- zo 26 mei 2013
- misc
- Jason K. Moore
- #photo, #python, #timestamp, #gexiv2, #pytz
Yumi and I just got back from a month long bicycle tour of Japan. We took two cameras with us. My Panasonic Lumix stopped autofocusing the first day of the trip, so we used her Canon Powershot for all the photographing. We dropped the Canon twice on the trip and now it makes a nasty noise when opening and closing, but it at least still takes photos.
We forgot to set the travel time on the camera so all 700+ of our photos had timestamps from Pacific Standard Time while under daylight saving time (Japan does not use DST). I wanted to correct the timestamps before uploading them to the web so I wrote a little Python script and used Yorba's Gexiv2 and the PyTZ package to sort things out.
To install Gexiv2 in Ubuntu 12.10 I ran these commands:
$ sudo aptitude install libexiv2-dev libtool libgirepository1.0-dev $ git clone git://git.yorba.org/gexiv2 $ cd gexiv2 $ ./configure --enable-introspection $ make $ sudo make install
And installed PyTZ with pip:
$ sudo pip install pytz
Then I wrote this script which searched my photo directory for the specific folders and photographs and adjusted the three timestamps in the photos (this didn't work with the few .mov files we took during the trip because Gexiv2 doesn't support .mov files, so that will need to be done manually or I need to find another tool):
#!/usr/bin/env python # -*- coding: utf-8 -*- """This script converts the time/date stamp of photographs we took in Japan to Japanese time/date from US Pacific time because we forgot to adjust the camera's internal clock at the beginning of the trip.""" import os import re from datetime import datetime import pytz from gi.repository import GExiv2 # The Canon PowerShot SX200 IS date format. DATE_FORMAT = '%Y:%m:%d %H:%M:%S' # We snapped photos in Japan while using the current PST time (which is under # Daylight Savings Time, but Japan is not). pacific_tz = pytz.timezone("America/Los_Angeles") japan_tz = pytz.timezone("Asia/Tokyo") # Get a list of file names that match the time we were there and the camera # filename format. directory = "/media/Data/My Pictures" pattern = r"/media/Data/My Pictures/2013-0[4-5]-\d\d/img_\d\d\d\d.jpg" files_to_adjust = [] for root, dirnames, filenames in os.walk(directory): for filename in filenames: path = os.path.join(root, filename) result = re.match(pattern, path) # if a matching file is found, save it to the list if result: files_to_adjust.append(path) # Now adjust the three timestamps (DateTime, DateTimeDigitized and # DateTimeOriginal) for each file and save the results. for filename in files_to_adjust: print('-' * 20) print(filename) print('-' * 20) metadata = GExiv2.Metadata(filename) for tag in metadata.get_exif_tags(): if 'DateTime' in tag: timestamp = metadata[tag] pacific_time = pacific_tz.localize(datetime.strptime(timestamp, DATE_FORMAT)) japan_time = pacific_time.astimezone(japan_tz) japan_timestamp = japan_time.strftime(DATE_FORMAT) print(tag) print('Current time stamp: {}.'.format(timestamp)) print('Adjusted time stamp: {}.'.format(japan_timestamp)) print('-' * 20) # Set the timestamp. metadata[tag] = japan_timestamp # Save the file with the adjusted timestamps. metadata.save_file()
The resulting output for an example file looked like this:
-------------------- /media/Data/My Pictures/2013-05-21/img_5484.jpg -------------------- Exif.Image.DateTime Current time stamp: 2013:05:21 00:51:31. Adjusted time stamp: 2013:05:21 16:51:31. -------------------- Exif.Photo.DateTimeDigitized Current time stamp: 2013:05:21 00:51:31. Adjusted time stamp: 2013:05:21 16:51:31. -------------------- Exif.Photo.DateTimeOriginal Current time stamp: 2013:05:21 00:51:31. Adjusted time stamp: 2013:05:21 16:51:31. --------------------
The correct 16 hour difference was applied correctly by PyTZ which takes care of DST differences. I then had to remove all of the photos from my Shotwell library and re-add them because the Shotwell database didn't automatically update with the new times. Once I did that, everything was now correct (except the .mov files, of course).
The photos of the trip can be found in my G+ photo album.