OCR

# OCR

An Android OCR app based on Tesseract that can recognize texts on images.

This app is now based(Since v3.0) on Tesseract 5 and the first of android app which is based on Tesseract 5.

###After downloading the Training data, the app does everything offline on your device.

Translate this app on [Hosted Weblate](https://hosted.weblate.org/projects/android-ocr).

🚀 Features
* Highest accuracy and fastest result.
* Extract text from images.
* Copy data to clipboard.
* Select any part of the data.
* Process multiple languages in single image(Required changes in settings).
* Based on latest Training Data.
* 120+ recognized languages.
* 3 Type of data i.e Standard, best, fast for more info click here.
* Math recognition/equation detection(Not Tested yet).
* Process images directly from the gallery on your device via the share menu.

Required Permissions

* Internet: Downloading the training data
* Access Internet state: To download the training data.
* Note: None of your data is sent to anywhere.

📓 License

* Tesseract and Tesseract Data: Apache 2.0.
* Tesseract4Android: Apache 2.0.
* ImageCropper: Apache 2.0.
* SpinnerDialog: [Apache 2.0

Made possible by the Tesseract4Android library.

Major Milestone: Jetpack Compose Migration

The app has undergone a complete architectural and UI overhaul, moving away from legacy XML-based layouts and Java-heavy logic to a modern, reactive Kotlin-first approach.

-Features & UI/UX Overhaul

•Material 3 Implementation: Fully adopted Material Design 3 (M3) standards, providing a modern, clean, and accessible user interface.
•Complete UI Migration: All screens have been rebuilt from the ground up using Jetpack Compose.
•Enhanced Home Experience