扫描模板纠偏

caoyd

扫描的时候,经常会出现偏移或者旋转,就算有纠偏效果也不好.看到一些软件,用第一张图片做模板,旋转模板中的红色区域作为区域模板,后面的图片都匹配区域模板,计算出平移量和旋转量.哪位大神能够出出力.感谢万分.

caoyd

KServer

建议 python opencv

[Python] 纯文本查看 复制代码

import cv2
import numpy as np

def deskew_image(image):
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    _, binary = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)
    contours, _ = cv2.findContours(binary, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    
    contour = max(contours, key=cv2.contourArea)
    rect = cv2.minAreaRect(contour)
    angle = rect[-1]
    
    if angle < -45:
        angle += 90
    
    (h, w) = image.shape[:2]
    center = (w // 2, h // 2)
    M = cv2.getRotationMatrix2D(center, angle, 1.0)
    rotated = cv2.warpAffine(image, M, (w, h), flags=cv2.INTER_CUBIC, borderMode=cv2.BORDER_REPLICATE)
    
    return rotated

# 读取原始图像
image = cv2.imread('scan_template.png')

# 纠偏图像
deskewed_image = deskew_image(image)

# 显示原始图像和纠偏后的图像
cv2.imshow("Original Image", image)
cv2.imshow("Deskewed Image", deskewed_image)
cv2.waitKey(0)
cv2.destroyAllWindows()