When I load grayscale, I use cv2.imread and give cv2.IMREAD_GRAYSCALE as the flag.
I just wondered what will happen if I train the model by color image from gray scale.
I Tried, and the training ended successfully.
Now, I’ve got something to ask.
What difference is between training with original grayscale image and color image converted to grayscale?
Is it better to train grayscale model with original grayscale images than to train color model with color images converted to grayscale?