[python] pdf 를 이미지로 저장

• poppler 다운로드 

https://github.com/oschwartz10612/poppler-windows/releases/tag/v23.11.0-0

• 압축 해제

C:\poppler-23.11.0

• pdf2image 설치

pip install pdf2image

def save_pdf_page_as_image(pdf_path, page_number, filename, output_dir):

    # PDF 파일에서 이미지로 변환
    name = os.path.splitext(filename)[0]
    output_path = os.path.join(output_dir, '{0}.png'.format(name))
    images = convert_from_path(pdf_path, first_page=page_number, last_page=page_number, poppler_path=r'C:\poppler-23.11.0\Library\bin')

    if images:
        # 첫 번째 이미지 선택 (페이지 번호에 해당하는 이미지)
        image = images[0]
        # 이미지 저장
        image.save(output_path, 'PNG')
    else:
        print(f'error.')

Image to base64, base64 to image

PIL.UnidentifiedImageError: cannot identify image file

import base64

import io
from io import BytesIO
from PIL import Image

byteImgIO = io.BytesIO()
byteImg = Image.open('./images/img_small.png')
byteImg.save(byteImgIO, "PNG")
byteImgIO.seek(0)
byteImg = byteImgIO.read()
v2 = base64.b64encode(byteImg)
v3 = base64.b64decode(v2)
# i = Image.open(BytesIO(v3))
# i = i.convert('L')
i = BytesIO(v3)
i.seek(0)
i = Image.open(i)
i.save('./images/a.png')
print('finish')

excel에 첨부된 이미지 추출.

# -*-coding:utf-8-*-

from openpyxl import load_workbook
from openpyxl_image_loader import SheetImageLoader


def main():
    filepath = './excel/sample.xlsx'
    load_wb = load_workbook(filepath, data_only=True)
    for sheet_name in load_wb.sheetnames:
        sheet = load_wb[sheet_name]

        image_loader = SheetImageLoader(sheet)
        # image = image_loader.get('C3')
        # image.show()
        print(sheet.max_row, sheet.max_column)
        for column in range(sheet.max_column):
            column = chr(ord('A') + column)
            for row in range(sheet.max_row):
                cell_id = '{0}{1}'.format(column, row)
                print(cell_id)
                if image_loader.image_in(cell_id):
                    print(cell_id, '->', image_loader.image_in(cell_id))


if __name__ == '__main__':
    main()