Download jTessBoxEditor – Download Free

jTessBoxEditor is an application designed to provide users with a companion product to the Tesseract OCR software package. It will provide methods for editing Box data generated by Tesseract 2.0x and 3.0x versions, and allow them to perform fully automated text recognition training on Tesseract.

jTessBoxEditor requires Java as the main prerequisite and can read most common image formats such as TIFF, JPEG, GIF, PNG, BMP and multi-page TIFF. Furthermore, in addition to the above image formats, the application also supports PDF format files.

Without the actual installation process, the user can simply open the JAR executable file provided in the deployment archive to run the application. The editor will use TIFF or Box file formats and Tesseract OCR training will require an image quality of at least 300 DPI.

The application will provide a basic set of hotkeys in its Box View window that one can use to increase editing speed. They will be able to move the box up/down/left/right, increase its width/height, switch previous/next box or edit its characters.

After the text generation process is complete, for each input UTF-8 text file, the editor will generate corresponding TIFF/Box file pairs, which users can input when training Tesseract OCR for text recognition. To eliminate bounding box overlapping issues, users can adjust letter tracking or character spacing in the image generator.