How to remove a polyline from an RGB image using numpy

Question

I am not sure what would be the best way to remove a vertical polyline from an image. This line starts from the very top of an image to the bottom. The input is an image with three channels and array of y indices of the line . The output would be the same input image but without the polyline. For example, if the first image shape is (5, 10,3) , then the final image shape should be (5, 9, 3) after removing the polyline.

Input example:

Image :

np.array([[[0, 20, 0, 0, 255, 0, 0, 0, 0, 0],
           [0, 20, 0, 0, 255, 0, 0, 0, 0, 0],
           [0, 56, 0, 0, 0, 255, 0, 0, 0, 0],
           [0, 58, 0, 0, 0, 255, 0, 0, 0, 0],
           [0, 0, 0, 0, 0, 0, 255, 0, 0, 0]],
          [[0, 0, 0, 0, 255, 0, 0, 0, 30, 0],
           [0, 0, 0, 0, 255, 0, 0, 0, 31, 0],
           [0, 0, 0, 0, 0, 255, 0, 0, 32, 0],
           [0, 0, 0, 0, 0, 255, 0, 0, 31, 0],
           [0, 0, 0, 0, 0, 0, 255, 0, 0, 0]],
          [[1, 0, 0, 0, 255, 0, 0, 0, 0, 0],
           [1, 0, 0, 0, 255, 0, 0, 0, 0, 0],
           [1, 0, 0, 0, 0, 255, 0, 0, 0, 0],
           [1, 0, 0, 0, 0, 255, 0, 0, 0, 0],
           [1, 0, 0, 0, 0, 0, 255, 0, 0, 0]]])

Note: the size here is (3, 5, 10) for illustration. The image shape would be (5, 10, 3). We wanna remove the white polyline

Polyline :

np.array([4, 4, 5, 5, 6])

Output Example

np.array([[[0, 20, 0, 0, 0, 0, 0, 0, 0],
           [0, 20, 0, 0, 0, 0, 0, 0, 0],
           [0, 56, 0, 0, 0, 0, 0, 0, 0],
           [0, 58, 0, 0, 0, 0, 0, 0, 0],
           [0, 0, 0, 0, 0, 0, 0, 0, 0]],
          [[0, 0, 0, 0, 0, 0, 0, 30, 0],
           [0, 0, 0, 0, 0, 0, 0, 31, 0],
           [0, 0, 0, 0, 0, 0, 0, 32, 0],
           [0, 0, 0, 0, 0, 0, 0, 31, 0],
           [0, 0, 0, 0, 0, 0, 0, 0, 0]],
          [[1, 0, 0, 0, 0, 0, 0, 0, 0],
           [1, 0, 0, 0, 0, 0, 0, 0, 0],
           [1, 0, 0, 0, 0, 0, 0, 0, 0],
           [1, 0, 0, 0, 0, 0, 0, 0, 0],
           [1, 0, 0, 0, 0, 0, 0, 0, 0]]])

Note: the size now is (3, 5, 9)

My code

def remove_polyline(image, polyline):
    h, w = image.shape[:2]
    res_image = np.zeros((h, w - 1))

    for i in range(h):
        left_section = image[i, :polyline[i]]
        right_section = image[i, polyline[i]+1:]
        res_image[i, :] = np.dstack((left_section, right_section))

    return res_image

I know this would work for grayscale images but not sure about 3 channel images

Answer 1

Note that your input has shape (3,5,10) instead of (5,10,3) . This answer assumes the latter:

def remove_polyline(s, polyline):
    mask = np.arange(s.shape[1]) != polyline[ :, None]
    return s[mask,:].reshape(s[:,:-1].shape).copy()

r = remove_polyline(image, polyline)
plt.imshow(r)

Output:

Answer 2

You can try this.

def remove_polyline(image, polyline):
    h, w, c = image.shape
    res_image = np.zeros((h, w - 1, c))

    for i in range(h):
        left_section = image[i, :polyline[i], :]
        right_section = image[i, polyline[i]+1:, :]
        res_image[i, :, :] = np.vstack((left_section, right_section))

    return res_image