opencv 求轮廓、外接矩形、凸包等-06

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
import cv2
import numpy as np


def cal_contours(image, thresh_value=0, mode=cv2.RETR_TREE, method=cv2.CHAIN_APPROX_SIMPLE):
    """
    :param image: 单通道图像, 8-bit or 32-bit floating point
    :param thresh_value: 二值化单通道图像
    :param mode: Contour retrieval mode, eg
    cv2.RETR_EXTERNAL：表示只提取最外面的轮廓; cv2.RETR_LIST：表示提取所有轮廓并将其放入列表;
    cv2.RETR_CCOMP:表示提取所有轮廓并将组织成一个两层结构，其中顶层轮廓是外部轮廓，第二层轮廓是“洞”的轮廓;
    cv2.RETR_TREE：表示提取所有轮廓并组织成轮廓嵌套的完整层级结构
    :param method:　Contour approximation method, eg
    cv2.CHAIN_APPROX_NONE：将轮廓中的所有点的编码转换成点;
    cv2.CHAIN_APPROX_SIMPLE：压缩水平、垂直和对角直线段，仅保留它们的端点
    :return:
    """
    cv_major_version = int(cv2.__version__.split('.')[0])  # 获取当前版本
    # multiple-channel, 8-bit or 32-bit floating point
    thresh_value, thresh_img = cv2.threshold(image, thresh_value, 255, cv2.THRESH_BINARY)

    if cv_major_version >= 4:
        # OpenCV 4 or a later version is being used.
        # an 8-bit single-channel image. Non-zero pixels are treated as 1's.
        contours, hier = cv2.findContours(thresh_img, mode, method)
    else:
        # OpenCV 3 or an earlier version is being used.
        # cv2.findContours has an extra return value.
        # The extra return value is the thresholded image, which is
        # unchanged, so we can ignore it.
        _, contours, hier = cv2.findContours(thresh_img, mode, method)
    return contours


if __name__ == '__main__':
    img = np.zeros((200, 200), dtype=np.uint8)
    img[50:150, 50:150] = 255

    cnts = cal_contours(img)
    color_img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
    img = cv2.drawContours(color_img, cnts, -1, (0, 255, 0), 2)
    cv2.imshow("contours", color_img), cv2.waitKey(), cv2.destroyAllWindows()

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
import cv2
import numpy as np

from chapter03.contours import cal_contours


def cal_min_area_rect(contour):
    """
    找到最小的带旋转的外接矩形
    :param contour:
    :return:
    """
    rect = cv2.minAreaRect(contour)  # Finds a rotated rectangle of the minimum area enclosing the input 2D point set.
    # calculate coordinates of the minimum area rectangle
    box = cv2.boxPoints(rect)  # Finds the four vertices of a rotated rect.
    # normalize coordinates to integers
    box = np.int0(box)  # numpy.int64. four vertices
    return box


if __name__ == '__main__':
    img = cv2.pyrDown(cv2.imread("hammer.jpg", cv2.IMREAD_UNCHANGED))
    gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    contours = cal_contours(gray_img, thresh_value=127, mode=cv2.RETR_EXTERNAL, method=cv2.CHAIN_APPROX_SIMPLE)
    for c in contours:
        # 1, find bounding box coordinates
        x, y, w, h = cv2.boundingRect(c)  # 轮廓的外接矩形
        cv2.rectangle(img, (x, y), (x + w, y + h), (0, 255, 0), 3)  # 没有旋转的外接矩形，绿色

        # 2, 找到最小的带旋转的外接矩形
        box = cal_min_area_rect(c)
        # draw contours
        cv2.drawContours(img, [box], 0, (0, 0, 255), 3)

        # 3, calculate center and radius of minimum enclosing circle
        (x, y), radius = cv2.minEnclosingCircle(c)
        # cast to integers
        center = (int(x), int(y))
        radius = int(radius)
        # draw the circle
        img = cv2.circle(img, center, radius, (0, 255, 255), 3)

    cv2.drawContours(img, contours, -1, (255, 0, 0), 3)
    cv2.imshow("contours", img)

cv2.waitKey()
cv2.destroyAllWindows()

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
import cv2
import numpy as np

from chapter03.contours import cal_contours

if __name__ == '__main__':
    img = cv2.pyrDown(cv2.imread("hammer.jpg", cv2.IMREAD_UNCHANGED))
    gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    contours = cal_contours(gray_img, 127, mode=cv2.RETR_EXTERNAL, method=cv2.CHAIN_APPROX_SIMPLE)
    for cnt in contours:  # 只有一个
        # 1, 轮廓，蓝色
        cv2.drawContours(img, [cnt], -1, (255, 0, 0), 3)
        # 2, 多边形，黄色
        epsilon = 0.01 * cv2.arcLength(cnt, True)  # 0.01 * 周长
        approx = cv2.approxPolyDP(cnt, epsilon, True)  # 以指定的精度近似多边形曲线。 (num, 1, 2)
        cv2.drawContours(img, [approx], -1, (0, 255, 255), 3)
        # 3, 凸包，红色
        hull = cv2.convexHull(cnt)  # 轮廓的凸包。(num1, 1, 2)
        cv2.drawContours(img, [hull], -1, (0, 0, 255), 3)

    cv2.imshow("hull", img)
    cv2.waitKey()
    cv2.destroyAllWindows()