LWSINet: A deep learning-based approach towards video script identification
| Autor: | Kaushik Roy, K. C. Santosh, Mridul Ghosh, Nibaran Das, Himadri Mukherjee, Sk Md Obaidullah |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Computer Networks and Communications
Computer science business.industry Salt (cryptography) Speech recognition Deep learning 020207 software engineering 02 engineering and technology Optical character recognition computer.software_genre Identification (information) Hardware and Architecture Scripting language ComputingMethodologies_DOCUMENTANDTEXTPROCESSING 0202 electrical engineering electronic engineering information engineering Media Technology Feature (machine learning) The Internet Noise (video) Artificial intelligence business computer Software |
| Zdroj: | Multimedia Tools and Applications. 80:29095-29128 |
| ISSN: | 1573-7721 1380-7501 |
| Popis: | Videos – a high volume of texts – broadcast via different media, such as television and the internet. Since Optical Character Recognition (OCR) engines are script-dependent, script identification is a precursor. Other than that, video script identification is not trivial as we have difficult issues, such as low resolution, complex background, noise, and blur effects. In this work, a deep learning-based system, which we call LWSINet: LightWeight Script Identification Network (6-layered CNN) is proposed to identify video scripts. For validation, we used a publicly available dataset named CVSI-15. Besides, the effects of three common noises namely, Salt & pepper, Gaussian and Poisson were considered on the scripts along with their hybridized metamorphosis. In our test results, we observed that the proposed CNN is coherent and robust enough to identify scripts in both scenarios, with and without noise. Further, we also employed other well-known handcrafted feature-based and deep learning approaches for a comparison. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink