I believe anyone who has been in touch with Java knows the track bar. It is called a slider control in Java. Today we will also implement it in OpenCV as a basic study.
Here, we will create a simple application to display the specified color. We will create a window to display colors, and three tracking bars for specifying B, G, and R colors. Slide the track bar and change the window color accordingly. By default, the initial color will be set to black.
We need to understand several functions:
createTrackbar(const string& trackbarname, const string&winname, int* value, int count ，TrackbarCallback onChange = 0, void* userdata = 0)
For the cv.createTrackbar() function, its parameters are defined as:
Parameter 1: Track bar name
Parameter 2: Window name
Parameter 3: Initial position of the slider
Parameter 4: the value that represents the maximum position of the slider
Parameter 5: The default value is 0, which points to the callback function
Parameter 6: The default value is 0, the data value passed by the user to the callback function
getTrackbarPos(const string& trackbarname, const string& winname)
Function role:
Get the value of the position of the slider
Parameter 1: progress bar name
Parameter 2: Window name
Return value: progress bar position
getTrackbarPos（ trackbarname，winname ）
Parameter definition:
trackbarname-the name of the trackbar.
winname-the name of the parent window of the track bar.
This function returns the current position of the specified track bar.
Another important application of the track bar is to use it as a button or switch. By default, OpenCV does not have button functions. Therefore, we can use the track bar to get such functions. In the application, we created a switch, the application can only run when the switch is ON, otherwise the screen is always black.
Now look at the code:
view plaincopy to clipboardprint?
1. import numpy as np
2. import cv2 as cv
3. def nothing(x):
4. pass
5. # Create a black image, a window
6. img = np.zeros((300,512,3), np.uint8)
7. cv.namedWindow('image')
8. # create trackbars for color change
9. cv.createTrackbar('R','image',0,255,nothing)
10.cv.createTrackbar('G','image',0,255,nothing)
11.cv.createTrackbar('B','image',0,255,nothing)
12.# create switch for ON/OFF functionality
13.switch = '0 : OFF \n1 : ON'
14.cv.createTrackbar(switch, 'image',0,1,nothing)
15.while(1):
16. cv.imshow('image',img)
17. k = cv.waitKey(1) & 0xFF
18. if k == 27:
19. break
20. # get current positions of four trackbars
21. r = cv.getTrackbarPos('R','image')
22. g = cv.getTrackbarPos('G','image')
23. b = cv.getTrackbarPos('B','image')
24. s = cv.getTrackbarPos(switch,'image')
25. if s == 0:
26. img[:] = 0
27. else:
28. img[:] = [b,g,r]
29.cv.destroyAllWindows()
We can see the effect:
Of course, this is only a color palette that can be adjusted. We have already talked about the function of using the mouse as a brush in OpenCV. Now let's create a drawing board and use the track bar to change the color of the brush, similar to Our computer's "drawing" software is the same.
In the previous tutorial, we used EVENT_FLAG_LBUTTON to indicate the current left button of the mouse is pressed, EVENT_MOUSEMOVE to indicate the mouse movement, and cv2.circle as the brush. In the last tutorial, our default pen color is red, and the penultimate parameter is the color representation, the code is:
cv2.circle(img, (x, y), 1, [0, 0, 255], 1)
But in this tutorial, in order to change the color of the brush at any time, we need to pass in dynamic variables;
cv2.circle(img, (x, y), 1, [b, g, r], 1)
b, g, r are determined by the track bar:
cv2.createTrackbar(‘R’, ‘image’, 0, 255, callback)
cv2.createTrackbar(‘B’, ‘image’, 0, 255, callback)
cv2.createTrackbar(‘G’, ‘image’, 0, 255, callback)
We must also note that for OpenCV, its color format is BGR, not RGB, so we need to pass in the parameters correctly.
Now we look at the code:
view plaincopy to clipboardprint?
1. import cv2
2. import numpy as np
3.
4. def callback(object):
5. pass
6.
7. img = np.zeros((400, 400, 3), np.uint8)
cv2.namedWindow('image')
8.
9. switch = '0:OFF\n1:ON'
10.cv2.createTrackbar(switch, 'image', 0, 1, callback)
11.
12.cv2.createTrackbar('R', 'image', 0, 255, callback)
13.cv2.createTrackbar('B', 'image', 0, 255, callback)
14.cv2.createTrackbar('G', 'image', 0, 255, callback)
15.
16.def Mouseback(event, x, y, flags, param):
17. if flags == cv2.EVENT_FLAG_LBUTTON and event == cv2.EVENT_MOUSEMOVE:
18. cv2.circle(img, (x, y), 1, [b, g, r], 1)
19.
20.img[:] = [255, 255, 255]
21.
22.while(True):
23. r = cv2.getTrackbarPos('R', 'image')
24. g = cv2.getTrackbarPos('G', 'image')
25. b = cv2.getTrackbarPos('B', 'image')
26. if cv2.getTrackbarPos(switch, 'image') == 1:
27. cv2.setMouseCallback('image', Mouseback)
28. else:
29. img[:] = [255, 255, 255]
30. if cv2.waitKey(10) & 0xFF == ord('q'):
31. break
32. cv2.imshow('image', img)
33.cv2.destroyAllWindows()
Experimental effect:
Press the Q key to exit, we can see that the color of the brush can be changed at any time.
Now let’s perform some more advanced operations. We will completely combine the code from the previous tutorial with the track bar. In the previous tutorial, we created a drawing board and used the Q key to switch brushes. The default is to draw a rectangle. Press Lower Q is a line drawing. Now we will change the color of the brush to complete more advanced operations. The thin drag bar can be used to adjust the thickness of the line, we still pass in its parameters:
cv2.circle(img, (x, y), thin, color, -1)
At the same time, for convenience, we also add the eraser function, which actually clears the drawing board to white.
Let's look at the code:
view plaincopy to clipboardprint?
1. import cv2
2. import numpy as np
3.
4. drawing = False
5. mode = True
6. ix, iy = -1, -1
7. def nothing(x):
8. pass
9.
10.def draw_circle(event, x, y, flags, param):
11. r = cv2.getTrackbarPos('R', 'image')
12. g = cv2.getTrackbarPos('G', 'image')
13. b = cv2.getTrackbarPos('B', 'image')
14. color = (b, g, r)
15. s = cv2.getTrackbarPos('eraser', 'image')
16. if s == 1:
17. color = (255, 255, 255)
18. thin = cv2.getTrackbarPos('thin', 'image')
19.
20. global ix, iy, drawing, mode
21. if event == cv2.EVENT_LBUTTONDOWN:
22. drawing = True
23. ix, iy = x, y
24. elif event == cv2.EVENT_MOUSEMOVE and flags == cv2.EVENT_FLAG_LBUTTON:
25. if drawing:
26. if mode:
27. cv2.rectangle(img, (ix, iy), (x, y), color, -1)
28. else:
29. cv2.circle(img, (x, y), thin, color, -1)
30. elif event == cv2.EVENT_LBUTTONUP:
31. drawing == False
32.
33.img = np.zeros((512, 512, 3), np.uint8)
34.img[:] = 255
35.cv2.namedWindow('image')
36.cv2.createTrackbar('R', 'image', 0, 255, nothing)
37.cv2.createTrackbar('G', 'image', 0, 255, nothing)
38.cv2.createTrackbar('B', 'image', 0, 255, nothing)
39.cv2.createTrackbar('eraser', 'image', 0, 1, nothing)
40.cv2.createTrackbar('thin', 'image', 1, 50, nothing)
41.
42.cv2.setMouseCallback('image', draw_circle)
43.
44.while (1):
45. cv2.imshow('image', img)
46. k = cv2.waitKey(1) & 0xFF
47. if k == ord('q'):
48. mode = not mode
49. elif k == 27:
50. break
Experimental effect:
Isn't it pretty? Press esc to exit, press q to switch between drawing rectangles and lines. We can also simply complete a drawing operation.
So this tutorial is over here, you still have to practice more usually, not as good as practice.
Check the article summary page https://blog.csdn.net/weixin_44237705/article/details/107864965
More openvino technical information can be exchanged in the group~
