So implementieren Sie die Erkennung maschinenlesbarer Zonen (MRZ) in Python

Maschinenlesbare Zone (MRZ) ist eine entscheidende Funktion, die in modernen Pässen, Visa und Personalausweisen zum Einsatz kommt. Es enthält wesentliche Informationen über den Dokumenteninhaber, wie z. B. seinen Namen, sein Geschlecht, seinen Ländercode und seine Dokumentennummer. Die MRZ-Erkennung spielt eine Schlüsselrolle bei der Grenzkontrolle, der Flughafensicherheit und den Check-in-Prozessen in Hotels. In diesem Tutorial zeigen wir, wie Sie das Dynamsoft Capture Vision SDK nutzen, um die MRZ-Erkennung unter Windows, Linux und macOS zu implementieren Plattformen. Dieser Leitfaden bietet einen schrittweisen Ansatz zur Nutzung der leistungsstarken Funktionen des SDK, um die plattformübergreifende MRZ-Erkennung nahtlos und effizient zu gestalten.

Python MRZ-Erkennungsdemo auf macOS

Voraussetzungen

• Dynamsoft Capture Vision-Testlizenz: Erhalten Sie einen 30-Tage-Testlizenzschlüssel für das Dynamsoft Capture Vision SDK.

• Python-Pakete: Installieren Sie die erforderlichen Python-Pakete mit den folgenden Befehlen:
  

  pip install dynamsoft-capture-vision-bundle opencv-python
  

  Wozu dienen diese Pakete?

  • dynamsoft-capture-vision-bundle ist das Dynamsoft Capture Vision SDK für Python.
  • opencv-python erfasst Kamerabilder und zeigt verarbeitete Bildergebnisse an.

Erste Schritte mit dem Dynamsoft Python Capture Vision-Beispiel

Das offizielle MRZ-Scanner-Beispiel zeigt, wie man mit dem Dynamsoft Capture Vision SDK in kurzer Zeit einen einfachen Python-basierten MRZ-Reader erstellt.

Werfen wir einen Blick auf den Quellcode und analysieren seine Funktionalität:

import sys
from dynamsoft_capture_vision_bundle import *
import os

class MRZResult:
    def __init__(self, item: ParsedResultItem):
        self.doc_type = item.get_code_type()
        self.raw_text=[]
        self.doc_id = None
        self.surname = None
        self.given_name = None
        self.nationality = None
        self.issuer = None
        self.gender = None
        self.date_of_birth = None
        self.date_of_expiry = None
        if self.doc_type == "MRTD_TD3_PASSPORT":
            if item.get_field_value("passportNumber") != None and item.get_field_validation_status("passportNumber") != EnumValidationStatus.VS_FAILED:
                self.doc_id = item.get_field_value("passportNumber")
            elif item.get_field_value("documentNumber") != None and item.get_field_validation_status("documentNumber") != EnumValidationStatus.VS_FAILED:
                self.doc_id = item.get_field_value("documentNumber")

        line = item.get_field_value("line1")
        if line is not None:
            if item.get_field_validation_status("line1") == EnumValidationStatus.VS_FAILED:
                line += ", Validation Failed"
            self.raw_text.append(line)
        line = item.get_field_value("line2")
        if line is not None:
            if item.get_field_validation_status("line2") == EnumValidationStatus.VS_FAILED:
                line += ", Validation Failed"
            self.raw_text.append(line)
        line = item.get_field_value("line3")
        if line is not None:
            if item.get_field_validation_status("line3") == EnumValidationStatus.VS_FAILED:
                line += ", Validation Failed"
            self.raw_text.append(line)

        if item.get_field_value("nationality") != None and item.get_field_validation_status("nationality") != EnumValidationStatus.VS_FAILED:
            self.nationality = item.get_field_value("nationality")
        if item.get_field_value("issuingState") != None and item.get_field_validation_status("issuingState") != EnumValidationStatus.VS_FAILED:
            self.issuer = item.get_field_value("issuingState")
        if item.get_field_value("dateOfBirth") != None and item.get_field_validation_status("dateOfBirth") != EnumValidationStatus.VS_FAILED:
            self.date_of_birth = item.get_field_value("dateOfBirth")
        if item.get_field_value("dateOfExpiry") != None and item.get_field_validation_status("dateOfExpiry") != EnumValidationStatus.VS_FAILED:
            self.date_of_expiry = item.get_field_value("dateOfExpiry")
        if item.get_field_value("sex") != None and item.get_field_validation_status("sex") != EnumValidationStatus.VS_FAILED:
            self.gender = item.get_field_value("sex")
        if item.get_field_value("primaryIdentifier") != None and item.get_field_validation_status("primaryIdentifier") != EnumValidationStatus.VS_FAILED:
            self.surname = item.get_field_value("primaryIdentifier")
        if item.get_field_value("secondaryIdentifier") != None and item.get_field_validation_status("secondaryIdentifier") != EnumValidationStatus.VS_FAILED:
            self.given_name = item.get_field_value("secondaryIdentifier")
    def to_string(self):
        msg = (f"Raw Text:\n")
        for index, line in enumerate(self.raw_text):
            msg += (f"\tLine {index + 1}: {line}\n")
        msg+=(f"Parsed Information:\n"
            f"\tDocumentType: {self.doc_type or ''}\n"
            f"\tDocumentID: {self.doc_id or ''}\n"
            f"\tSurname: {self.surname or ''}\n"
            f"\tGivenName: {self.given_name or ''}\n"
            f"\tNationality: {self.nationality or ''}\n"
            f"\tIssuingCountryorOrganization: {self.issuer or ''}\n"
            f"\tGender: {self.gender or ''}\n"
            f"\tDateofBirth(YYMMDD): {self.date_of_birth or ''}\n"
            f"\tExpirationDate(YYMMDD): {self.date_of_expiry or ''}\n")
        return msg
def print_results(result: ParsedResult) -> None:
    tag = result.get_original_image_tag()
    if isinstance(tag, FileImageTag):
        print("File:", tag.get_file_path())
    if result.get_error_code() != EnumErrorCode.EC_OK:
        print("Error:", result.get_error_string())        
    else:
        items = result.get_items()
        print("Parsed", len(items), "MRZ Zones.")
        for item in items:
            mrz_result = MRZResult(item)
            print(mrz_result.to_string())

if __name__ == '__main__':

    print("**********************************************************")
    print("Welcome to Dynamsoft Capture Vision - MRZ Sample")
    print("**********************************************************")

    error_code, error_message = LicenseManager.init_license("LICENSE-KEY")
    if error_code != EnumErrorCode.EC_OK and error_code != EnumErrorCode.EC_LICENSE_CACHE_USED:
        print("License initialization failed: ErrorCode:", error_code, ", ErrorString:", error_message)
    else:
        cvr_instance = CaptureVisionRouter()
        while (True):
            image_path = input(
                ">> Input your image full path:\n"
                ">> 'Enter' for sample image or 'Q'/'q' to quit\n"
            ).strip('\'"')

            if image_path.lower() == "q":
                sys.exit(0)

            if image_path == "":
                image_path = "../Images/passport-sample.jpg"

            if not os.path.exists(image_path):
                print("The image path does not exist.")
                continue
            result = cvr_instance.capture(image_path, "ReadPassportAndId")
            if result.get_error_code() != EnumErrorCode.EC_OK:
                print("Error:", result.get_error_code(), result.get_error_string())
            else:
                parsed_result = result.get_parsed_result()
                if parsed_result is None or len(parsed_result.get_items()) == 0:
                    print("No parsed results.")
                else:
                    print_results(parsed_result)
    input("Press Enter to quit...")

Erklärung

• Die Methode LicenseManager.init_license initialisiert das Dynamsoft Capture Vision SDK mit einem gültigen Lizenzschlüssel.
• Die CaptureVisionRouter-Klasse verwaltet Bildverarbeitungsaufgaben und koordiniert verschiedene Bildverarbeitungsmodule. Seine Capture-Methode verarbeitet das Eingabebild und gibt das Ergebnis zurück.
• Die ReadPassportAndId ist eine integrierte Vorlage, die den Verarbeitungsmodus angibt. Das SDK unterstützt verschiedene Verarbeitungsmodi, wie z. B. MRZ-Erkennung, Dokumentkantenerkennung und Barcode-Erkennung.
• Die Methode get_parsed_result ruft das MRZ-Erkennungsergebnis als Wörterbuch ab. Die MRZResult-Klasse extrahiert und verpackt die relevanten MRZ-Informationen. Da diese Klasse in verschiedenen Anwendungen wiederverwendet werden kann, wird empfohlen, sie in eine utils.py-Datei zu verschieben.

Im nächsten Abschnitt verwenden wir OpenCV, um die MRZ-Erkennungsergebnisse zu visualisieren und die erkannten MRZ-Zonen auf dem Passbild anzuzeigen.

Visualizing Machine Readable Zone Location in a Passport Image

In the code above, result is an instance of the CapturedResult class. Calling its get_recognized_text_lines_result() method retrieves a list of TextLineResultItem objects. Each TextLineResultItem object contains the coordinates of the detected text line. Use the following code snippet to extract the coordinates and draw contours on the passport image:

cv_image = cv2.imread(image_path)
line_result = result.get_recognized_text_lines_result()

items = line_result.get_items()
for item in items:
    location = item.get_location()
    x1 = location.points[0].x
    y1 = location.points[0].y
    x2 = location.points[1].x
    y2 = location.points[1].y
    x3 = location.points[2].x
    y3 = location.points[2].y
    x4 = location.points[3].x
    y4 = location.points[3].y
    del location

    cv2.drawContours(
        cv_image, [np.intp([(x1, y1), (x2, y2), (x3, y3), (x4, y4)])], 0, (0, 255, 0), 2)

cv2.imshow(
    "Original Image with Detected MRZ Zone", cv_image)
cv2.waitKey(0)
cv2.destroyAllWindows()

Scanning and Recognizing MRZ in Real-time via Webcam

Scanning and recognizing MRZ in real-time via webcam requires capturing a continuous image stream. We can use the OpenCV library to capture frames from the webcam and process them with the Dynamsoft Capture Vision SDK. The following code snippet demonstrates how to implement real-time MRZ recognition using a webcam:

from dynamsoft_capture_vision_bundle import *
import cv2
import numpy as np
import queue
from utils import *


class FrameFetcher(ImageSourceAdapter):
    def has_next_image_to_fetch(self) -> bool:
        return True

    def add_frame(self, imageData):
        self.add_image_to_buffer(imageData)


class MyCapturedResultReceiver(CapturedResultReceiver):
    def __init__(self, result_queue):
        super().__init__()
        self.result_queue = result_queue

    def on_captured_result_received(self, captured_result):
        self.result_queue.put(captured_result)


if __name__ == '__main__':
    errorCode, errorMsg = LicenseManager.init_license(
        "LICENSE-KEY")
    if errorCode != EnumErrorCode.EC_OK and errorCode != EnumErrorCode.EC_LICENSE_CACHE_USED:
        print("License initialization failed: ErrorCode:",
              errorCode, ", ErrorString:", errorMsg)
    else:
        vc = cv2.VideoCapture(0)
        if not vc.isOpened():
            print("Error: Camera is not opened!")
            exit(1)

        cvr = CaptureVisionRouter()
        fetcher = FrameFetcher()
        cvr.set_input(fetcher)

        # Create a thread-safe queue to store captured items
        result_queue = queue.Queue()

        receiver = MyCapturedResultReceiver(result_queue)
        cvr.add_result_receiver(receiver)

        errorCode, errorMsg = cvr.start_capturing("ReadPassportAndId")

        if errorCode != EnumErrorCode.EC_OK:
            print("error:", errorMsg)

        while True:
            ret, frame = vc.read()
            if not ret:
                print("Error: Cannot read frame!")
                break

            fetcher.add_frame(convertMat2ImageData(frame))

            if not result_queue.empty():
                captured_result = result_queue.get_nowait()

                items = captured_result.get_items()
                for item in items:

                    if item.get_type() == EnumCapturedResultItemType.CRIT_TEXT_LINE:
                        text = item.get_text()
                        line_results = text.split('\n')
                        location = item.get_location()
                        x1 = location.points[0].x
                        y1 = location.points[0].y
                        x2 = location.points[1].x
                        y2 = location.points[1].y
                        x3 = location.points[2].x
                        y3 = location.points[2].y
                        x4 = location.points[3].x
                        y4 = location.points[3].y
                        cv2.drawContours(
                            frame, [np.intp([(x1, y1), (x2, y2), (x3, y3), (x4, y4)])], 0, (0, 255, 0), 2)

                        delta = y3 - y1
                        for line_result in line_results:
                            cv2.putText(
                                frame, line_result, (x1, y1), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 1, cv2.LINE_AA)
                            y1 += delta

                        del location

                    elif item.get_type() == EnumCapturedResultItemType.CRIT_PARSED_RESULT:
                        mrz_result = MRZResult(item)
                        print(mrz_result.to_string())

            if cv2.waitKey(1) & 0xFF == ord('q'):
                break

            cv2.imshow('frame', frame)

        cvr.stop_capturing()
        vc.release()
        cv2.destroyAllWindows()

Explanation

• The FrameFetcher class implements the ImageSourceAdapter interface to feed frame data into the built-in buffer.
• The MyCapturedResultReceiver class implements the CapturedResultReceiver interface. The on_captured_result_received method runs on a native C++ worker thread, sending CapturedResult objects to the main thread where they are stored in a thread-safe queue for further use.
• A CapturedResult contains several CapturedResultItem objects. The CRIT_TEXT_LINE type represents recognized text lines, while the CRIT_PARSED_RESULT type represents parsed MRZ data.

Running the Real-time MRZ Recognition Demo on Windows

Source Code

https://github.com/yushulx/python-mrz-scanner-sdk/tree/main/examples/official

Das obige ist der detaillierte Inhalt vonSo implementieren Sie die Erkennung maschinenlesbarer Zonen (MRZ) in Python. Für weitere Informationen folgen Sie bitte anderen verwandten Artikeln auf der PHP chinesischen Website!