图像处理的二进制阈值

Question

I am new to python and I learning image processing. A very basic concept of Thresholding is causing some confusion. I understand that it can be implemented as:

if AvgPix > T:
   pix  = 255
else:
   pix  = 0

I found this code to calculate the threshold value for the image-

for eachrow in imagearray:
        for eachpix in eachrow:
            avg = reduce(lambda x,y: x + y ,eachpix[:3]) / len(eachpix[:3])
            balanceAr.append(avg)
threshold  = reduce(lambda x,y: x + y ,balanceAr) / len(balanceAr) 

I found that when eachpix = array([239, 228, 176, 255], dtype=uint8) , the avg getting calculated is 131 as per this code. Shouldn't it be = (239+228+176)/3 = 214.3

Please help me in understanding this. If I am missing something!

Answer 1

I wasn't able to reproduce the problem:

from functools import reduce

pix = [239, 228, 176, 255]

imagearray = [
    [pix, pix, pix],
    [pix, pix, pix],
    [pix, pix, pix]
]

balanceAr = []

for eachrow in imagearray: 
    for eachpix in eachrow: 
        avg = reduce(lambda x,y: x + y, eachpix[:3]) / len(eachpix[:3])
        balanceAr.append(avg) 

threshold = reduce(lambda x,y: x + y, balanceAr) / len(balanceAr)

print(threshold) # 214.3

And by the way, you can get an average of a list without reduce() :

avg = sum(my_list) / len(my_list)

So it can be boiled down to this:

pix = [239, 228, 176, 255]

imagearray = [
    [pix, pix, pix],
    [pix, pix, pix],
    [pix, pix, pix]
]

def avg(lst):
    return sum(lst) / len(lst)

threshold = avg( [avg([avg(px[:3]) for px in row]) for row in imagearray] ) 

print(threshold) # 214.3