Kooka

Kooka is an open-source optical character recognition (OCR) frontend originally developed for the KDE desktop environment. It provides a graphical interface for scanning documents with SANE-compatible scanners, performing basic image corrections, and running OCR to extract editable text.

Designed to integrate with KDE’s scanning workflow, Kooka supports multiple OCR backends through a plug-in architecture and can work with common engines such as Tesseract and GOCR. Users can preview scans, rotate or deskew images, adjust contrast, and crop before recognition.

Output from Kooka can be saved as plain text, and may be exported to additional formats via available plug-ins or backend options. The project emphasizes an end-to-end workflow from capture to text extraction, suitable for archival tasks or creating editable documents.

Kooka’s development has historical significance within KDE’s utility tools for document handling and image processing. Availability varies by Linux distribution and release, and some modern environments may rely on alternative scanning or OCR solutions. Nonetheless, Kooka remains cited in KDE documentation as an example of desktop-integrated OCR tooling.

See also: KDE, Optical character recognition, SANE, Tesseract.