jaidedai/easyocr is a Python OCR (optical character recognition) library that finds and reads the characters in an image and turns them into a string. It extracts the text regions from an image along with the recognized strings. It has 30,093 GitHub stars, the license is Apache-2.0, and the primary language is Python. The latest release recorded in the repository is v1.7.2, and the provided information contains no changelog for this version, so what changed cannot be confirmed.
The breadth of this library lies in its languages and writing systems. EasyOCR supports more than 80 languages and several writing systems such as Latin, Chinese, Arabic, Devanagari, and Cyrillic. The README states that the input can be an image file, an OpenCV image array, image bytes, or a source image URL, and that it can also be run from the command line. The results include the text, bounding boxes, and confidence scores. In environments without a GPU, you can set gpu=False to run on the CPU.
No figures are given for recognition accuracy or processing speed. The momentum score assigned by the pipeline is 3.1 out of 10, while the novelty, technical, and quality scores are not given. The repository was created on March 14, 2020. The range of language support, the Apache-2.0 license, and the configuration that runs without a GPU are confirmed, but the actual recognition performance is hard to gauge from this information alone.