使用

Question

我想从这张图片中的表中提取数据，我使用 cv2 和 pytesseract 但我没有得到可靠的结果。这是我的代码和我的图像。

在此输入图片描述

import cv2
import pytesseract
from PIL import Image

def preprocess_image(image_path):
    # Load the image using OpenCV
    img = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
    
    # Apply thresholding (binarization)
    _, thresh = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    
    # Invert the image colors
    inverted_img = cv2.bitwise_not(thresh)
    
    # Convert the thresholded and inverted image back to PIL image format
    pil_img = Image.fromarray(inverted_img)
    
    return pil_img

def image_to_text(path):
    # Path to your Tesseract executable
    pytesseract.pytesseract.tesseract_cmd = r"C:\Program Files\Tesseract-OCR\tesseract.exe"
    
    # Preprocess the image
    img = preprocess_image(path)
    
    # Perform OCR using Tesseract
    text = pytesseract.image_to_string(img, config='--psm 4 --oem 2 ', lang='rus')
    
    return text


print(image_to_text('concatenated_image3.jpg'))

Answer 1

要使用 Tesseract OCR 检测并打印图像中的文本，您可以使用 Python 和 OpenCV 进行图像预处理，并使用 Pytesseract 进行 OCR。

import cv2
import pytesseract
import re

class TextDetector:
    def __init__(self, image_path, language='rus'):
        self.image_path = image_path
        self.language = language
        self.text_results = None

    def preprocess_image(self):
        img = cv2.imread(self.image_path)
        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
        _, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
        return thresh

    def detect_text_boxes(self):
        img = self.preprocess_image()
        results = pytesseract.image_to_data(img, output_type=pytesseract.Output.DICT, lang=self.language)

        img_height, img_width = img.shape[:2]
        text_data = []
        for i in range(len(results['text'])):
            x, y, w, h = results['left'][i], results['top'][i], results['width'][i], results['height'][i]
            text = results['text'][i]

            # Filter out non-alphanumeric characters using regex
            filtered_text = re.sub(r'[^A-Za-z0-9А-Яа-я]', '', text)

            # Only consider alphanumeric text
            if filtered_text.isalnum():
                text_data.append({'Text': filtered_text, 'Bounding Box': (x, y, w, h)})

        self.text_results = text_data

    def get_text_results(self):
        if self.text_results is None:
            self.detect_text_boxes()
        return self.text_results


if __name__ == "__main__":
    detector = TextDetector('KnPrd3fG.png')
    text_results = detector.get_text_results()

    # Print detected text
    for result in text_results:
        print(f"Detected Text: {result['Text']}")

以下是如何使用 Pandas 将检测到的文本保存到 CSV 文件中：

import pandas as pd

class TextProcessor:
    def __init__(self, text_detector):
        self.text_detector = text_detector
        self.text_data = None

    def process_text(self):
        text_results = self.text_detector.get_text_results()
        if text_results:
            df = pd.DataFrame(text_results)
            self.text_data = df

    def save_to_csv(self, file_path):
        if self.text_data is None:
            self.process_text()
        if self.text_data is not None:
            self.text_data.to_csv(file_path, index=False)
            print(f"Data saved to '{file_path}' successfully.")
        else:
            print("No text data to save. Please run 'process_text' first.")


if __name__ == "__main__":
    detector = TextDetector('KnPrd3fG.png')
    processor = TextProcessor(detector)
    processor.save_to_csv('text_data.csv')