用智能手机或数码相机拍摄的照片有丰富的附加信息元数据，通常以 EXIF 格式存储这些元数据，EXIF 即Exchangeable Image File Format，有不同的标准版本。
EXIF 元数据可能包含关于设备模型、尺寸、图片的日期及其位置的信息。

安装exif 库

pip install exif

读取图片

from exif import Image
img_path = 'images/image_exif.jpg'
with open(img_path, 'rb') as src:
    img = Image(src)
    print (src.name, img)

输出内容为：

images/image_exif.jpg <exif._image.Image object at 0x00000223EBC62430>

通过PTL库读取该输出即可获取图像内容：

import PIL
image = PIL.Image.open(img_path)
image

检查图片中是否含有EXIF数据：

if img.has_exif:
 info = f"has the EXIF { img.exif_version}"
else:
 info = "does not contain any EXIF information"
print(f"Image { src.name}: { info}")

若有，提示：

Image images/image_exif.jpg:  has the EXIF 0220

没有：

# Image without EX
with open('images/foto_no_exif.jpg', "rb") as src:
 img = Image(src)
 if img.has_exif:
    info = f" has the EXIF { img.exif_version}"
 else:
    info = "does not contain any EXIF information"
 print(f"Image { src.name}: { info}")

输出：

Image images/foto_no_exif.jpg: does not contain any EXIF information

获取图像中的EXIF元数据

# Read again photo with exif info
with open(img_path, “rb”) as src:
 img = Image(src)
img.list_all()

该列表显示了元数据中所有的标签内容：

['image_description',
 'make',
 'model',
 'orientation',
 'x_resolution',
 'y_resolution',
 'resolution_unit',
 'software',
 'datetime',
 'y_and_c_positioning',
 '_exif_ifd_pointer',
 '_gps_ifd_pointer',
 'compression',
 'jpeg_interchange_format',
 'jpeg_interchange_format_length',
 'exposure_time',
 'f_number',
 'exposure_program',
 'photographic_sensitivity',
 'exif_version',
 'datetime_original',
 'datetime_digitized',
 'components_configuration',
 'exposure_bias_value',
 'max_aperture_value',
 'metering_mode',
 'light_source',
 'flash',
 'focal_length',
 'maker_note',
 'user_comment',
 'flashpix_version',
 'color_space',
 'pixel_x_dimension',
 'pixel_y_dimension',
 '_interoperability_ifd_Pointer',
 'file_source',
 'scene_type',
 'custom_rendered',
 'exposure_mode',
 'white_balance',
 'digital_zoom_ratio',
 'focal_length_in_35mm_film',
 'scene_capture_type',
 'gain_control',
 'contrast',
 'saturation',
 'sharpness',
 'subject_distance_range',
 'gps_latitude_ref',
 'gps_latitude',
 'gps_longitude_ref',
 'gps_longitude',
 'gps_altitude_ref',
 'gps_timestamp',
 'gps_satellites',
 'gps_img_direction_ref',
 'gps_map_datum',
 'gps_datestamp']

获取图像中的地理坐标

将经纬度坐标转换为十进制数值表示：

def decimal_coords(coords, ref):
 decimal_degrees = coords[0] + coords[1] / 60 + coords[2] / 3600
 if ref == "S" or ref == "W":
     decimal_degrees = -decimal_degrees
 return decimal_degrees

获取十进制经纬度坐标以及其他信息函数：

def image_coordinates(img_path):
    with open(img_path, 'rb') as src:
        img = Image(src)
    if img.has_exif:
        try:
            img.gps_longitude
            coords = (decimal_coords(img.gps_latitude,
                      img.gps_latitude_ref),
                      decimal_coords(img.gps_longitude,
                      img.gps_longitude_ref))
        except AttributeError:
            print 'No Coordinates'
    else:
        print 'The Image has no EXIF information'
    print(f"Image { src.name}, OS Version:{ img.get('software', 'Not Known')} ------")
    print(f"Was taken: { img.datetime_original}, and has coordinates:{ coords}")

输出内容：

Image images/DSCN0012.jpg, OS Version:Nikon Transfer 1.1 W ------
Was taken: 2008:10:22 16:29:49, and has coordinates:(43.46715666666389, 11.885394999997223)

参考：

https://medium.com/spatial-data-science/how-to-extract-gps-coordinates-from-images-in-python-e66e542af354