Отрывок: We also tested image mirroring augmentation technique but it resulted in quality degradation, because fragments of slanted text lines bleeding from the opposite page side started to mess up with the regular ones. Gaussian blurring also didn’t help us in this problem. The random elastic deformations allowed us to produce better results on handwritten images, but on printed ones results got worse and, after all, we refused to use them. From Table ...
Название : | U-Net-bin: hacking the document image binarization contest |
Авторы/Редакторы : | Bezmaternykh, P.V. Ilin, D.A. Nikolaev, D.P. |
Ключевые слова : | historical document processing binarization DIBCO deep learning U-Net architecture training dataset augmentation document analysis |
Дата публикации : | Окт-2019 |
Издательство : | Новая техника |
Библиографическое описание : | Bezmaternykh, P.V. U-Net-bin: hacking the document image binarization contest / P.V. Bezmaternykh, D.A. Ilin, D.P. Nikolaev // Computer Optics. – 2019. – Vol. 43(5). – P. 825-832. – DOI: 10.18287/2412-6179-2019-43-5-825-832. |
Серия/номер : | 43;5 |
Аннотация : | Image binarization is still a challenging task in a variety of applications. In particular, Document Image Binarization Contest (DIBCO) is organized regularly to track the state-of-the-art techniques for the historical document binarization. In this work we present a binarization method that was ranked first in the DIBCO`17 contest. It is a convolutional neural network (CNN) based method which uses U-Net architecture, originally designed for biomedical image segmentation. We describe our approach to training data preparation and contest ground truth examination and provide multiple insights on its construction (so called hacking). It led to more accurate historical document binarization problem statement with respect to the challenges one could face in the open access datasets. A docker container with the final network along with all the supplementary data we used in the training process has been published on Github. |
URI (Унифицированный идентификатор ресурса) : | https://dx.doi.org/10.18287/2412-6179-2019-43-5-825-832 http://repo.ssau.ru/handle/Zhurnal-Komputernaya-optika/UNetbin-hacking-the-document-image-binarization-contest-80243 |
Другие идентификаторы : | Dspace\SGAU\20191117\80243 |
Располагается в коллекциях: | Журнал "Компьютерная оптика" |
Файлы этого ресурса:
Файл | Описание | Размер | Формат | |
---|---|---|---|---|
430516.pdf | Основная статья | 3.14 MB | Adobe PDF | Просмотреть/Открыть |
Показать полное описание ресурса
Просмотр статистики
Поделиться:
Все ресурсы в архиве электронных ресурсов защищены авторским правом, все права сохранены.