Python OpenCV – drawKeypoints()函数

在本文中，让我们讨论 OpenCV 的 drawKeypoints()函数。使其脱颖而出的图像的区别品质被称为图像中的关键点。特定图像的关键点让我们能够识别物体并比较图像。可以使用多种技术和算法来检测图片中的关键点。我们利用 OpenCV 中的 drawKeypoints() 方法能够在给定图片上绘制已识别的关键点。输入图片、关键点、颜色和标志被发送到 drawKeypoints() 方法。关键点是检测中最重要的方面。即使在图像被修改后，关键点仍然保持不变。到目前为止，我们只能使用 SIRF_create()函数，因为 surf函数已获得专利。

Syntax of drawKeypoints() function:

drawKeypoints(input_image, key_points, output_image, colour, flag)

parameters:

• input_image : The image  which  is turned into grayscale and then the key points are extracted using the  SURF  or SIFT algorithms is called input image.
• key_points : The key points obtained from the input picture after using the algorithms are referred to as keypoints.
• output_image :  image on which the keypoints are drawn.
• colour : the colour of the keypoints.
•  flag : drawing features are represented by the flag.

示例 1：

此示例从导入 OpenCV 和 matplotlib 包开始。我们读取图像，将其转换为灰度，然后应用 SIRF_create() 算法来帮助我们检测图像中的关键点。 drawKeypoints()函数接受许多参数并在图像上绘制关键点。可以更改标志。在下面的示例中，我们使用 cv2.DRAW_MATCHES_FLAGS_NOT_DRAW_SINGLE_POINTS 作为标志。输出图像被绘制。除了使用 cv2.SIRF_create() ，还可以使用 cv2.xfeatures2d.SIFT_create() ，在少数版本的 OpenCV 中它可能不起作用。 cv2.xfeatures2d.SURF_create() 算法也是如此。

注意：红色为 (255,0,0)，蓝色为 (0,0,255)，绿色为 (0,255,0)。

使用的图像：

Python3

# importing packages
import cv2
import matplotlib.pyplot as plt
 
# reading image using the imread() function
imageread = cv2.imread('img1.jpeg')
 
# input image is converted to gray scale image
imagegray = cv2.cvtColor(imageread, cv2.COLOR_BGR2GRAY)
 
# using the SIRF algorithm to detect key
# points in the image
features = cv2.SIFT_create()
 
keypoints = features.detect(imagegray, None)
 
# drawKeypoints function is used to draw keypoints
output_image = cv2.drawKeypoints(imagegray, keypoints, 0, (0, 0, 255),
                                 flags=cv2.DRAW_MATCHES_FLAGS_NOT_DRAW_SINGLE_POINTS)
 
# displaying the image with keypoints as the
# output on the screen
 
plt.imshow(output_image)
 
# plotting image
plt.show()

输出：

示例 2：

此示例与上一个示例类似，只是我们将颜色更改为红色 (255,0,0) 并将标志更改为 cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS。

Python3

# importing packages
import cv2
import matplotlib.pyplot as plt
 
# reading image using the imread() function
imageread = cv2.imread('img1.jpeg')
 
# input image is converted to gray scale image
imagegray = cv2.cvtColor(imageread, cv2.COLOR_BGR2GRAY)
 
# using the SIRF algorithm to detect key
# points in the image
features = cv2.SIFT_create()
 
keypoints = features.detect(imagegray, None)
 
# drawKeypoints function is used to draw keypoints
output_image = cv2.drawKeypoints(imagegray, keypoints, 0, (255, 0, 0),
                                 flags=cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)
 
# displaying the image with keypoints as the
# output on the screen
plt.imshow(output_image)
 
# plotting image
plt.show()

输出：

示例 3：

此示例与上一个示例类似，只是我们将颜色更改为绿色 (0,255,0) 并将标志更改为 cv2.DRAW_MATCHES_FLAGS_DEFAULT。

Python3

import cv2
import matplotlib.pyplot as plt
 
# reading image using the imread() function
imageread = cv2.imread('img1.jpeg')
 
# input image is converted to gray scale image
imagegray = cv2.cvtColor(imageread, cv2.COLOR_BGR2GRAY)
 
# using the SIRF algorithm to detect key
# points in the image
features = cv2.SIFT_create()
 
keypoints = features.detect(imagegray, None)
 
# drawKeypoints function is used to draw keypoints
output_image = cv2.drawKeypoints(imagegray, keypoints, 0, (0, 255, 0),
                                 flags=cv2.DRAW_MATCHES_FLAGS_DEFAULT)
 
# displaying the image with keypoints as
# the output on the screen
plt.imshow(output_image)
 
# plotting image
plt.show()

输出：