python-ocr

Optical Character Recognition for images, Pdfs, zip files, tif files.

What you can expect from this repository:

• Efficient ways to get textual information from your documents like images, pdfs, zip files.

Quick Tour

Get text from documents and save results in JSON.

Installation

Developer mode

pip install python-ocr

For tesseractOcr process

storage_type='local/aws' #currently only local and aws supported. local storage_path='Desired path of your OS where you want to store the output' # for local storage. local storage_path='S3 bucket' # for AWS storage (CASE SENSTIVE).

e.g. for Storing output to AWS

config={'storage_type':'AWS','storage_path:'your-bucket-name'}

from ocr import TesseractOcrProcessor
process=TesseractOcrProcessor(config)

For EasyOcr process

from ocr import EasyOcrProcessor
process=EasyOcrProcessor(config)

storage_type: type of storage local or aws.

storage_path: storage path is path where user wants to store the output result.

# Path of file
PATH=''

# reading image files
process.process_image(PATH)

# reading pdf files
process.process_pdf(PATH)

# reading zip files
process.process_zip(PATH)

Documentation:

The full package documentation is available here.

First of all, you have to create dict of storage_type and storage_path.

1. storage_type: storage type is type of storage where the user wants to store the output result. It may be local or aws.

2. storage_path: storage path is path where the user wants to store the output result.

   • if you want to store the file in local system than give the path of folder where user wants to store the result as storage_path.

   • if user wants to store the result in aws than in storage_path you have to give the bucket name.

config={'storage_type':'','storage_path':''}

Now create the object of EasyOcrProcessor which take the config as a object parameter.

process = EasyOcrProcessor(config)

Image process:

To read the text from image user have to call the process_image method of EasyOcrProcessor and pass the path of image file as a parameter in it. process_image method store the output at the storage_path.

process.process_image(PATH)

Pdf process:

To read the text from pdf file user have to call the process_pdf method of EasyOcrProcessor and pass the path of pdf file as a parameter in it. process_pdf method convert each page of pdf into images and create the result of each page and store the result at the storage_path.

process.process_pdf(PATH)

Zip process:

To read the text from zip file user have to call the process_zip method of EasyOcrProcessor and pass the path of zip file as a parameter in it. Zip should contain only files with valid extensions. process_zip method extract each file of zip one by one and save the result at the storage path.

process.process_zip(PATH)

Result output:

[{
        "left": 125,
        "top": 141,
        "right": 259,
        "bottom": 161,
        "text": "Folin MGA-5875",
        "confidence": 0.3961432168382489
    },
    {
        "left": 1115,
        "top": 140,
        "right": 1272,
        "bottom": 161,
        "text": "OM8 N0 : 2126-0006",
        "confidence": 0.41482855467690777
    },
    {
        "left": 1281,
        "top": 139,
        "right": 1498,
        "bottom": 165,
        "text": "Epiration Datc 12/31/2024",
        "confidence": 0.40780972855935615
}]