序列的方法，目的是将序列归类。它被广泛应用于各个领域。其中一个广为人知的用例是利用监控摄像头的视频分析和检测潜在的非法行为攻击。另一个用例涉及姿势序列分析，目的是创建虚拟教练环境。具体来说，我们的研究将以这一具有挑战性的问题为目标，旨在为移动设备创建这一环境。我们将介绍一些适合在移动设备上有效进行姿势估计和动作识别的解决方案。我们将展示如何利用基于卷积神经网络的轻量级模型来有效解决姿势估计问题，并利用动态时间扭曲算法来解决动作识别问题。
{"title":"On Mobile Pose Estimation and Action Recognition Design and Implementation","authors":"M. Aslanyan","doi":"10.1134/s1054661824010036","DOIUrl":"https://doi.org/10.1134/s1054661824010036","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>Human pose estimation (PE, tracking body pose on-the-go) is a computer vision-based technology that identifies and controls specific points on the human body. These points represent our joints and special points over the body determining the sizes, distances, angle of flexion, and type of the motion. Knowing this in a specific exercise is the basis of work for rehabilitation and physiotherapy, fitness and self-coaching, augmented reality, animation and gaming, robot management, surveillance and human activity analysis. Implementing such capabilities may use special suits or sensor arrays to achieve the best result, but massive use of PE is related to devices that many users own: namely smartphones, smartwatches, and earbuds. The body pose estimation system starts with capturing the initial data. In dealing with motion detection, it is necessary to analyze a sequence of images rather than a still photo. Different software modules are responsible for tracking 2D key points, creating a body representation, and converting it into a 3D space. Action recognition on the other hand is a way to analyze the sequence of estimated pose data with the aim to categorize sequence under the classes. It is widely used various fields. One of the widely known use cases is analysis and detection of potential attacks of illegal action using video from the surveillance cameras. Another use case involves analysis of the sequence of pose with the aim of creating a virtual coaching environment. Specifically, our research will target this challenging issue and aim to create this environment for mobile devices. We will describe some of the solutions that are suitable for effectively pose estimation and action recognition on mobile devices. We will show how lightweight models based on convolution neural networks can be used to efficiently solve pose estimation issue and address action recognition problem with the dynamic time warping algorithm.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"8 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140603433","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 : 2024-04-10DOI: 10.1134/s1054661824010115
A. H. Kostanyan
Abstract
This paper is devoted to determining the degree of compliance of a given string with a pattern represented as a grammar, the terminal symbols of which are fuzzy properties of the characters of the base alphabet. In the case when the pattern is specified as a context-free grammar in the Chomsky normal form, the matching degree is calculated by applying a fuzzy version of the Cocke–Younger–Kasami (CYK) algorithm in cubic time depending on the length of the input string. The proposed algorithm becomes a linear time algorithm for the subclass of the automata grammars, which can be considered as finite automata with fuzzy properties of alphabetic characters on transitions. This work may find application in bioinformatics to classify deoxyribonucleic acid (DNA) sequences using fuzzy prototypes described in one way or another. Other applications are related to fuzzy analysis of natural languages, pattern recognition and determination of fuzzy regularity of a string.
{"title":"Algorithms for Matching Strings with Fuzzy Context-Free and Automata Patterns","authors":"A. H. Kostanyan","doi":"10.1134/s1054661824010115","DOIUrl":"https://doi.org/10.1134/s1054661824010115","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>This paper is devoted to determining the degree of compliance of a given string with a pattern represented as a grammar, the terminal symbols of which are fuzzy properties of the characters of the base alphabet. In the case when the pattern is specified as a context-free grammar in the Chomsky normal form, the matching degree is calculated by applying a fuzzy version of the Cocke–Younger–Kasami (CYK) algorithm in cubic time depending on the length of the input string. The proposed algorithm becomes a linear time algorithm for the subclass of the automata grammars, which can be considered as finite automata with fuzzy properties of alphabetic characters on transitions. This work may find application in bioinformatics to classify deoxyribonucleic acid (DNA) sequences using fuzzy prototypes described in one way or another. Other applications are related to fuzzy analysis of natural languages, pattern recognition and determination of fuzzy regularity of a string.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"68 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590432","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 : 2024-04-10DOI: 10.1134/s1054661824010024
L. Aslanyan, Yu. Shoukourian
Abstract
The biennial “Computer Sciences and Information Technologies” conference in Yerevan in September of 2023 was the 14th in this series. This event is of regional interest, provide a networking ecosystem to scientists in areas such as the Mathematical Logic and Logical Reasoning, Discrete Mathematics, Pattern Recognition and Cognitive Sciences, and Applications. 17 selected papers of this conference, included in Special Issue of PRIA journal provide up to date research results and research topics that make a sensitive input to the field known as “Logic Combinatorial Pattern Recognition”. Results from Algebra and Mathematical Logic help in structural and semantic areas of pattern recognition, graph theoretical investigations help with clustering and image analysis, other selected papers and results are devoted directly to semantic pattern recognition, to cognitive sciences in general, or they provide application platforms where different artificial intelligence technologies are converging.
{"title":"Special Issue “Selected Papers of the 14th International Conference Computer Science and Information Technologies”","authors":"L. Aslanyan, Yu. Shoukourian","doi":"10.1134/s1054661824010024","DOIUrl":"https://doi.org/10.1134/s1054661824010024","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The biennial “Computer Sciences and Information Technologies” conference in Yerevan in September of 2023 was the 14th in this series. This event is of regional interest, provide a networking ecosystem to scientists in areas such as the Mathematical Logic and Logical Reasoning, Discrete Mathematics, Pattern Recognition and Cognitive Sciences, and Applications. 17 selected papers of this conference, included in Special Issue of PRIA journal provide up to date research results and research topics that make a sensitive input to the field known as “Logic Combinatorial Pattern Recognition”. Results from Algebra and Mathematical Logic help in structural and semantic areas of pattern recognition, graph theoretical investigations help with clustering and image analysis, other selected papers and results are devoted directly to semantic pattern recognition, to cognitive sciences in general, or they provide application platforms where different artificial intelligence technologies are converging.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"26 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590540","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 : 2024-04-10DOI: 10.1134/s105466182401019x
A. A. Smirnova, N. G. Grafeeva, M. A. Tokman
Abstract
This study explores an approach to recognizing texts of ancient printed books written in Church Slavonic, utilizing computer vision and neural network methods. We achieved a classification accuracy of 99.13% for Church Slavonic alphabet characters, including punctuation marks, and 98.58% for superscript signs. The study developed an application variant enabling the conversion of Church Slavonic text images into an editable format. We formed datasets comprising image samples of letters and superscript signs, featuring no fewer than 200 examples per letter and at least 150 images per sign.
{"title":"Recognition of Church Slavonic Texts Using Machine Learning Methods","authors":"A. A. Smirnova, N. G. Grafeeva, M. A. Tokman","doi":"10.1134/s105466182401019x","DOIUrl":"https://doi.org/10.1134/s105466182401019x","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>This study explores an approach to recognizing texts of ancient printed books written in Church Slavonic, utilizing computer vision and neural network methods. We achieved a classification accuracy of 99.13% for Church Slavonic alphabet characters, including punctuation marks, and 98.58% for superscript signs. The study developed an application variant enabling the conversion of Church Slavonic text images into an editable format. We formed datasets comprising image samples of letters and superscript signs, featuring no fewer than 200 examples per letter and at least 150 images per sign.</p>","PeriodicalId":35400,"journal":{"name":"PATTERN RECOGNITION AND IMAGE ANALYSIS","volume":"169 1","pages":""},"PeriodicalIF":1.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140590541","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号