2020-08-19 18:03:17 +00:00
|
|
|
# manga translator
|
2020-08-24 01:50:57 +00:00
|
|
|
get japanese manga from url to translate manga image using SickZil(tensorflow model), ocr(pytesseract ocr and nhocr) and googletrans
|
2020-08-19 14:30:51 +00:00
|
|
|
|
2020-08-24 02:02:42 +00:00
|
|
|
This project code is still under processing on the colab environment. For use, run colab in below url.
|
2020-08-19 14:30:51 +00:00
|
|
|
https://colab.research.google.com/drive/1XbR7fNXtT4TGlLI1FBcCQv7Gj5mlDvwb?usp=sharing
|
2020-08-19 14:34:05 +00:00
|
|
|
|
|
|
|
|
# result
|
2020-08-19 17:21:24 +00:00
|
|
|
|
2020-08-19 17:21:46 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-19 17:47:36 +00:00
|
|
|
|
2020-08-19 14:39:32 +00:00
|
|
|
|
2020-08-24 02:08:26 +00:00
|
|
|
# Workflow
|
2020-08-24 02:10:15 +00:00
|
|
|
- use gallery-dl to get managa from url
|
|
|
|
|
- do text segmentation from manga image using SickZil
|
2020-08-24 02:11:23 +00:00
|
|
|
- use opencv to crop text image based on text segmentation results
|
2020-08-24 02:10:15 +00:00
|
|
|
- get text from image using pytesseract ocr and nhocr
|
|
|
|
|
- translating using googletrans
|
|
|
|
|
- use pil to place translated text
|
2020-08-24 02:08:26 +00:00
|
|
|
|
|
|
|
|
# Supported Languages (for destination translation)
|
|
|
|
|
For translating, it uses google trans. It supports :
|
|
|
|
|
afrikaans, albanian, amharic, arabic, armenian, azerbaijani, basque, belarusian, bengali, bosnian, bulgarian, catalan, chichewa, chinese (simplified), chinese (traditional), corsican, croatian, czech, danish, dutch, english, esperanto, estonian, filipino, filipino, finnish, french, frisian, galician, georgian, german, greek, gujarati, haitian creole, hausa, hawaiian, hindi, hungarian, icelandic, igbo, indonesian, irish, italian, japanese, kazakh, khmer, korean, kurdish (kurmanji), kyrgyz, lao, latin, latvian, lithuanian, luxembourgish, macedonian, malagasy, malay, malayalam, maltese, maori, marathi, mongolian, myanmar (burmese), nepali, norwegian, pashto, persian, polish, portuguese, punjabi, romanian, russian, samoan, scots gaelic, serbian, sesotho, shona, sindhi, sinhala, slovak, slovenian, somali, spanish, sundanese, swahili, swedish, tajik, tamil, thai, turkish, ukrainian, urdu, uzbek, vietnamese, welsh, xhosa, yiddish, yoruba, zulu
|
|
|
|
|
|
|
|
|
|
# Supported url
|
|
|
|
|
gallery-dl is used to download. Its support sites are:
|
|
|
|
|
- [text-detection](https://github.com/mikf/gallery-dl/blob/master/docs/supportedsites.rst)
|
|
|
|
|
|
2020-08-19 17:34:35 +00:00
|
|
|
# Acknowledgement and References
|
2020-08-19 17:35:34 +00:00
|
|
|
- [SickZil-Machine](https://github.com/KUR-creative/SickZil-Machine)
|
2020-08-19 17:34:35 +00:00
|
|
|
- [OpenCV with Python wrapper](https://pypi.org/project/opencv-python/)
|
|
|
|
|
- [Google Translate API for Python](https://pypi.org/project/googletrans/)
|
2020-08-19 17:35:34 +00:00
|
|
|
- [Tesseract](https://github.com/tesseract-ocr/tesseract)
|
2020-08-19 17:34:35 +00:00
|
|
|
- [Pytesseract](https://pypi.python.org/pypi/pytesseract)
|
|
|
|
|
- [nhocr](https://github.com/fireae/nhocr)
|
2020-08-23 15:43:14 +00:00
|
|
|
- [text-detection](https://github.com/qzane/text-detection)
|
|
|
|
|
|
2020-08-19 17:35:34 +00:00
|
|
|
|
