Geotagging Photos

Thanks, I got it working, although I must say I am gobsmacked at how slow the export is, and how it ties up effectively 100% of my computer resources.

Is it normal for the exported images to more than double in size ?

Definitely not! Are they just doubling in terms of the file size, or is the image itself doubling in dimensions (width and height in pixels)? If the latter, it would explain the former and also the slowdown.

No, the dimensions are unchanged, 6000*4000 in the original, and the same in the exported, horizontal and vertical resolution unchanged at 300dpi, Bit depth unchanged at 24.

But as an example, original is 2.3 mb in size, exported is 5.3 mb

My guess is that your export settings must have quality at 100%, which would have the program re-sampling the existing jpeg with no output compression. The original jpegs from whatever program you converted with probably have a moderate default compression setting. I’d recommend setting the quality much lower- for purposes like these, you’re not losing any important image data even with pretty heavy compression. I put my lightroom export settings for iNat to 2000x2000 maximum dimensions and limit quality to 1MB (iNat resizes all photos to just a little larger than 2000x2000 anyways).

Here’s a relevant discussion on what’s going on with the compression differences:

Thanka, I have another batch from my outing yesterday, but am so far behind on photo processing and inat entry, I’m not sure when I will get to those.

I used to use a free google app called myTracks on my phone to track my location. It did a good job of tracking my path with an overlay on Google Earth. But it was cumbersome to get the geo data on to my photos, really drew down the battery on the phone so that 4 hours recording was the practical maximum and the last straw was that Google stopped supporting it.

So last year I purchased a geo data logger Holux RCV-3000 for less than $100 and it’s been great. The unit is dead simple. It is a very small unit the size of a small box of matches. There is only an on and off switch and three lights. The three lights are for bluetooh, satellite lock and low battery warning. Battery is rated to last 20 hours and I’ve never had a problem. It charges via a USB connection so a battery pack would probably allow for two weeks of daily use. It comes with Holux EZ-tour software for reading the logs. I connect the data logger to my computer via the USB port and download the track data. I have not been able to connect via bluetooth and the software manual is not very helpful on bluetooth and the USB connection. I had to find the COM port for the USB through trial and error.

Once the data is uploaded to the EZ-tour software you can add JPEG images from a file or folder and it will write the GPS data to the photos. The software allows the input of a time zone and time difference from UTM satellite time (the RCV-3000 gets UTM time from the satellites) and the camera. For simplicity I always adjust my camera clock every two weeks or so to the second using world time on the computer. The camera time drifts by up to a minute a month. I no longer bother with daylight saving time and use standard time year round. I friend of mine uses Greenwich time all the time on his camera.

The EZ-tour software can export a KMZ file. The only shortcoming is that the software does not write the GPS data to RAW files so I have to manually enter the GPS data into my RAW files from the JPEG photos. This isn’t too big a deal as my RAW processor has a goof EXIF editor. I recently found a few other applications that can read the RCV-3000 data logger so I’ll experiment with them and see if they can handle RAW files.

I find that sometimes the location needs to be adjusted to reflect the actual location of the organism. If the area has good Google earth mapping I will adjust the location but if the mapping is poor I will leave it as it is.


I have an add-on gizmo on my camera that tracks GPS.

Reducing this did resolve the photo output size issue. It is still shockingly slow and ties up 100% of my machine resources though. (70 minutes to process 60 photos which averaged about 2mb in size) with nothing else running other than a browser, on a powerful machine that does not have this issue with any other program.

I’ve been using the Nikon Coolpix S9700 - mostly because it has a 30x zoom so I can get a photo without having to get too close and scare off what I’m photographing. It does have a built in GPS, but I find it’s not very accurate and takes so long to get a fix that the species had flown/run/swum away. I’ve recently realized I can link this to my phone using the Nikon Windows Mobility app. So I take the photo with my camera and on the spot can download the image onto my Android. Then using the iNat app, I “choose image” instead of “take photo” and select the photo I just took with my camera to record my observation.
I’d really like the ability to use an external GPS instead of the default one in my phone to get better accuracy, but I haven’t found a way to do that yet. I’ve found other apps that allow this, like Qfield which, using an extra app that can provide a “mock location” from an external GPS. I haven’t looked into this to see if might work with iNat

1 Like

Getting all photos as accurately tagged as possible is my goal too. My iPhone GPS is almost always satisfactory, and I thought that would be used for the location recorded on my Olympus TG-5 and Canon SX-730, but it seems the Canon only uses - and needs - the bluetooth connnection with the phone in order to get internet to use its own GPS, and the Olympus doesn’t use the phone at all, except to send its own GPS to the phone. I think.
I bought a hand-held GPS thinking that would be useful, but it may be an old model because it doesn’t have any way of connecting with anything, it seems. So Ihaven’t yet invested the time it would take to learn to use it to make its own internal maps.

I’ve tried doing the phone GPS tracking and then then synching with the photos, it simply is not working for me.

I have no idea where the problem is, but the locations I am getting in my photos are not even close to accurate. I have ones from this weekend where the location is 50km away from where I was. It’s not a time synch issue between the pics and the GPS track as I never was anywhere near where the locations that are coming up.

I’m not sure I am interested in trying to diagnose if the tracker is wrong, the software that is linking the GPS tracks into the pics is wrong etc.

It’s simply a drag on time, when I can estimate manually and get very close, even if that slows me down too.

Quick tip: open Google maps (or similar) on your phone before you take a picture with it. From what I can tell, phone cameras passively record whatever location the phone is currently registering, so you have to get something else running that will specifically engage your GPS. Any photos taken subsequently will have a more accurate location tag. Sticking with the phone theme, another thing to do is to screenshot your Google maps location periodically so you have that as reference later during uploads.


A quick update, the Holux M-241 is non-functional after the week number rollover earlier this year and Holux doesn’t have any firmware update to fix the problem. I’m not sure if the plus version works or not. I have gone back to using my BT747 unit. It’s about the size of a matchbox and has a rechargeable battery. It’s been kicking around the desert with me for 11 years and it still runs great.