Optical Recognition of Handwritten Logic Formulas Using Neural Networks
In this paper, we present a handwritten character recognition (HCR) system that aims to recognize first-order logic handwritten formulas and create editable text files of the recognized formulas. Dense feedforward neural networks (NNs) are utilized, and their performance is examined under various tr...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/10/22/2761 |
| _version_ | 1797510610775179264 |
|---|---|
| author | Vaios Ampelakiotis Isidoros Perikos Ioannis Hatzilygeroudis George Tsihrintzis |
| author_facet | Vaios Ampelakiotis Isidoros Perikos Ioannis Hatzilygeroudis George Tsihrintzis |
| author_sort | Vaios Ampelakiotis |
| collection | DOAJ |
| description | In this paper, we present a handwritten character recognition (HCR) system that aims to recognize first-order logic handwritten formulas and create editable text files of the recognized formulas. Dense feedforward neural networks (NNs) are utilized, and their performance is examined under various training conditions and methods. More specifically, after three training algorithms (backpropagation, resilient propagation and stochastic gradient descent) had been tested, we created and trained an NN with the stochastic gradient descent algorithm, optimized by the Adam update rule, which was proved to be the best, using a trainset of 16,750 handwritten image samples of 28 × 28 each and a testset of 7947 samples. The final accuracy achieved is 90.13%. The general methodology followed consists of two stages: the image processing and the NN design and training. Finally, an application has been created that implements the methodology and automatically recognizes handwritten logic formulas. An interesting feature of the application is that it allows for creating new, user-oriented training sets and parameter settings, and thus new NN models. |
| first_indexed | 2024-03-10T05:33:37Z |
| format | Article |
| id | doaj.art-bac261d9ab684d67a521b64faad04007 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T05:33:37Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-bac261d9ab684d67a521b64faad040072023-11-22T23:06:43ZengMDPI AGElectronics2079-92922021-11-011022276110.3390/electronics10222761Optical Recognition of Handwritten Logic Formulas Using Neural NetworksVaios Ampelakiotis0Isidoros Perikos1Ioannis Hatzilygeroudis2George Tsihrintzis3Department of Computer Engineering and Informatics, University of Patras, 26504 Patras, GreeceDepartment of Computer Engineering and Informatics, University of Patras, 26504 Patras, GreeceDepartment of Computer Engineering and Informatics, University of Patras, 26504 Patras, GreeceDepartment of Informatics, University of Piraeus, 18534 Piraeus, GreeceIn this paper, we present a handwritten character recognition (HCR) system that aims to recognize first-order logic handwritten formulas and create editable text files of the recognized formulas. Dense feedforward neural networks (NNs) are utilized, and their performance is examined under various training conditions and methods. More specifically, after three training algorithms (backpropagation, resilient propagation and stochastic gradient descent) had been tested, we created and trained an NN with the stochastic gradient descent algorithm, optimized by the Adam update rule, which was proved to be the best, using a trainset of 16,750 handwritten image samples of 28 × 28 each and a testset of 7947 samples. The final accuracy achieved is 90.13%. The general methodology followed consists of two stages: the image processing and the NN design and training. Finally, an application has been created that implements the methodology and automatically recognizes handwritten logic formulas. An interesting feature of the application is that it allows for creating new, user-oriented training sets and parameter settings, and thus new NN models.https://www.mdpi.com/2079-9292/10/22/2761optical character recognitionlogic formulasneural networksresilient propagationOpenCVEncog |
| spellingShingle | Vaios Ampelakiotis Isidoros Perikos Ioannis Hatzilygeroudis George Tsihrintzis Optical Recognition of Handwritten Logic Formulas Using Neural Networks Electronics optical character recognition logic formulas neural networks resilient propagation OpenCV Encog |
| title | Optical Recognition of Handwritten Logic Formulas Using Neural Networks |
| title_full | Optical Recognition of Handwritten Logic Formulas Using Neural Networks |
| title_fullStr | Optical Recognition of Handwritten Logic Formulas Using Neural Networks |
| title_full_unstemmed | Optical Recognition of Handwritten Logic Formulas Using Neural Networks |
| title_short | Optical Recognition of Handwritten Logic Formulas Using Neural Networks |
| title_sort | optical recognition of handwritten logic formulas using neural networks |
| topic | optical character recognition logic formulas neural networks resilient propagation OpenCV Encog |
| url | https://www.mdpi.com/2079-9292/10/22/2761 |
| work_keys_str_mv | AT vaiosampelakiotis opticalrecognitionofhandwrittenlogicformulasusingneuralnetworks AT isidorosperikos opticalrecognitionofhandwrittenlogicformulasusingneuralnetworks AT ioannishatzilygeroudis opticalrecognitionofhandwrittenlogicformulasusingneuralnetworks AT georgetsihrintzis opticalrecognitionofhandwrittenlogicformulasusingneuralnetworks |