python - numpy.logic_and 代码示例中的解释

Question

我正在尝试重现此代码 example ，但是运行代码时出现异常:

import cv2
import numpy as np


def find_image(im, tpl):
    im = np.atleast_3d(im)
    tpl = np.atleast_3d(tpl)
    H, W, D = im.shape[:3]
    h, w = tpl.shape[:2]

    # Integral image and template sum per channel
    sat = im.cumsum(1).cumsum(0)
    tplsum = np.array([tpl[:, :, i].sum() for i in range(D)])

    # Calculate lookup table for all the possible windows
    iA, iB, iC, iD = sat[:-h, :-w], sat[:-h, w:], sat[h:, :-w], sat[h:, w:]
    lookup = iD - iB - iC + iA
    # Possible matches
    np.logical_and(True, False)
    npnd= np.logical_and(*[lookup[..., i] == tplsum[i] for i in range(D)])
    possible_match = np.where(npnd)

    # Find exact match
    for y, x in zip(*possible_match):
        if np.all(im[y+1:y+h+1, x+1:x+w+1] == tpl):
            return (y+1, x+1)

    raise Exception("Image not found")

img1 = cv2.imread('c:/temp/1.png',0)
img2 = cv2.imread('c:/temp/3.png',0)
tplImg = cv2.imread('c:/temp/2.png',0)

find_image(img1, tplImg)

异常发生在这一行:

npnd= np.logical_and(*[lookup[..., i] == tplsum[i] for i in range(D)])

它失败并显示以下异常消息:

C:\Python36-32>python.exe cvtest.py
Traceback (most recent call last):
  File "cvtest.py", line 33, in <module>
    find_image(img1, tplImg)
  File "cvtest.py", line 19, in find_image
    npnd= np.logical_and(*[lookup[..., i] == tplsum[i] for i in range(D)])
ValueError: invalid number of arguments

我做错了什么？如何修复它以使此代码正常工作？

Answer 1

错误是因为np.logical_and采用 2 个输入和一个可选的输出参数。然而，由于调用 *[lookup[..., i] == tplsum[i] for i in range(D)] 你有 D 参数。当D为2时一切正常，当它为3时你有一个Bug，当它为<2或>3时你会得到异常。

>>> import numpy as np

>>> np.logical_and(*[[1]])
ValueError: invalid number of arguments
>>> np.logical_and(*[[1], [2]])
array([ True], dtype=bool)
>>> np.logical_and(*[[1], [2], np.array([1])])
array([1])
>>> np.logical_and(*[[1], [2], np.array([1]), 3])
ValueError: invalid number of arguments

这可以通过使用 np.logic_and.reduce 来修复，它接受任意数量的数组:

np.logical_and.reduce([lookup[..., i] == tplsum[i] for i in range(D)])