Python OCR.parse_image Exemples

Exemple #1

Fichier : lambda_handler.py Projet : BlakeJordan/tesseract-ocr-lambda-function

def lambda_handler(base_64_string: str, context):
    ocr = OCR(debug_mode=False, aws_request_id=context.aws_request_id)
    (recognized_text, status_code,
     conf) = ocr.parse_image(base_64_string=base_64_string)

    return {
        'conf': json.dumps(conf),
        'text': json.dumps(recognized_text),
        'statusCode': json.dumps(status_code)
    }

Exemple #2

Fichier : lambda_handler.py Projet : sethepeterson/tesseract-ocr-lambda-function

def lambda_handler(base_64_string: str, context):
    ocr = OCR(debug_mode=False, aws_request_id=context.aws_request_id)
    (status_code, recognized_text,
     confidence_values) = ocr.parse_image(base_64_string=base_64_string)

    return {
        'statusCode': json.dumps(status_code),
        'recognizedText': json.dumps(recognized_text),
        'confidenceValues': json.dumps(confidence_values)
    }

Exemple #3

    print('=====================================')

    # Convert file to encoded Base64 string.
    file_path = test_files_path.format(file_name)
    with open(file_path, 'rb') as image_file:
        base_64_string = base64.b64encode(image_file.read())

        # Create OCR object.
        ocr = OCR(debug_mode=True)

        # Create temp_files directory.
        os.makedirs(ocr.temp_files_directory_path, exist_ok=True)

        # Get output.
        (_, recognized_text,
         _) = ocr.parse_image(base_64_string=base_64_string)

        # Delete temp_files directory.
        shutil.rmtree(ocr.temp_files_directory_path)

        # Case: test passed.
        if recognized_text == expected_text:
            print(test_passed.format(file_name))

        # Case: test failed.
        else:
            print(test_failed.format(file_name, expected_text,
                                     recognized_text))

print('\n\n\n')

Exemple #4

Fichier : ocr_tester.py Projet : BlakeJordan/tesseract-ocr-lambda-function

# Execute test cases.
print('\n')
for file_name, expected_text in test_cases:

    # Convert file to encoded Base64 string.
    file_path = test_files_path.format(file_name)
    with open(file_path, 'rb') as image_file:
        base_64_string = base64.b64encode(image_file.read())

        # Create OCR object.
        ocr = OCR(debug_mode=True)

        # Create temp_files directory.
        os.makedirs(ocr.temp_files_directory_path, exist_ok=True)

        # Get output.
        (recognized_text, _, _) = ocr.parse_image(base_64_string=base_64_string)

        # Delete temp_files directory.
        shutil.rmtree(ocr.temp_files_directory_path)

        # Case: test passed.
        if recognized_text == expected_text:
            print(test_passed.format(file_name))

        # Case: test failed.
        else:
            print(test_failed.format(file_name, expected_text, recognized_text))
            
print('\n\n\n')