Compress photos in bulk with pillow python module

compress_images

Compress photos in bulk with pillow python module

  • Images can be compressed to reduce storage space. This can be particularly useful while storing images in cloud storage.
  • Images can be compressed or resized using pillow python library in python. It can be installed using pip with the command python -m pip install Pillow
  • Pillow is a python library for image processing
  • To reduce the size of an image, it can be resized to lower resolution or jpeg compression can be increased
  • thumbnail function of pillow library can be used to resize an image.
  • save function of pillow library can be used to control the jpeg image file compression percentage
  • piexif python library can be used to extract the exif data (device, gps, etc) of the original image and use it in the compressed file
  • The source code below compresses the images in the folder D:\\test1 and saves the compressed images in D:\\test2. The file metadata like modified time and last access time is set in the compressed image using the os.stat and os.utime functions
import os
from PIL import Image
import piexif
import glob
import pathlib

def copyMetadata(srcImgPath, destImgPath):
    # Open the source image
    srcImg = Image.open(srcImgPath)

    # Extract EXIF data from the source image
    if "exif" in srcImg.info:
        exifData = piexif.load(srcImg.info['exif'])

        # bug handling for 41729 exif tag (scenetype)
        if 41729 in exifData['Exif'] and isinstance(exifData['Exif'][41729], int):
            exifData['Exif'][41729] = str(exifData['Exif'][41729]).encode('utf-8')
    else:
        exifData = {}
    
    # Open the destination image
    destImg = Image.open(destImgPath)

    # Save the destination image with the source's EXIF data
    destImg.save(destImgPath, exif=piexif.dump(exifData))

    # Copy file attributes for last accessed time and modified time
    srcImgStats = os.stat(srcImgPath)

    # set file ownership information (applicable only in unix systems)
    # os.chown(destImgPath, srcImgStats.st_uid, srcImgStats.st_gid)

    # set modified time and last accessed time
    os.utime(destImgPath, (srcImgStats.st_atime, srcImgStats.st_mtime))

def compressImage(srcImgPath, destImgPath, maxWidth, quality=100):
    # open source image
    image = Image.open(srcImgPath)

    # get the image dimensions
    width, height = image.size

    # Resize image if necessary
    if width > maxWidth:
        newHeight = height*(maxWidth/width)
        image.thumbnail((maxWidth, newHeight))

    # Save the new image
    outFname = destImgPath
    print(f"saving {outFname}")
    image.save(outFname, 'JPEG', quality=quality)

    # copy metadata from source image to destination image
    copyMetadata(srcImgPath, destImgPath)

# Example usage
inpFolder = r"D:\\test1"
outFolder = r"D:\\test2"
maxWidth = 1280
qualityPerc = 100

# iterate through all images of the input folder 
for imgIter, imgPath in enumerate(glob.glob(inpFolder+r"\\*.jpg")):
    print(f"{imgIter} : processing {imgPath}")
    outFileName = pathlib.Path(imgPath).name
    outFilePath = os.path.join(outFolder, outFileName)
    compressImage(imgPath, outFilePath, maxWidth, qualityPerc)

print("process complete...")

Video

Video on this post can be seen here

References

Comments