'''
"doc_scanner.py" 文件中的 DocScanner 类提供了文档扫描的核心功能，功能包括
    - 加载图片
    - 找到文档的角落
    - 裁剪图片

- 通过将文档扫描功能封装在一个独立的类中，我们可以将其作为一个模块，方便在其他项目中重用。
- 这种模块化的设计使得代码更加模块化、可扩展和可测试。
- 它可以独立于其他部分进行开发、测试和维护，而不会对其他组件产生影响。
- 此外，将文档扫描功能单独放在一个文件中，使得代码结构更清晰、易于理解和维护。
- 通过将不同的功能分割到不同的文件中，我们可以更好地组织代码，减少文件的复杂性，并促进团队合作开发。
'''
import cv2 as cv
import numpy as np

class DocScanner:
    def __init__(self):
        pass

    def load_image(self, file_path):
        """加载图片并找到文档的四个角落。"""
        img = cv.imread(file_path)
        img_gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
        thresh, binary_img = cv.threshold(img_gray, 127, 255, cv.THRESH_BINARY)
        contours, hierarchy = cv.findContours(binary_img, cv.RETR_EXTERNAL, cv.CHAIN_APPROX_SIMPLE)
        
        max_area = 0
        extreme_pnts = None
        for contour in contours:
            peri = cv.arcLength(contour, True)
            approx = cv.approxPolyDP(contour, 0.02 * peri, True)
            if len(approx) == 4:
                area = cv.contourArea(approx)
                if area > max_area:
                    max_area = area
                    extreme_pnts = approx
        
        corners = self.order_points(extreme_pnts.reshape(4, 2))
        return img, corners

    def order_points(self, pts):
        """对给定的四个点进行排序，返回排序后的点。"""
        rect = np.zeros((4, 2), dtype="float32")
        s = pts.sum(axis=1)
        rect[0] = pts[np.argmin(s)]
        rect[2] = pts[np.argmax(s)]
        diff = np.diff(pts, axis=1)
        rect[1] = pts[np.argmin(diff)]
        rect[3] = pts[np.argmax(diff)]

        print(f"0:左上角:{rect[0]},因为x,y的坐标和是: np.min(s)={np.min(s)}")
        print(f"1:右上角:{rect[1]},因为x,y的坐标差是: np.min(diff)={np.min(diff)}")
        print(f"2:右下角:{rect[2]},因为x,y的坐标和是: np.max(s)={np.max(s)}")
        print(f"3:左下角:{rect[3]},因为x,y的坐标差是: np.max(diff)={np.max(diff)}")

        return rect

    def crop_image(self, img, corners):
        """根据四个角落裁剪图片。"""
        top_left_corner = corners[0]
        top_right_corner = corners[1]
        bottom_right_corner = corners[2]
        bottom_left_corner = corners[3]

        width, height = self.get_image_dimensions((top_left_corner, top_right_corner, bottom_right_corner, bottom_left_corner))

        dst_points = np.array([
            [0, 0],
            [width - 1, 0],
            [width - 1, height - 1],
            [0, height - 1]
        ], dtype="float32")
            
        matrix = cv.getPerspectiveTransform(corners.astype("float32"), dst_points)
        dst = cv.warpPerspective(img, matrix, (width, height))
        return dst

    def get_image_dimensions(self, corners):
        """计算图片的宽度和高度。"""
        top_left_corner, top_right_corner, bottom_right_corner, bottom_left_corner = corners
        
        top_width = np.linalg.norm(top_right_corner - top_left_corner)
        bottom_width = np.linalg.norm(bottom_right_corner - bottom_left_corner)
        
        left_height = np.linalg.norm(bottom_left_corner - top_left_corner)
        right_height = np.linalg.norm(bottom_right_corner - top_right_corner)
        
        width = int(max(top_width, bottom_width))
        height = int(max(left_height, right_height))
        
        return width, height