https://github.com/shreeshrii/tessdata_jav_java

Tesseract 4.0.0 training data for Javanese Script (Aksara Jawa)
https://github.com/shreeshrii/tessdata_jav_java

Last synced: 2 months ago
JSON representation

Tesseract 4.0.0 training data for Javanese Script (Aksara Jawa)

• Host: GitHub
• URL: https://github.com/shreeshrii/tessdata_jav_java
• Owner: Shreeshrii
• License: apache-2.0
• Created: 2018-07-19T18:01:08.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2020-03-23T04:04:35.000Z (about 5 years ago)
• Last Synced: 2025-01-26T05:11:26.361Z (4 months ago)
• Language: Shell
• Size: 210 MB
• Stars: 3
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# tessdata_jav_java

Tesseract 4.0.0 training data for Javanese Script (Aksara Jawa)

Built in response to [this issue](https://github.com/tesseract-ocr/langdata/issues/126)

## Traineddata files

* [jav1.traineddata](https://github.com/Shreeshrii/tessdata_jav_java/blob/master/tessdata_best/jav1.traineddata)

* [jav2.traineddata](https://github.com/Shreeshrii/tessdata_jav_java/blob/master/tessdata_best/jav2.traineddata)

## To run training for Javanese script

* Clone this repo

* To further continue training with existing data, run `./plustrain.sh`

* To customize for your own training, update training text in langdata and fonts lists in `makedata.sh`

* Training text needs to be in utf-8 encoding.

* Unicode fonts supporting the Javanese code-range need to be used.

* Source code changes will be needed in tesseract, additions will be similar to Khmer/Mynamar or Thai.

## Custom bash scripts - run in following order

./makeeval.sh

./makedata.sh

./mergedata.sh

./plustrain.sh