The Global Positioning
System is a network of 24 satellites. GPS receivers use information transmitted by at least three of these satellites to calculate the user's exact location. Courtesy of Garmin.
A Garmin GPS unit that
generates a track log.
Ricoh's 500SE is GPS
enabled and sold mainly to industry and other professionals.
The Nokia N95 includes
integrated GPS and a 5-megapixel camera.
The Sony GPS-CS1 unit
tracks your location every 15 seconds so the GPS Image Tracker software can match the time and location data in your images to infer where each photo was taken. The Picture Motion Browser software, supplied with certain Sony cameras, displays your photos on an online map.
Click here to see how a KMZ file works with Google Earth.
When you see an interesting photo you are usually curious about who or what the subject is, and when and where it was taken. Imagine driving Route 66 and mapping your photos showing where each was taken along the route. The good news is that the information you need to do this can be stored in an image's Exif information. For example, the latitude, longitude, heading and elevation of the camera can be recorded—called geotagging or geocoding. With this information in an image file you can post it on an interactive map that shows where it was taken and even the direction in which the camera was pointed. The bad news is that this information is rarely stored in image files at the time they are captured. Digital cameras and camera phones, with few exceptions, have not yet integrated the necessary GPS (Global Positioning System) technology. Ideally a camera would have GPS built-in or at least be connectible to a GPS unit. Unfortunately few cameras have this feature, and even then GPS receivers don't work well under a heavy forest canopy or indoors. Even with a GPS equipped camera a lot of photos could be missing the location information you expect them to have.
However, systems have been developed that let you circumvent the lack of a GPS device in or attached to your camera until the camera companies catch up. The location at which a photo was taken can be determined using a GPS receiver, by manually looking up the coordinates of a street address, or by simply clicking the location on a map. The number of photos you plan on geotagging influences your choice of which method to use. Manually tagging a few images is easy, but it soon becomes tiresome when you tag a lot of them. In this section we'll discuss how you can use these techniques while focusing on Google Earth and Google's related Picassa2 program.
Latitude and longitude—A Refresher
To relate a photo to a location you need to know at least the latitude and longitude of where it was taken. Here is a quick refresher:
- Latitude specifies north/south locations in degrees, ranging from 0° at the
Equator to 90° N for the North Pole, and -90° S for the South Pole.
- Longitude specifies east/west locations starting from an imaginary north south
line called the Prime Meridian running through Greenwich, England. Longitude ranges from 0° degrees at the Prime Meridian to +180° degrees eastward and -180° westward.
Each degree of both latitude and longitude is sub-divided into 60 minutes, each of which is farther divided into 60 seconds. A longitude is thus specified as 123° 27' 30" E and latitude as 35° 17' 40" N.
Finding locations—Dragging and Dropping
One of the easiest ways to map photos that don't contain location information is to use a photo sharing site such as Flickr. On that site you open an album of your photos and select Organize > Map to display a filmstrip of your images and a map. You can zoom the map to the desired level of detail and drag and drop images on the spot where they were taken. Markers appear at each spot on the map and by clicking one of them you and your visitors see the photo taken there.
Photoshop Elements 5 uses a different approach. You drag and drop photos into a map that's displayed in the program, then upload the map and images to a Web site.
Finding Locations—Manual Geotagging
Using a site such as Flickr or SmugMug, or using a program such as Picasa2, you can add latitude and longitude to images using tags, or by embedding them into the image file's Exif header. (It's better to embed it into the Exif header because it then travels wherever the image travels). The first step is to determine the latitude and longitude at which a photo was taken.
If you recall where you took a photo you can find that spot's latitude and longitude using Google Earth or Flickr's mapping system. For example, on Google Maps (www.maps.google.com) double click the location to center it in the map, then click Link to this page. The link to the center of the map is displayed on the browser's address line and the portion following the ll is the latitude and longitude. It looks like this:
The lowest tech way of keeping track of locations is to use a GPS unit and take a photo of its screen every time you take a picture. This way you have a record to which you can refer when geotagging images.
Once you know the location, you then geotag the images.
- In Flickr, you geotag images using the following format:
- In SmugMug you select one or more photos and select Edit Geography from the Photo Tools drop-down menu and then type the latitude or longitude in or even look up the location information for a street address.
- In Picasa you select one or more photos and then select the menu commands Tools > Geotag >Geotag With Google Earth. When the map appears you zoom and pan it to center the desired spot under cross hairs and then click the Geotag or Geotag All button to embed the location in the photos Exif header.
Finding Locations—Matching Time and Track
Even when your camera is GPS challenged, there is another way to automatically add location information to Exif headers. It's done using software that matches times stored in an image's Exif header with times stored in a track log created by a separate GPS that you carry along while shooting. When the software finds matching times, it looks up in the track log where you were at that time and stores the location in the image's Exif header. If a photo falls between two log point, or if the GPS looses signal for awhile, a photo's position is estimated using the two closest log points. This is how the Sony GPS-CS1 unit works. For this procedure to work accurately you must set the camera to match the GPS unit's time and date. Keep in mind that a camera's clock doesn't change automatically when you cross time zones.
Adding images to a map
Once images have their location information embedded, it's easy to place them on maps. On some sites you just upload them and when the site finds the location information, it adds them to the map. (Some can even do this if you have tagged the images with place names). On most maps icons represent each photo's position and you either hover over one or click it to see the image.
One approach to geotagging images
shows great promise. NXP's swGPST system records raw location data at the time you take each picture and stores it in the image, but doesn't process the information in the field. Processing the raw GPS data into a specific latitude and longitude is done back at your computer when you connect to their Website. This approach is faster, cheaper, and less power hungry than having both location capture and processing done in the camera.
A photo of a US Geologic Survey marker on Montecito Peak outside of Santa Barbara mapped to Google Earth using SmugMug.
Photoshop Elements 5 lets you drag and drop images onto a map and then upload the map to a Web site. Visitors see the map and icons they
click to see the photos.