Mini -Project #3-1 명함 인식 프로그램 만들기(pyside6,openai,opencv)

 

단순 동기로 시작함
리멤버 앱 사용하는데 python으로 명함 인식 하는 프로그램 만드는게 가능하지 않을까로 시작
pyside6, opencv를 공부하면서 작성
결국 openai가 다 만듬 끝(업그레이트 지속 예정)
주의 프로젝트 폴더명이 한글이면 잘 되어서 중간에 영어이름으로 변경함

https://github.com/carcase2/-Python_Mini_businesscard

GitHub - carcase2/-Python_Mini_businesscard: 명함 인식하는것 만들어보자

import sys
import cv2
import pytesseract
from PySide6.QtWidgets import QApplication, QLabel, QVBoxLayout, QWidget, QPushButton, QFileDialog
from PySide6.QtGui import QImage, QPixmap

# Ensure you have the correct Tesseract executable path
pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'

def read_business_card(image_path):
    # Load the image
    image = cv2.imread(image_path)
    
    if image is None:
       print("Error: Could not load the image. Please check the file path.")
       return ""

    # Preprocess the image
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    _, threshold = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)

    # Extract text using OCR
    text = pytesseract.image_to_string(threshold, lang='kor')

    return text

class BusinessCardReader(QWidget):
    def __init__(self):
        super().__init__()

        self.label = QLabel()
        self.label.setWordWrap(True)

        self.button = QPushButton('Open Image')
        self.button.clicked.connect(self.load_image)

        layout = QVBoxLayout()
        layout.addWidget(self.button)
        layout.addWidget(self.label)

        self.setLayout(layout)
        self.setWindowTitle('Business Card Reader')

    def load_image(self):
        options = QFileDialog.Options()
        options |= QFileDialog.ReadOnly
        image_path, _ = QFileDialog.getOpenFileName(self, "Open Image", "", "Images (*.png *.xpm *.jpg *.bmp *.jpeg)", options=options)
        
        if image_path:
            text = read_business_card(image_path)
            self.label.setText(text)

def main():
    app = QApplication(sys.argv)
    window = BusinessCardReader()
    window.show()
    sys.exit(app.exec())

if __name__ == '__main__':
    main()

실행화면

샘플 명함

 

---

결과

한글로 인식 했을떄와 영어로 인식 했을때

---

과정

1-1. 시작

1-2. 결과

import sys
import cv2
import pytesseract
from PySide6.QtWidgets import QApplication, QLabel, QVBoxLayout, QWidget
from PySide6.QtGui import QImage, QPixmap

# Ensure you have the correct Tesseract executable path
pytesseract.pytesseract.tesseract_cmd = r'/path/to/tesseract.exe'

def read_business_card(image_path):
    # Load the image
    image = cv2.imread(image_path)

    # Preprocess the image
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    _, threshold = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)

    # Extract text using OCR
    text = pytesseract.image_to_string(threshold)

    return text

def main():
    app = QApplication(sys.argv)

    # Load the business card image
    image_path = 'path/to/business_card.jpg'
    text = read_business_card(image_path)

    # Display the extracted text using PySide6
    label = QLabel()
    label.setText(text)
    label.setWordWrap(True)

    layout = QVBoxLayout()
    layout.addWidget(label)

    window = QWidget()
    window.setLayout(layout)
    window.setWindowTitle('Business Card Reader')
    window.show()

    sys.exit(app.exec())

if __name__ == '__main__':
    main()

2-1 image_path를 어떻게 넣지를 몰라서 다시 물어봄

2-2 이미지를 불어오는 단추를 만들어서 불러오고 싶음

import sys
import cv2
import pytesseract
from PySide6.QtWidgets import QApplication, QLabel, QVBoxLayout, QWidget, QPushButton, QFileDialog
from PySide6.QtGui import QImage, QPixmap

# Ensure you have the correct Tesseract executable path
pytesseract.pytesseract.tesseract_cmd = r'/path/to/tesseract.exe'

def read_business_card(image_path):
    # Load the image
    image = cv2.imread(image_path)

    # Preprocess the image
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    _, threshold = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)

    # Extract text using OCR
    text = pytesseract.image_to_string(threshold)

    return text

class BusinessCardReader(QWidget):
    def __init__(self):
        super().__init__()

        self.label = QLabel()
        self.label.setWordWrap(True)

        self.button = QPushButton('Open Image')
        self.button.clicked.connect(self.load_image)

        layout = QVBoxLayout()
        layout.addWidget(self.button)
        layout.addWidget(self.label)

        self.setLayout(layout)
        self.setWindowTitle('Business Card Reader')

    def load_image(self):
        options = QFileDialog.Options()
        options |= QFileDialog.ReadOnly
        image_path, _ = QFileDialog.getOpenFileName(self, "Open Image", "", "Images (*.png *.xpm *.jpg *.bmp *.jpeg)", options=options)
        
        if image_path:
            text = read_business_card(image_path)
            self.label.setText(text)

def main():
    app = QApplication(sys.argv)
    window = BusinessCardReader()
    window.show()
    sys.exit(app.exec())

if __name__ == '__main__':
    main()

3-1 예외가 발생했습니다. ModuleNotFoundError No module named 'cv2' 발생함) 

- cv2가 설치가 안되어서 발생

4-1 pytesseract.pytesseract.TesseractNotFoundError: /path/to/tesseract.exe is not installed or it's not in your PATH. See README file for more information. 

- tessearct 설치 했는데 왜 안되지?

'/path/to/tesseract.exe'  라인을 그대로 적어서 문제가된 실제 teeseract 설치된 경로를 넣어줘야 하는거었음

5-1 한글이 인식이 안되어서 문제가됨

- 한글 데이타가 없어서 문제가 됨(설치하면 끝)

https://github.com/tesseract-ocr/tessdata_best/blob/main/kor.traineddata

 

def read_business_card(image_path):
    # Load the image
    image = cv2.imread(image_path)

    # Check if the image is loaded successfully
    if image is None:
        print("Error: Could not load the image. Please check the file path.")
        return ""

    # Preprocess the image
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    _, threshold = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)

    # Extract text using OCR with the Korean language setting
    text = pytesseract.image_to_string(threshold, lang='kor')

    return text

---

 

이미지 처리 하는 법을 더 공부해서 명함 인식 확율을 높여 봐야 겠다.