Opened 10 years ago
Closed 10 years ago
#11285 closed defect (othersoftware)
Geotagged images are displayed at wrong (rounded?) position
Reported by: | brainsail | Owned by: | team |
---|---|---|---|
Priority: | normal | Milestone: | |
Component: | Core | Version: | |
Keywords: | template_report | Cc: |
Description
What steps will reproduce the problem?
- Open geotagged images
- Open the track that was recorded with the same device as the photos
What is the expected result?
Image symbols should appear exactly on the track.
What happens instead?
Many image symbols are 10 - 20 m off the track. Looks like they are on a raster, as if the position was rounded too much.
Please provide any additional information below. Attach a screenshot if possible.
Repository Root: http://josm.openstreetmap.de/svn Build-Date: 2015-03-02 02:30:58 Last Changed Author: Don-vip Revision: 8109 Repository UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b Relative URL: ^/trunk URL: http://josm.openstreetmap.de/svn/trunk Last Changed Date: 2015-03-02 00:13:51 +0100 (Mon, 02 Mar 2015) Last Changed Rev: 8109 Identification: JOSM/1.5 (8109 en) Linux Ubuntu 14.04.2 LTS Memory Usage: 1381 MB / 3461 MB (1155 MB allocated, but free) Java version: 1.7.0_75, Oracle Corporation, OpenJDK 64-Bit Server VM Java package: openjdk-7-jre:amd64-7u75-2.5.4-1~trusty1 VM arguments: [-Djosm.restart=true, -Djava.net.useSystemProxies=true] Dataset consistency test: No problems found Plugins: - buildings_tools (30955) - terracer (30892)
Attachments (3)
Change History (8)
by , 10 years ago
Attachment: | josmbug.png added |
---|
comment:1 by , 10 years ago
Resolution: | → othersoftware |
---|---|
Status: | new → closed |
by , 10 years ago
Attachment: | img_0300.jpg added |
---|
comment:2 by , 10 years ago
Resolution: | othersoftware |
---|---|
Status: | closed → reopened |
The exif position has sufficient precision. Attached img_300.jpg and the the track.
comment:3 by , 10 years ago
Resolution: | → othersoftware |
---|---|
Status: | reopened → closed |
Yes, but it's not on the GPX track.
comment:4 by , 10 years ago
Resolution: | othersoftware |
---|---|
Status: | closed → reopened |
I've found there are two different ways of decoding these coordinates. One is correct, the other is off.
- exif-viewer.com
48,44.924128199999984N = 48.74873547
9,5.6574479999999738E = 9.0942908
matches the point in the track
<trkpt lat="48.7487355" lon="9.0942908">
<time>2015-03-28T13:58:22Z</time>
- exifdata.com
48.748611 degrees N
9.094167 degrees E
This is where josm shows it.
I don't know anything about the format of coordinates in exif. The data in this image was encoded by Qt. I hope you can support this encoding.
comment:5 by , 10 years ago
Resolution: | → othersoftware |
---|---|
Status: | reopened → closed |
They include the position twice:
$ exiftool -v img_0300.jpg ... | 3) GPSInfo (SubDirectory) --> | + [GPS directory with 4 entries] | | 0) GPSLatitudeRef = N | | 1) GPSLatitude = 48 44 55 (48/1 44/1 55/1) | | 2) GPSLongitudeRef = E | | 3) GPSLongitude = 9 5 39 (9/1 5/1 39/1) JPEG APP1 (2990 bytes): + [XMP directory, 2961 bytes] | XMPToolkit = GStreamer | ExposureCompensation = 0/1 | GPSLatitude = 48,44.924128199999984N | GPSLongitude = 9,5.6574479999999738E ...
The correct value as XMP metadata and again the incorrectly (rounded) in the standard EXIF format.
It isn't our problem that the metadata is inconsistent, this is simply broken. Given two values, I would rather rely on EXIF.
You can write a script that fixes the images files. Alternatively you may submit a patch that adds a workaround.
This is a problem of the recording device. If you think otherwise, please attach the gpx track and 1-2 images.