How to use EasyOCR tool to recognize image text in Python

What is EasyOCR?

Description: EasyOCR is a python module for extracting text from images. It is a general-purpose OCR that can read both natural scene text and documents. Dense text in . Currently supports over 80 languages ​​and all popular writing scripts including: Latin, Chinese, Arabic, Sanskrit, Cyrillic, and more.

EasyOCR is an optical character recognition (OCR) tool implemented in PyTorch.

Q: What can you do with EasyOCR?

Description: EasyOCR supports two ways to run, one is a commonly used CPU, and the other requires GPU support and the CUDA environment needs to be installed. We use It can perform language and text recognition in pictures, such as picture recognition in mini programs and vehicle license plate recognition (i.e., car debt management system).

Install EasyOCR

In the command window, use pip to install the stable version of EasyOCR.

pip install easyocr

When you use EasyOCR

import easyocr

reader = easyocr.Reader(
    ['ch_sim', 'en'],
    gpu=False,
    model_storage_directory='model/.',
    user_network_directory='model/.',
)
result = reader.readtext('examples/chinese.jpg')

to execute the above code, the detection and recognition model will be automatically downloaded to the specified directory through the network.

• ['ch_sim', 'en'],: Specify the recognized language

• gpu=False,: Set whether to use GPU (EasyOCR It runs more efficiently on GPU. Set False when there is no GPU or insufficient GPU memory)

• model_storage_directory='model/.',: Detect and identify the storage path of the model (when not set) Stored in the ~/.EasyOCR/model directory by default)

The recognition resultresult is a list, and each item in the list is a length of 3 recognition results, for example ([[189, 75], [469, 75], [469, 165], [189, 165]], 'Yuyuan Road', 0.3754989504814148) , which are bounding box, detected text and confidence values ​​respectively.

easyocr-server

EasyOCR server is a server for extracting text from images. It is a general-purpose OCR that can read both natural scene text and dense text in documents. Currently 80 languages ​​are supported and growing.

Installation steps

Step 0. Download easyocr-server from GitHub and install it.

git clone https://github.com/hekaiyou/easyocr-server.git

Step 1. Install easyocr, bottle and gevent modules using PyPI.

cd easyocr-server
pip install -r requirements.txt

Verify installation

python main.py

• Browser: http://localhost:8080/ocr/

• CMD: curl http://localhost:8080/ocr/ -F "language=en" -F "img_file=@examples/english.png"

After successful verification, you should be able to see the printed inference results in the browser.

Deployment service through Docker

We provide a Dockerfile to build the image.

docker build -t easyocr-server:latest .

Run it.

docker run -it -v {DATA_DIR}:/workspace/model -p 8083:8080 easyocr-server:latest

The above is the detailed content of How to use EasyOCR tool to recognize image text in Python. For more information, please follow other related articles on the PHP Chinese website!