FIELD: information technology.
SUBSTANCE: optical character recognition systems and methods using specialized confidence functions implemented based on a neural network. An example of the method includes obtaining a grapheme image; calculating, by a neural network, a feature vector representing a grapheme image in the image feature space; and calculating a confidence vector associated with an image of the grapheme, where each element of the confidence vector displays distance in the feature space of images between the feature vector and the class center from the set of classes, wherein said class is identified by the confidence vector element index.
EFFECT: high efficiency of optical character recognition, including optical recognition of grapheme, by using confidence functions to minimize errors.
25 cl, 7 dwg
| Title | Year | Author | Number |
|---|---|---|---|
| NEURAL NETWORK TRAINING BY MEANS OF SPECIALIZED LOSS FUNCTIONS | 2018 |
|
RU2707147C1 |
| HANDWRITING RECOGNITION USING NEURAL NETWORKS | 2020 |
|
RU2757713C1 |
| CHARACTER RECOGNITION USING A HIERARCHICAL CLASSIFICATION | 2018 |
|
RU2693916C1 |
| DIFFERENTIAL CLASSIFICATION WITH MULTIPLE NEURAL NETWORKS | 2017 |
|
RU2652461C1 |
| METHOD FOR PROCESSING IMAGES BY CONVOLUTIONAL NEURAL NETWORKS | 2020 |
|
RU2771442C1 |
| TEXT RECOGNITION USING ARTIFICIAL INTELLIGENCE | 2017 |
|
RU2691214C1 |
| IDENTIFICATION OF WRITING SYSTEMS USED IN DOCUMENTS | 2021 |
|
RU2792743C1 |
| IDENTIFICATION OF FIELDS ON AN IMAGE USING ARTIFICIAL INTELLIGENCE | 2018 |
|
RU2695489C1 |
| SYMBOLS RECOGNITION WITH THE USE OF ARTIFICIAL INTELLIGENCE | 2017 |
|
RU2661750C1 |
| OPTICAL CHARACTER RECOGNITION BY MEANS OF COMBINATION OF NEURAL NETWORK MODELS | 2020 |
|
RU2768211C1 |
