My phone is ‘accurate’ when out in the open. Poor in trees.
At first I took it as literally un acurate in trees and got confused trying to figure out what you meant then I realised what you meant. It did actually sound like you were actually in trees and found it inaccurate in trees. cause it said in trees I took it literally at first But then realised you mean in trees as in places with Trees. I thought it was quiet funny that I took your comment literally. I spent 5 or 10 minutes trying to work it out cause it didn’t make sense and I couldn’t think of any other reason why someone would put in trees unless they actually meant literally in trees
Luckily I know what you mean now so Im not comfused so I don’t have to ask why you were in trees
I couldn’t help but mention cause it was funny that I read it that way at first until I realised what you meant
I have an m6 I don’t know about the location accuracy with my camara cause you have to transfer your photos to your smartphone then you can edit the satalite gps location and then you can transfer it back to your camara so mine doesn’t do location only my smartphone does.
I normaly had location services on on my phone when I used to use it for taking pictures and the map would tell me where it is I took the picture and I usually prefer satalite imagery I didn’t notice much difference in accuracy when it came to trees and would usually be almost dead on if it wasn’t I would edit the location and put it at the spot where I know exactly I saw what I saw to make it more acurate but mine was usually fine most of the time regardless of weather its in trees or not
There would seem to be two ways of looking at this.
Could FairPhone sort the problem? Having seen the EXIF data from some of my photos, can you tell in what way(s) FairPhone is misinterpreting the standards?
Could the OU sort the problem? If sites such as Strava, whose main business isn’t in handling photos, can correctly interpret the EXIF data from both my devices, why can’t the OU do this on a site intended to handle photographs?
I think I have mentioned on a number of occasions that with the previous version of ispot we looked at this same issue several times and found that each library that was tried would only read certain camera/phone files but never all of them - even individual manufacturers had some devices that could be read and others that could not be read correctly.
Either Strava have managed to find a library that can read them all (in which case perhaps you could ask them what they use) or they are using more than one library and somehow checking which one is reading each image correctly and using that data. As far as I recall the errors were fairly obvious but this may require quite a lot more coding. If we were going down that route then potentially other error checking could also be developed such as checking if land based plants are being put in the sea.
The Mike Kovarik software (exifr) looks very interesting - it even suggests it can read location data from HEIC/HEIF files. Its JavaScript so would need to be worked into the client-side (web browser) code. I’ll have a play and see what I can come up with. Thanks Thistle for asking, and thanks Stava for answering!
Looks interesting, if it is implementable then would need to test with wide range of image files from cameras and phones. Would be interesting to see what other data e.g. IPTC info on copyright and subject i.e. if you have already identified the image and put it in the subject field etc.
I’ve just updated the JavaScript EXIF reading library that we’re already using but its still not getting the location data from the Fairphone image submitted earlier, although the data is there because I can see it with EXIF Pilot.
It appears that although the GPS data is in the “right place” in the image file header, it sometime contains “odd” characters that trip up the JavaScript that reads it and converts it to decimal to populate the form fields. The new(er) exfir library apparently does a much better job of processing the header so I’m looking forward to getting that working in iSpot.
I’ve just made some more code change - replacing the EXIF library and changed some of the code that grabs the EXIF GPS data and converts it - and I believe iSpot will now be able to grab location data from a wider variety of cameras and phones. I’d be grateful if users can test this for me (changes so far only on the acceptance server) and in the observation please say what make/model of camera you’ve used.
Today I had the need to correct a location. It seemed to remember the last Observation location
My camera is usually VERY accurate
I will test it again in a minute.
.
PERFECT retests -
Olympus camera system and
Android (Samsung A41) very precise
I have received another new UK species dictionary from NBN. Unless anyone has a strong objection I want to deploy this to the acceptance server this morning - doing so will wipe all of the test submissions you’ve kindly made on that server. Please let me know ASAP if you don’t want me to wipe the server and install a new dictionary.
OK after no objections received the acceptance server now contains the very latest UK & Ireland species dictionary as provided by NBN earlier this week.
Unfortunately its throwing a 500 error for all recent observations - old ones seem to be OK - which is strange.
Been there done that… First test ob in.
Works fine…
I will not do much testing UNLESS there is reasonable feedback IN the test site
…
Later. OK it’s NOT working. The 500 error needs fixing before I can get any further
It is possible to add observations and test the New Dictionary but opening Obs creates the error
