Python Globbing 图像目录

Question

I am working on a project currently and so far I have generated a folder of images (png format) where I need to iterate through each image and do some operation on it using PIL.

I have the operation correctly working by manually linking the file path into the script. To iterate through each image I tried using the following

frames = glob.glob("*.png")

But this yields a list of the filenames as strings.

PIL requires a filepath for the image to be loaded and thus further used

filename = input("file path:")
image = Image.open(filename)
callimage = image.load()

How can I convert the strings from the glob.glob list and have it used as an argument for the Image.open method?

Thanks for your feedback!

I am on python 3.6.1 if that holds any relevance.

Answer 1

Solution with os package:

import os

source_path = "my_path"

image_files = [os.path.join(base_path, f) for f in files for base_path, _, files in os.walk(source_path) if f.endswith(".png")]

for filepath in image_files:
    callimage = Image.open(filepath).load()
    # ...

Solution using glob :

import glob

source_path = "my_path"

image_files = [source_path + '/' + f for f in glob.glob('*.png')]

for filepath in image_files:
    callimage = Image.open(filepath).load()
    # ...

Answer 2

Use the pathlib library instead of glob.glob() so that you don't have to worry about building the file's path. The Image.open() method also takes a pathlib.Path object as the first argument.

from pathlib import Path

frames = Path('path_to_folder')

for file in frames.glob('*.png'):
    callimage = Image.open(file).load()

Answer 3

The builtin map()<\/code> is a general way to do something, in this case load<\/code> , for a list of things:
map( func, [a, b, c ...] )<\/code> is roughly
[func(a), func(b), func(c) ...]<\/code> . So

def loadit( png ): 
    return Image.open( png ).load()

images = map( loadit, glob( "*.png" ))
for image in images: ...