识别图像中旋转的文本

Question

我使用EASYOCR检测图像中的文本，并使用TESSERACT进行识别，但无法检测到旋转的文本。如何检测旋转的文本？ 我使用了这段代码：

# # Text detection with easyocr and recognition with tesseract
import cv2
import easyocr
import pytesseract
pytesseract.pytesseract.tesseract_cmd = "C:\\Program Files\\Tesseract-OCR\\tesseract.exe"

# Load image
image = cv2.imread('image 18.jpg')

# Convert image to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

# Initialize the EasyOCR text detector
reader = easyocr.Reader(['en'])

# Detect text in image
results = reader.readtext(gray)

# Image preprocessing for Tesseract
_, processed_image = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)

# Recover text regions detected by EasyOCR
for (bbox, _, _) in results:
    # Extract coordinates from bounding box
    x_min, y_min = map(int, bbox[0])
    x_max, y_max = map(int, bbox[2])
    # Vérifier les limites de l'image
    x_min = max(0, x_min)
    y_min = max(0, y_min)
    x_max = min(image.shape[1], x_max)
    y_max = min(image.shape[0], y_max)
    # # Extract text region
    if x_max>x_min and y_max>y_min :
        text_region = processed_image[y_min:y_max, x_min:x_max]

        # Text recognition with Tesseract
        text = pytesseract.image_to_string(text_region, lang='eng')

        # # Draw the bounding box on the image
        cv2.rectangle(image, (x_min, y_min), (x_max, y_max), (0, 255, 0), 2)

        # Display text detected by Tesseract
        cv2.putText(image, text, (x_min, y_min - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 0, 255), 2)


# Display image with bounding boxes and detected text
cv2.imwrite('Text Detection_2.jpg', image)

结果：

Answer 1

微调 PaddelOCR

文档： https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.3/doc/doc_en/vertical_and_multilingual_datasets_en.md

笔记本 https://gist.github.com/leonbora167/049ac6622b7a2fb5c23ec48070af486f