Dynamical System Modeling and Stability Investigation�DSMSI-2023

Dedicated to the 77th anniversary of the outstanding Ukrainian scientist

professor Denys Khusainov

December 19-21, 2023, Kyiv, Ukraine

The system for recognizing useful information of the client's ID-card based on machine learning technologies

​

Oleksii Bychkov, Liudmyla Zubyk, Dmytro Gololobov, Yaroslav Isaienkov, Ganna Grynkevych, Anastasiia Ivanytska

The relevance of research

2022 IEEE Third International Conference on System Analysis & Intelligent Computing (SAIC)

2

Scientific task: to find the approach to recognizing text from the image of ID card, which will allow to develop the dictionary formation model for interaction with the server

The approaches to evaluating the effectiveness of information technology :

1. Models based on Tesseract technology of recognition
2. Models used Paddle Optical Character Recognition ion (OCR)

The actual problems:

1. the existing approaches to text recognition often require image pre-processing models
2. the quality of the input image, its brightness, position relative to the camera and size
3. the complex output format after text recognition

Contemporary scientists:

1. Q. Uddin, T. Bui, T. Nguyen, A. Jatowt, M. Coustaty, Ch. Padole, U. Sh. Verma, Pr. Gujral, Mr. Kumar, C. Madan Kumar, D. Arulanantham, Ivanytska A., Zubyk L, V. Martsenyuk, O. Bychkov, K. Merkulova, Y. Zhabska, Dakhno, N., Barabash, O., Shevchenko, H., Leshchenko, O., Dudnik A, M. Kozlenko, V. Sendetskyi, [13] E. Zhang, V. A. Putra, K. Leung
2. M. R. M. Ribeiro, D. Julio, Phan Van Hoai, H.-Th. Duong, N. S. Yusman, Kr. Olejniczak, M. Sulc, D. Yan, S. Shen, D. Wang, F. Borisyuk, A. Gordo, D. Yu, X. Li, X. Zu, Ch. Zhang

2022 IEEE Third International Conference on System Analysis & Intelligent Computing (SAIC)

3

The aim of this study

develop the model for forming the dictionary based on text recognition technologies from document images and increase the accuracy of information transfer in the client-service system

The following tasks:

1. analysis of text recognition technologies on documents;
2. select the optimal module from two approaches for recognizing text images
3. determine the stages of developing the model of the dictionary formation which based of text recognition technologies, which was selected;
4. evaluate the adequacy of the proposed model on test data.

Text recognition tools based on machine learning technologies

Figure 1: Sequence of data processing recognition system

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Text recognition tools based on machine learning technologies

5.1

where S – number of substitutions;

D– number of deletions;

I – number of insertions;

N – number of characters in reference text.

N=S+D+C (where C- number of correct characters).

​

5.2

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Table 1�The quality of recognition according to the CER and WER parameters with address date of Chinese user’s information

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Text recognition tools based on machine learning technologies

Figure 2: The result of text recognition by the Paddle OCR system

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Development of the text processing model for dictionary formation

The formation of the dictionary of useful information, which looks like this:

"name":"张*舒",

"sex":"女",

"nation":"汉",

"birth":"197***512",

"address":"广州市番**洛溪新城吉祥道十幢之二601房",

"idcard":"441*******0427

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Development of the text processing model for dictionary formation

The next stages of forming the dictionary of useful information:

​

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Steps for forming system responses when working with the server:

Dynamical System Modeling and Stability Investigation, DSMSI-2023

Conclusions

As the result of the research, the model of forming dictionary basing on text recognition technologies, that uses machine learning technologies, was developed and implemented. The following stages of research were performed.

Firstly, existing approaches to recognition of text information from user documents in client-server systems were analyzed in order to solve the problem of user identification. Two approaches were selected for text recognition algorithm.

Secondly, the comparative analysis of the Optical Character Recognition library using Tesseract Engine and Paddle Optical Character Recognition was held. The feasibility and effectiveness of using Paddle OCR for analyzing documents with Chinese characters was substantiated.

Thirdly, the model of the dictionary formation which based of text recognition technologies was developed and tested.

Finally, the adequacy of the model of the dictionary formation was evaluated. Its effectiveness was verified on test data and the adequacy of the model was assessed based CER and WER. In contrast to existing approaches, the model showed the increase in text recognition accuracy by 1.8-11.3%, depending on the quality of the source image. The result is implemented in client-server applications for product solutions of Zhejiang Jimi IoT Technology Co., Ltd.

Dynamical System Modeling and Stability Investigation, DSMSI-2023

CONTACT US�for collaborative machine learning and data mining analysis��a.titova.wk@gmail.com�l.v.zubyk19@gmail.com �gololobov.dma@meta.ua �yisaienkov@gmail.com�Ggrynkevych@ukr.net ��