ディレクトリ以下のJPEGファイルを自動分析するコード(とりまパナ機のみ対応)

jpeg_exif_analysis.py

import glob
import os
from pprint import pprint
from typing import Dict, Any

from PIL import Image
from PIL.ExifTags import TAGS
from PIL.MpoImagePlugin import MpoImageFile

# Exif情報を取得するためのシグネチャ
exif_signature = b'Exif\x00\x00'

# DecompressionBombWarning対策
Image.MAX_IMAGE_PIXELS = 1000000000

if __name__ == '__main__':
    root_path = 'E:/整理済/画像/・記念/撮影画像まとめ'

    lens_focal_count: Dict[str, Dict[str, int]] = {}  # lens_focal_count[レンズ名][焦点距離] = 回数

    # 再帰的に読み込む
    for path in glob.glob(os.path.join(root_path, '**', '*.jpg'), recursive=True):
        print(path)
        with open(path, 'rb') as f:
            # Pillowを使い、焦点距離・メーカー名・メーカーノートを取得する
            im: MpoImageFile = Image.open(path)
            temp: Dict[str, Any] = {}
            for tag_id, value in im.getexif().items():
                temp[TAGS.get(tag_id, tag_id)] = value
            if 'FocalLengthIn35mmFilm' not in temp or 'Make' not in temp or 'MakerNote' not in temp:
                continue
            focal_length_by_35 = temp['FocalLengthIn35mmFilm']
            maker = temp['Make']
            maker_note = temp['MakerNote']

            # メーカー名に応じた処理を実施する
            if maker == 'Panasonic':
                # TIFF IFDの読み込み時、オフセット計算に使用する定数を取得する
                raw_data = f.read()
                raw_exif_offset = raw_data.find(exif_signature) + len(exif_signature)

                # TIFF IFDを順次読み込み、レンズ名に関わるものを検索する
                ifd_count = int.from_bytes(maker_note[12:14], byteorder='little')
                for i in range(0, ifd_count):
                    pointer = 12 + 2 + i * 12
                    ifd_data = maker_note[pointer:pointer + 12]
                    tag_id = int.from_bytes(ifd_data[:2], byteorder='little')
                    if tag_id == 0x51:
                        # 取得処理を実施
                        lens_name_length = int.from_bytes(ifd_data[4:8], byteorder='little')
                        lens_name_offset = raw_exif_offset + int.from_bytes(ifd_data[8:], byteorder='little')
                        lens_name_bin = raw_data[lens_name_offset:lens_name_offset + lens_name_length]
                        lens_name = lens_name_bin.decode('ASCII').replace('\0', '')

                        # Dictに記録する
                        if lens_name not in lens_focal_count:
                            lens_focal_count[lens_name] = {}
                        if focal_length_by_35 not in lens_focal_count[lens_name]:
                            lens_focal_count[lens_name][focal_length_by_35] = 0
                        lens_focal_count[lens_name][focal_length_by_35] += 1

    # 集計処理を実施
    lens_count: Dict[str, int] = {}
    all_count = 0
    for lens_name, data in lens_focal_count.items():
        if lens_name not in lens_count:
            lens_count[lens_name] = 0
        for focal_length_by_35, count in lens_focal_count[lens_name].items():
            lens_count[lens_name] += count
            all_count += count

    # 結果を出力する
    print(f'\n写真枚数：{all_count}\n')
    print('レンズ名,使用率(％),焦点距離毎の使用率(％)')
    lens_count2 = sorted(lens_count.items(), key=lambda x: x[1], reverse=True)
    for lens_name, count in lens_count2:
        print(lens_name, end=',')
        print(round(100.0 * count / all_count, 1), end=',')
        focal_count = sorted(lens_focal_count[lens_name].items(), key=lambda x: x[1], reverse=True)
        for focal_length_by_35, count2 in focal_count:
            print(f'{focal_length_by_35}mm({round(100.0 * count2 / count, 1)}) ', end='')
        print('')