删除np数组中值为0 255的像素

Question

I am using Opencv and python 3.6 I have an image with this dimensions:

depth = depth[0:240, 96:320]

So an image of (240,224). The image is in grayscale, and I don't want consider the 2 extreme values 0 / 255 of the pixels intensity.

I already have a loop in my script. I think that I have to add a line where I take off these pixels like this:

for b in depth: #cicle already existing
    if b < 255 and b > 0:  #added line
        a = a+1
        if a <= 121 and a >= 118:
            for f in range(25, 223):
                #print((x+b)%33)
                soundArr[int((f-25)/6)] = soundArr[int((f-25)/6)]+b[f]

but I obtain the error:

The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

So how do I delete those pixels?

Answer 1

If I understand the question correctly you can use numpy.where in order to find the indices of 0 and 255 in your depth image. Then when you iterate over the image you can disregard the results from numpy.where.