RTS to Latin
Transliteration Service

Drop the image / box pair here!

Upload Files

Use Sample Image

Download Files

 
By: Dr. Marius E. Penteliuc
Box Editor Settings
App Appearance
Adjust the appearance of the interface.
App Language
Choose what language to use in the interface.
Toolbar Actions
Choose actions to display in the toolbar.
Image View
Adjust the size of the image display area.
Editor Tools
Choose which convenience features to make available in the editor.
macOS Keyboard Layout

Users on macOS can download and install a keyboard layout made for them. Download it from the Language Models section

On Image Load
Perform actions when a new image is loaded.
Alerts and Warnings
Enable or surpress the following messages.
Workflow
Choose what convenience features you would like to use.
Keyboard Shortcuts
Customize the keyboard shortcuts for frequently used actions.
Upcoming Feature

In a future update you will be able to add and customize the keyboard shortcuts. For now, you can use these defaults and swap them around.

Recognition Model
Choose a language to use for OCR text suggestions.
Known Issue

Not all recognition models are fully supported yet. Right-to-left scripts do not properly display in the input field, especially when mixed with left-to-right text.

RTS Keyboard Layout

We've made a custom keyboard layout for macOS to allow for easier typing of Cyrillic characters.

Download RTS Keyboard Layout
Text Highlighting
Highlight text in the input field based on custom regex patterns.

Patterns only apply at the character level. Edit the cells under Name, Color, and Pattern columns to modify a highlighter's properties.

Add Pattern
 

How to use the Box Editor

To begin, upload an image or use the provided sample. Select the buttons in the toolbar to perform certain actions such as identifying lines of text and generating suggestions.

Once text boxes are generated for all lines of text, you can start editing as needed. Either manually type in the box dimensions in the coordinate fields or drag the handles on the image to change the size or position of a text box. Use the tools on the right to split boxes into columns, draw new ones, or delete existing boxes.

When the document is complete, download the .box file and .gt.txt files and use them to train your OCR model.

Toolbar Actions:

Toggle invisibles on or off.
Detect all lines of text and draw boxes for them.
Redetect text in boxes and regenerate output.
Redetect text for selected box and regenerate output.
Choose a language model for OCR suggestions.
Navigate to the previous page in a PDF document. new
Navigate to the next page in a PDF document. new
PDF Support

You can now upload and tag PDF documents. The app will display navigation controls and automatically separate tags and ground truth by pages.

Image Actions:

Zoom in on the image.
Zoom out of the image.
Split boxes into columns.
Draw new boxes for text.
Delete one or more boxes.

Reset

In case of unexpected behavior, you can reset all settings to their default values.

Report Issue

About

This is a powerful Box Editor for Tesseract allowing for efficient generation, editing, and managing text boxes in WordStr format. The tool uses a custom trained model for authoring the Romanian Transitional Script.

Version:
1.4.1
Updated:
July 22nd, 2023

For info and support, please visit the following links:

Author: Dr. Marius E. Penteliuc
Source Code: GitHub Repository
Project Website: ROTLA Team
Custom OCR Model: Download RTS updated
Training Dataset: Kaggle

Acknowledgement

This work was supported by a grant from the Romanian Ministry of Research, Innovation and Digitization, CCCDI - UEFISCDI, project number PN-III-P2-2.1-PED-2021-0693, within PNCDI III.

OK