robertknight/ocrs
Rust library and CLI tool for OCR (extracting text from images)
What it solves
Ocrs is designed to extract text from images (OCR) while reducing the need for manual image preprocessing that is typically required by older engines like Tesseract. It aims to provide a modern, easy-to-compile OCR engine that works across various platforms, including WebAssembly.
How it works
The project uses neural network models trained in PyTorch, which are then exported to ONNX format and executed via the RTen engine. This machine-learning-heavy pipeline allows it to handle a wide variety of image types, such as scanned documents, screenshots, and photos containing text.
Who it’s for
It is intended for developers who need a Rust-based OCR library or CLI tool that is easy to deploy and cross-platform compatible.
Highlights
- Cross-platform support: Compatible with multiple platforms, including WebAssembly.
- ML-driven pipeline: Uses PyTorch and ONNX to minimize preprocessing effort.
- Open datasets: Trained on open and liberally licensed datasets.
- CLI capabilities: Includes a command-line tool that supports reading from files or the system clipboard and outputting text, JSON layout information, or annotated images.
Related
- Project
- Project
- Project
- Project
- Project