CarlitosDiaz Posted at 2018-10-16 04:04
This is the GPS information of a picture taken on top of a table.

MSL = AGLish

Yea, it looks like the correct barometer-based altitude is no longer embedded in the image tags.

DJI - this a real problem if true.

The new firmware fixed the altitude issue for me. Thanks to DJI for the quick turn around on this.

SWscanning Posted at 2018-10-16 08:10
The new firmware fixed the altitude issue for me. Thanks to DJI for the quick turn around on this.

Yes, thanks for fixing the GPS altitude, however for those of us with the need for greater precision the barometer-based elevation is a must have, an is present in P4P, and Mavic Pro, and the older Mavic 2 firmware.

I'm also having the same problem!!! I'm very very sad because I bought this new Mavic 2 pro and my client wants to know all the altitude reached in each position! how can you solve this? Is there any prediction for this???

Guys, I just posted a bad review to BH readers for not buying any of the Mavic models until you can fix this problem! this is lying to your customers!!! I bought something that I'm stuck to know!!

I've been a week doing a job with 600 images that a client need to reference all the data taken and know what???

SHAME ON YOU DJI!!!!

DJI Susan Posted at 2018-9-17 00:29
Sorry for keeping you waiting. Our engineers are working on this and we'll release on our website as soon as it is ready. We'll keep an eye on the status as well. Thanks for your support and patience.

Hi Susan,
do you have some new updates regarding this issue which is still causing a lot of troubles during mapping mission. Your engineers have found the problem ? If yes, do you have a rough estimate on the expected time for a new firmware update?
Best Regards,
Daniele

TenenteDAN1986 Posted at 2018-10-30 08:05
Hi Susan,
do you have some new updates regarding this issue which is still causing a lot of troubles during mapping mission. Your engineers have found the problem ? If yes, do you have a rough estimate on the expected time for a new firmware update?
Best Regards,

Hi Daniele, the GPS altitude issue has been figured out in the firmware update, please kindly update to the latest firmware and see whether it works now, thank you!

EduardoB Posted at 2018-10-17 17:36
Guys, I just posted a bad review to BH readers for not buying any of the Mavic models until you can fix this problem! this is lying to your customers!!! I bought something that I'm stuck to know!!

I've been a week doing a job with 600 images that a client need to reference all the data taken and know what???

Hi EduardoB, as mentioned in 42#, the issue has been fixed by the firmware update, you may update to V01.00.01.00 and see whether it works now. We're sorry for the inconvenience caused.

BTW, if you need one's attention, please click the reply button under one's message so that he/she can get the reminder and handle it timely, thanks!

DJI Susan Posted at 10-31 22:41
Hi Daniele, the GPS altitude issue has been figured out in the firmware update, please kindly update to the latest firmware and see whether it works now, thank you!

Hello again Susan,
thanks for your reply. I'm running the last firmware on my MP2 (version checked with both DJIGO4 app and DJI assistant). However, the problem  has not been solved yet. Take a look of the attached snapshots.  I have highlighted with a red circle both the exif and Xmp metadata information . As you can see in the exif GPS altitude is equal to 0 despite I was flying at 60 m (altitude reading from my smartphone with DJIGO4). Moreover in the Xmp metadata also the relative altitude (derived from barometric sensor, I suppose) seems wrong with a strange negative value (-33.14 m).

Any news??? Its terrible !! After update altitude at 0 and wrong coordinates!!!!

DJI Susan Posted at 10-31 22:48
Hi EduardoB, as mentioned in 42#, the issue has been fixed by the firmware update, you may update to V01.00.01.00 and see whether it works now. We're sorry for the inconvenience caused.

BTW, if you need one's attention, please click the reply button under one's message so that he/she can get the reminder and handle it timely, thanks!

Nothing fixed- even with latest firmware!!!! With last update it seems to every photo 0m alt than some miles that showed before!

fanse71ed566 Posted at 11-11 08:53
Nothing fixed- even with latest firmware!!!! With last update it seems to every photo 0m alt than some miles that showed before!

Yes it's ridiculous.  So many problems with imagery  from Hasselblad camera/MP2, not to mention  the other compatibility issues regarding HEVC/H.265 video codecs with  video editing softwares... after  I
have spent 1400 euros I'm seriously thinking to return my MP2 back! For a company like DJI which has a turnover of approximately 17 billion of dollars  I really suppose that their engineers can handle this firmware problems!

TenenteDAN1986 Posted at 11-9 05:06
Hello again Susan,
thanks for your reply. I'm running the last firmware on my MP2 (version checked with both DJIGO4 app and DJI assistant). However, the problem  has not been solved yet. Take a look of the attached snapshots.  I have highlighted with a red circle both the exif and Xmp metadata information . As you can see in the exif GPS altitude is equal to 0 despite I was flying at 60 m (altitude reading from my smartphone with DJIGO4). Moreover in the Xmp metadata also the relative altitude (derived from barometric sensor, I suppose) seems wrong with a strange negative value (-33.14 m).

Hi TenenteDAN1986, the altitude reading from DJI GO 4 is the relative height, but the GPS altitude of the picture is the height above sea level, you may check the take-off altitude and confirm.
For the H.264/H.265, you may choose the proper one according to the configuration of your PC. Sorry about this.

fanse71ed566 Posted at 11-11 08:53
Nothing fixed- even with latest firmware!!!! With last update it seems to every photo 0m alt than some miles that showed before!

Hi there, we're sorry for the inconvenience. Could you upload some original pictures from SD card and upload to Dropbox or Google Drive, then share the link with us here? We'd like to check the exact status.

DJI Susan Posted at 11-12 00:00
Hi there, we're sorry for the inconvenience. Could you upload some original pictures from SD card and upload to Dropbox or Google Drive, then share the link with us here? We'd like to check the exact status.

Hi Susan,
thanks again for your reply! Yes  the altitude reading from DJI GO 4 is the relative height, however since I have lunch my MP from the beach (i.e. take-off altitude = 0) the reading should match the GPS altitude (height above sea level with an error of +-10m). However, considerign this aspect I would prefer to use the internal pressure sensor in order to get a more accurate estimate of altitude based on barometric pressure (amost constant during a flight of 15-20 m, with god weather conditions). Usually, with my Phantom 4 advanced I use the barometer values (extracted  from Xmp metadata ) because also in the Phantom4 the GPS altitude is very innacurate.  I have attached (Google drive link) some pictures from MP2 so you can check the huge inconsistencies regarding the altitude values stored in the metadata (exif and xmp).
I hope this helps to solve the problem maybe in a further firmware update...

https://drive.google.com/open?id ... CiHAYtIbf6hxpeRI_41


P.s.: For the H.264/H.265, I can choose only h.264 since Hevc codec is quite RAM consuming and I have not the correct codec in mi Adobe products. However, I get good results (after spending a lot of time with converting tools) by recording in 4k D-log (so h.265) and the transcoding the footages to h.264 (FULL HD)...-(This is another problem so maybe we can deal with this aspects in another post)

TenenteDAN1986 Posted at 11-12 12:32
Hi Susan,
thanks again for your reply! Yes  the altitude reading from DJI GO 4 is the relative height, however since I have lunch my MP from the beach (i.e. take-off altitude = 0) the reading should match the GPS altitude (height above sea level with an error of +-10m). However, considerign this aspect I would prefer to use the internal pressure sensor in order to get a more accurate estimate of altitude based on barometric pressure (amost constant during a flight of 15-20 m, with god weather conditions). Usually, with my Phantom 4 advanced I use the barometer values (extracted  from Xmp metadata ) because also in the Phantom4 the GPS altitude is very innacurate.  I have attached (Google drive link) some pictures from MP2 so you can check the huge inconsistencies regarding the altitude values stored in the metadata (exif and xmp).
I hope this helps to solve the problem maybe in a further firmware update...

Received with thanks! I'll forward to our engineers to check.

DJI Susan Posted at 11-14 01:55
Received with thanks! I'll forward to our engineers to check.

Ok  Susan, good!
We hope for a better outcome than in post #18

Cheers,
Daniele

TenenteDAN1986 Posted at 11-14 08:19
Ok  Susan, good!
We hope for a better outcome than in post #18

I‘ll keep following up, hope we can figure it out this time.

I really hope this gets fixed soon. My images are all taken at a hundred and some million meters above sea level. SpaceX must be jealous.
What a disappointment to see such an error in a new product (Mavic 2 Enterprise).

DJI Susan Posted at 11-15 23:17
I‘ll keep following up, hope we can figure it out this time.

Susan.....i cant expressed how let down am feeling on these issue....especially as i submitted a proposal to a client using the M2P as a solution for their video/photo/ mapping needs in a portable package.....Now am being forced to look for alternatives for this potential big supply.

If its a deliberate attempt to force us to buy your other products like P4P then its a big shame and thumps down to DJI.......hoping you prioritize this and get possitive reviews like in BnH

fans51675b83 Posted at 11-18 06:17
Susan.....i cant expressed how let down am feeling on these issue....especially as i submitted a proposal to a client using the M2P as a solution for their video/photo/ mapping needs in a portable package.....Now am being forced to look for alternatives for this potential big supply.

If its a deliberate attempt to force us to buy your other products like P4P then its a big shame and thumps down to DJI.......hoping you prioritize this and get possitive reviews like in BnH

Im in the same situation as you regarding the M2P drone a group of costumers relying on me delivering Ortophotos, 3D models etc..
Words can't describe the disappointment and frustration towards DJI for this. If it shows out to be a "cheap" trick to have us purchase other drones this is my absolute last drone purchase with DJI..

Expect a solution from DJI latest November 30th, 2018. Otherwise I see no other way than to claim my money back for my recent purchase of the M2P.

Regards
Simon Vammen
Denmark

SimonID Posted at 11-17 11:39
I really hope this gets fixed soon. My images are all taken at a hundred and some million meters above sea level. SpaceX must be jealous.
What a disappointment to see such an error in a new product (Mavic 2 Enterprise).

Hi SimonID, I understand your situation and very sorry for the troubles. Do you encounter similar issue when using Mavic 2 Enterprise? If yes, please kindly send us the original file in the SD card, I'll forward to the team for double check, thank you so much!

fans51675b83 Posted at 11-18 06:17
Susan.....i cant expressed how let down am feeling on these issue....especially as i submitted a proposal to a client using the M2P as a solution for their video/photo/ mapping needs in a portable package.....Now am being forced to look for alternatives for this potential big supply.

If its a deliberate attempt to force us to buy your other products like P4P then its a big shame and thumps down to DJI.......hoping you prioritize this and get possitive reviews like in BnH

I sincerely apologize for the troubles. The above data has been forwarded to our engineers and I'll keep following up.

fans51675b83 Posted at 11-18 06:17
Susan.....i cant expressed how let down am feeling on these issue....especially as i submitted a proposal to a client using the M2P as a solution for their video/photo/ mapping needs in a portable package.....Now am being forced to look for alternatives for this potential big supply.

If its a deliberate attempt to force us to buy your other products like P4P then its a big shame and thumps down to DJI.......hoping you prioritize this and get possitive reviews like in BnH

This is not the experience we want you to have. Our engineers are locating the root cause, we'll come back as long as there is an update.

DJI Susan Posted at 11-20 01:34
Hi SimonID, I understand your situation and very sorry for the troubles. Do you encounter similar issue when using Mavic 2 Enterprise? If yes, please kindly send us the original file in the SD card, I'll forward to the team for double check, thank you so much!

Yes, the M2E shows the exact same problem. I sent you a link to a sample image via PN

SimonID Posted at 11-21 01:47
Yes, the M2E shows the exact same problem. I sent you a link to a sample image via PN

Received with thanks! Could you provide us with the DNG file, the version of aircraft and controller?  I've forwarded to our engineers to check, thank you!

DJI Susan Posted at 11-23 03:03
Received with thanks! Could you provide us with the DNG file, the version of aircraft and controller?  I've forwarded to our engineers to check, thank you!

Hello again Susan,
I saw the new update for the Mavic 2 firmware (https://forum.dji.com/thread-172949-1-1.html#post_1655609). However, from the release note I have not found any information regarding the fix for altitude/exif issues... Do you have some good news for us that they have forgotten to add in the firmware's description?
Also the barrel distortion when recording in D-log seems has not been addressed....

Tomorrow I will try to update  the firmware throughout DJI assistant since from DJI Go app and Android device still persists the connection error that makes impossible the updating without a pc...another annoying issue for Android users...

Because I suppose that nothing has been done for the altitude metadata problem, please  could you explain us what are the problem that are causing so much troubles to DJI's engeneers ? Because add simple reading from the barometer into exif metadata in my opinion seems a quite trivial task for your team...
Best regards

TenenteDAN1986 Posted at 11-24 09:29
Hello again Susan,
I saw the new update for the Mavic 2 firmware (https://forum.dji.com/thread-172949-1-1.html#post_1655609). However, from the release note I have not found any information regarding the fix for altitude/exif issues... Do you have some good news for us that they have forgotten to add in the firmware's description?
Also the barrel distortion when recording in D-log seems has not been addressed....

Cant you see its a deliberate attempt by DJI to withhold that and made that available in another product......Think about it.....so now u have dji mapper....or something like that ...........so u have zoom....pro.....enterprise......mapper....

Surveying and mapping is a big business involving  the pix4d....dronedeploy etc cartel. So why wont DJI join the party?

Remember that mavic pro didnt have this issue...........cheap flight controllers dont have this issues....pixwack etc.........

maybe we need to exploit other geotagging devices and hang it on the drone and do a sync using time for geotag data..........

TenenteDAN1986 Posted at 11-24 09:29
Hello again Susan,
I saw the new update for the Mavic 2 firmware (https://forum.dji.com/thread-172949-1-1.html#post_1655609). However, from the release note I have not found any information regarding the fix for altitude/exif issues... Do you have some good news for us that they have forgotten to add in the firmware's description?
Also the barrel distortion when recording in D-log seems has not been addressed....

.

fans51675b83 Posted at 11-24 10:10
Cant you see its a deliberate attempt by DJI to withhold that and made that available in another product......Think about it.....so now u have dji mapper....or something like that ...........so u have zoom....pro.....enterprise......mapper....

Surveying and mapping is a big business involving  the pix4d....dronedeploy etc cartel. So why wont DJI join the party?

I really hope that your thought is different from  Dji marketing strategy, because in that case DJI would commit a fraud toward a lot of customers without expliciting that the new series are only usable for photography/filming and not for mapping. I don't like to have a third party product   to add  simple geotagging info. I hope that DJI will be able to solve this problem to keep the trust of ours costumers.

SimonID Posted at 11-21 01:47
Yes, the M2E shows the exact same problem. I sent you a link to a sample image via PN

Hi Simon, I just got an update from our engineers that the issue will be optimized in the future firmware update, and we'll release as soon as it is ready. Appreciate your attention.

TenenteDAN1986 Posted at 11-24 09:29
Hello again Susan,
I saw the new update for the Mavic 2 firmware (https://forum.dji.com/thread-172949-1-1.html#post_1655609). However, from the release note I have not found any information regarding the fix for altitude/exif issues... Do you have some good news for us that they have forgotten to add in the firmware's description?
Also the barrel distortion when recording in D-log seems has not been addressed....

Hi TenenteDAN1986, yes, the altitude issue has been optimized in the previous firmware update, and it was not included in the release note. Also, could you export the corresponding log in the aircraft when the issue appears? Our engineers would like to have more details to locate the root cause, thank you!

For the distortion when recording in D-log, it's a normal phenomenon. The video captured in the DLOG-M and HLG color modes is uniformly encoded using 10bits H.265 for greater post-processing space. But 10bits video cannot complete distortion correction in the aircraft directly, it needs to be post-processed, you may use the third-party processed software to fix. If you want an image that is not distorted too much, you may use the normal color mode for video shooting.
Regarding the firmware update, it is recommended to update via DJI Assistant 2 and see whether it works well.

DJI Susan Posted at 11-25 23:10
Hi TenenteDAN1986, yes, the altitude issue has been optimized in the previous firmware update, and it was not included in the release note. Also, could you export the corresponding log in the aircraft when the issue appears? Our engineers would like to have more details to locate the root cause, thank you!

For the distortion when recording in D-log, it's a normal phenomenon. The video captured in the DLOG-M and HLG color modes is uniformly encoded using 10bits H.265 for greater post-processing space. But 10bits video cannot complete distortion correction in the aircraft directly, it needs to be post-processed, you may use the third-party processed software to fix. If you want an image that is not distorted too much, you may use the normal color mode for video shooting.

Sorry Susan,
but  I joined this forum to explain my issue with altitude with the previous firmaware and like me also other usera have noted the same problem. I tell you again that only a 0 value is stored in the exif metadata of photos shooted with mavic 2 pro with firmware version  = v01.00.01.00 . I have updated just now both the aircraft and RC to v01.00.02.00 with dji assistant. Tomorrow I will shoot some picture and I send them with log files to you because I am sure that the problem still persists.

Regards
Daniele

TenenteDAN1986 Posted at 11-26 11:24
Sorry Susan,
but  I joined this forum to explain my issue with altitude with the previous firmaware and like me also other usera have noted the same problem. I tell you again that only a 0 value is stored in the exif metadata of photos shooted with mavic 2 pro with firmware version  = v01.00.01.00 . I have updated just now both the aircraft and RC to v01.00.02.00 with dji assistant. Tomorrow I will shoot some picture and I send them with log files to you because I am sure that the problem still persists.

Thank you so much! Look forward to your update.

DJI Susan Posted at 11-26 19:11
Thank you so much! Look forward to your update.

Hello Susan,
so finally I have both a very good and a little bad news for you and specially for MP2 mappers.
The new firmware (01.00.02.00) shows a very important improvement  compared to the old one regarding altitude information. In fact, even if I can't see any description on the official overview/release of the new firmware, I am happy to note that now in the Xmp metadata the relative altitude (estimate from the barometer) shows correct values  (see the two images on the left, one on the ground and the other at about 60 m).




However, the GPS altitude (above the MSL) stored in the exif is still very inaccurate with an error of 50 m (the right side of the images above). Despite the fact that my home point is no more than 2 m above the sea level the value reported is equal to 50.8 m. For this reason also the Absolute altitude stored in the Xmp is affected by this error.
Inaccuracies in the GPS positioning relative to the height  is  a common problem that affect also Phantom 4 and other non-DJI products with consumer-grade GPS receiver, so I can't consider it as a problem due to DJI's engineers. Maybe could be reduced because I don' t believe that on board GPS is so bad in estimating altitude (or at least not so worse than a smartphone GPS, which usually gives more acceptable results).
Indeed, considering this aspect I suggest to DJI of inserting in the exif (that are easier to read compared to Xmp) the relative altitude values directly read from the barometric pressure sensor instead of GPS altitude.

I am attaching  the original photo  (both dng and Jpeg) and please Susan, let me know if you still need logs, but I suppose that for now a agreement point has been reached.

https://drive.google.com/open?id ... dBZQYtu-NJxdrZTM23S

Kind regards and thanks for your support.

Daniele

TenenteDAN1986 Posted at 11-27 09:01
Hello Susan,
so finally I have both a very good and a little bad news for you and specially for MP2 mappers.
The new firmware (01.00.02.00) shows a very important improvement  compared to the old one regarding altitude information. In fact, even if I can't see any description on the official overview/release of the new firmware, I am happy to note that now in the Xmp metadata the relative altitude (estimate from the barometer) shows correct values  (see the two images on the left, one on the ground and the other at about 60 m).

I believe the GPS altitude is relative to the ellipsoid, and depending on where you are the difference between the ellipsoid height and MSL can be large. For the GPS reading to be accurate the drone would need to have a precise GEOID model for reference.

Here's a good explanation: https://eos-gnss.com/elevation-for-beginners/

TenenteDAN1986 Posted at 11-27 09:01
Hello Susan,
so finally I have both a very good and a little bad news for you and specially for MP2 mappers.
The new firmware (01.00.02.00) shows a very important improvement  compared to the old one regarding altitude information. In fact, even if I can't see any description on the official overview/release of the new firmware, I am happy to note that now in the Xmp metadata the relative altitude (estimate from the barometer) shows correct values  (see the two images on the left, one on the ground and the other at about 60 m).

Hi Daniele, thanks so much for keeping us updated. Could you refer to 72# and also export the corresponding log in the aircraft when the issue appears? I'll forward to our engineers for further analysis, thank you!

DJI Susan Posted at 11-29 00:38
Hi Daniele, thanks so much for keeping us updated. Could you refer to 72# and also export the corresponding log in the aircraft when the issue appears? I'll forward to our engineers for further analysis, thank you!

Hi Susan,
this is the link
https://drive.google.com/open?id ... 3jPT1MGGX6OslF83g3t
for the log of the fly when I shooted the photos of post #71.
So you can check thew large  error of GPS altitude data.
Best regards,
Daniele

TenenteDAN1986 Posted at 11-30 06:37
Hi Susan,
this is the link
https://drive.google.com/open?id=1y0_kdZDa6Ts_S3jPT1MGGX6OslF83g3t

I just looked at one of your images (DJI_0029.DNG)
Exiftool reports following:

Absolute Altitude               : +50.90
Relative Altitude               : +0.00
GPS Altitude                    : 50.8 m Above Sea Level
GPS Latitude                    : 42 deg 21' 1.95" N
GPS Longitude                   : 10 deg 55' 17.02" E

According to Google Earth Pro that given position is 6m above MSL.

Compute now the GPS altitude with reference to the GEOID (EGM96) we receive:

Latitude = 42.3505403° N = 42° 21' 1.95" N
Longitude = 10.9213955° E = 10° 55' 17.02" E
GPS ellipsoidal height = 6 (meters)
Geoid height = 48.357 (meters)

Taking into account the GPS accuracy may vary depending on varoius factors (i.e. VDOP)
we see that the EXIF GPS altitude data (50.8m) from your image is very close to the computed value.
The difference is only 2.44m, which is totally acceptable.

So, the data in the provided image is quite plausible.

Research the topic by reading the link provided in post #76.
If you do want crosscheck the data i provided, look here: GEOID height calculator.

Keule Posted at 11-30 08:28
I just looked at one of your images (DJI_0029.DNG)
Exiftool reports following:

Hi Keule,
thank you so much for your very informative post. So great news for everyone because the GPS altitude data are good but are referred to ellipsoid height  and so we have to covert them to orthometric height.  Also this post provide a very user-friendly description of the problem http://www.esri.com/news/arcuser/0703/geoid1of3.html

So now the question is why on our smartphones and handheld GPS receiver we can read a GPS altitude  that roughly match  the orthometric height? Are there geoid models that can be applied  automatically by our devices ? If so would be great if DJI could do the same. Of course a general geoid  model can only provide gross values but maybe could be more comfortable that covert the ellipsoid height by ourselves, for each images...or not?