new_predicted_result_GWT2180 = pytesseract.image_to_string(gaussian_blur_license_plate, lang ='eng',

config ='--oem 3 -l eng --psm 6 -c tessedit_char_whitelist = ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789')

filter_new_predicted_result_GWT2180 = "".join(new_predicted_result_GWT2180.split()).replace(":", "").replace("-", "")

print(filter_new_predicted_result_GWT2180)

# specify path to the license plate images folder as shown below

path_for_license_plates = os.getcwd() + "/license-plates/**/*.jpg"

list_license_plates = []

predicted_license_plates = []

for path_to_license_plate in glob.glob(path_for_license_plates, recursive = True):

    license_plate_file = path_to_license_plate.split("/")[-1]

    license_plate, _ = os.path.splitext(license_plate_file)

    '''

    Here we append the actual license plate to a list

    '''

    list_license_plates.append(license_plate)

    '''

    Read each license plate image file using openCV

    '''

    img = cv2.imread(path_to_license_plate)

    '''

    We then pass each license plate image file

    to the Tesseract OCR engine using the Python library 

    wrapper for it. We get back predicted_result for 

    license plate. We append the predicted_result in a

    list and compare it with the original the license plate

    '''

    predicted_result = pytesseract.image_to_string(img, lang ='eng',

    config ='--oem 3 --psm 6 -c tessedit_char_whitelist = ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789')

    filter_predicted_result = "".join(predicted_result.split()).replace(":", "").replace("-", "")

    predicted_license_plates.append(filter_predicted_result)

print("Actual License Plate", "\t", "Predicted License Plate", "\t", "Accuracy")

print("--------------------", "\t", "-----------------------", "\t", "--------")

def calculate_predicted_accuracy(actual_list, predicted_list):

    for actual_plate, predict_plate in zip(actual_list, predicted_list):

        accuracy = "0 %"

        num_matches = 0

        if actual_plate == predict_plate:

            accuracy = "100 %"

        else:

            if len(actual_plate) == len(predict_plate):

                for a, p in zip(actual_plate, predict_plate):

                    if a == p:

                        num_matches += 1

                accuracy = str(round((num_matches / len(actual_plate)), 2) * 100)

                accuracy += "%"

        print("     ", actual_plate, "\t\t\t", predict_plate, "\t\t  ", accuracy)

calculate_predicted_accuracy(list_license_plates, predicted_license_plates)

# Read the license plate file and display it

test_license_plate = cv2.imread(os.getcwd() + "/license-plates / GWT2180.jpg")

plt.imshow(test_license_plate)

plt.axis('off')

plt.title('GWT2180 license plate')

resize_test_license_plate = cv2.resize(

    test_license_plate, None, fx = 2, fy = 2, 

    interpolation = cv2.INTER_CUBIC)

grayscale_resize_test_license_plate = cv2.cvtColor(

    resize_test_license_plate, cv2.COLOR_BGR2GRAY)

gaussian_blur_license_plate = cv2.GaussianBlur(

    grayscale_resize_test_license_plate, (5, 5), 0)