python 批量裁剪文档图片

批量把大于400像素的文档图片裁剪矫正。

请确保你已经安装了OpenCV库。如果没有，请使用以下命令安装：

pip install opencv-python-headless

import cv2
import numpy as np
import os

def order_points(pts):
    # 初始化坐标点，顺序为左上，右上，右下，左下
    rect = np.zeros((4, 2), dtype="float32")

    # 左上点有最小的和，右下点有最大的和
    s = pts.sum(axis=1)
    rect[0] = pts[np.argmin(s)]
    rect[2] = pts[np.argmax(s)]

    # 计算点的差，右上点有最小的差，左下点有最大的差
    diff = np.diff(pts, axis=1)
    rect[1] = pts[np.argmin(diff)]
    rect[3] = pts[np.argmax(diff)]

    return rect

def four_point_transform(image, pts):
    # 获取坐标点，并将它们分成左上，右上，右下，左下
    rect = order_points(pts)
    (tl, tr, br, bl) = rect

    # 计算输入的w和h值
    widthA = np.sqrt(((br[0] - bl[0]) ** 2) + ((br[1] - bl[1]) ** 2))
    widthB = np.sqrt(((tr[0] - tl[0]) ** 2) + ((tr[1] - tl[1]) ** 2))
    maxWidth = max(int(widthA), int(widthB))

    heightA = np.sqrt(((tr[0] - br[0]) ** 2) + ((tr[1] - br[1]) ** 2))
    heightB = np.sqrt(((tl[0] - bl[0]) ** 2) + ((tl[1] - bl[1]) ** 2))
    maxHeight = max(int(heightA), int(heightB))

    # 透视变换的目标点
    dst = np.array([
        [0, 0],
        [maxWidth - 1, 0],
        [maxWidth - 1, maxHeight - 1],
        [0, maxHeight - 1]], dtype="float32")

    # 计算变换矩阵并应用它
    M = cv2.getPerspectiveTransform(rect, dst)
    warped = cv2.warpPerspective(image, M, (maxWidth, maxHeight))

    return warped
	
def crop_and_warp_image(image_path):
    image = cv2.imread(image_path)
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    edges = cv2.Canny(gray, threshold1=100, threshold2=200)
    contours, _ = cv2.findContours(edges, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    
    if not contours:
        return None

    max_contour = max(contours, key=cv2.contourArea)
    x, y, w, h = cv2.boundingRect(max_contour)

    if w > 400 and h > 400:
        # 这里假设max_contour是矩形的近似轮廓
        epsilon = 0.05 * cv2.arcLength(max_contour, True)
        approx = cv2.approxPolyDP(max_contour, epsilon, True)
        
        if len(approx) == 4:
            # 转换为四点坐标
            pts = approx.reshape(4, 2)
            warped = four_point_transform(image, pts)
            return warped
        else:
            print(f"Cannot find four corners for {image_path}")

    return None

cut_dir = 'cut'
os.makedirs(cut_dir, exist_ok=True)
image_files = [f for f in os.listdir('.') if f.lower().endswith(('.png', '.jpg', '.jpeg', '.tiff', '.bmp', '.gif'))]

for image_file in image_files:
    warped_image = crop_and_warp_image(image_file)
    if warped_image is not None:
        cv2.imwrite(os.path.join(cut_dir, f'cut_{image_file}'), warped_image)