Support Website Contact Support Blog

Processing issue Step 1 and Step 2 densification

Images (GoPro Hero4Black 3.0 3000x2250 RGB) used for initial processing (Step 1) were taken in wide mode.

I thought a data display issue or a graphics card driver not up to date but it seems that this is not the case

As you will see in the images attached below, the LAS file generated in Step 2 is not good. In XYZ export, it’s the same problem.

The points of densification of the cloud are not random, they are on in lines (spaced in X of 6 cm and spaced in Y of 50 cm).

The images come from GoPro 4 (WIDE mode), we do not use the Pixhawk flight logs, we have failed with GCP points from a geometer survey.

Do you know where this “bug” comes from?

Thank you in advance for your clarification.

STEP 1:

STEP 2:

Densified point cloud in Global Mapper :

Hi Alain,

Has the project been created with the same version as it was processed? Example: created in 4.3X and processed in the same one?

We have observed a similar isolated case when the project has been initially created in 3.X.X  and then re-processed in 4.X.X.

If this is the case could you please try to re-create the project from scratch in the latest available version and process step 1 and 2 and see if this behavior persists?

Best,
Ina

Hi Ina,

Thank you for answering me

I actually created the project with version 4.1.25 of Pix4 D and then re-processed step 1 and 2 with versions 4.2.27 and 4.3.31.

I will try to re-create the project from scratch in the latest available version and I will keep you informed.

Best,

Alain

Hi,

Can you let us know if you were able to reprocess the project with 4.3.31? :slight_smile:

What are your results? 
Cheers!

Hi,

First, I tried to reprocess the project with 4.2.27 and today I have the same trouble.


Non densified cloud

Densified cloud

Today, I will try with 4.3.31.

Best

Alain

 

Hi Alain,

Thank you for getting back to us. 
Let us know once you have the results from 4.3.31.

Best

Hi,

The results after the Step 1 are not good. (processed with 4.3.31)

If I do not add GCP after Step 1, the non densified cloud is ok but if I add GCP and I Reoptimize, the non densified cloud is not ramdom.

Look for the following screen capture:

 

Best

Alain

Hi Alain,

Can I ask you to share this project with us? Our team will investigate the issue! :slight_smile:

Thank you in advance

Hi Beata,

Our project is on the Pix4DCloud:

https://cloud.pix4d.com/pro/project/392950/map?shareToken=53a623fa-513b-4608-a594-c4332219fd86

https://cloud.pix4d.com/pro/project/392950/map?shareToken=5301e95a-95e5-43cb-9ff3-c13873a2d4ab

Thank you

Alain

Thank you very much!

I will come back to you as soon as I have some news.

Cheers

Hi there Alain. Thanks for uploading your project to the cloud. Unfortunately, the project doesn’t have the GCPs and we cannot reproduce the issue unless we have them. Could you please send us the GCP coordinates file? Or even just paste the coordinates in this thread?

We are investigating the issue and we’ll keep you posted.

Thanks heaps!

Best regards

Amritha

(Pix4D Test engineer)

Hi Amritha,

Here is the measured gcps position from 1_initial_params:

962528.73600000003352761269  4679332.37700000032782554626  5.57000000000000028422
962513.94099999999161809683  4679366.65799999982118606567  2.47000000000000019540
962491.59299999999348074198  4679365.10500000044703483582  6.61000000000000031974
962540.71400000003632158041  4679358.62999999988824129105  2.24000000000000021316
962572.36899999994784593582  4679404.31199999991804361343  1.17999999999999993783
962583.51199999998789280653  4679400.37200000043958425522  1.41999999999999992895
962584.87300000002142041922  4679401.41000000014901161194  1.25000000000000000000
962585.74899999995250254869  4679400.22800000011920928955  1.36000000000000009770
962584.34900000004563480616  4679399.17300000041723251343  1.38999999999999990230
962583.90899999998509883881  4679425.57000000029802322388  0.61999999999999999556
962636.78899999998975545168  4679438.56500000040978193283  1.38999999999999990230
962588.06599999999161809683  4679461.18699999991804361343  0.76000000000000000888
962588.37399999995250254869  4679491.92300000041723251343  0.71999999999999997335
962587.32200000004377216101  4679534.06599999964237213135  0.79000000000000003553
962701.98699999996460974216  4679515.77799999993294477463  1.25000000000000000000
962742.06700000003911554813  4679517.83000000007450580597  3.87999999999999989342
962623.61399999994318932295  4679586.68499999959021806717  0.93000000000000004885
962642.77099999994970858097  4679578.11799999978393316269  1.01000000000000000888
962687.74899999995250254869  4679576.82299999985843896866  0.95999999999999996447
962706.27599999995436519384  4679574.34900000039488077164  2.62999999999999989342
962680.82099999999627470970  4679608.08700000029057264328  2.27000000000000001776

Here is the estimated_gcps_position from 1_initial_params:

0, 962528.75248984445352107286, 4679332.36160616297274827957, 5.52988149239296955528
2, 962491.57524972106330096722, 4679365.12647271435707807541, 6.61622156345513001696
5, 962583.50459214905276894569, 4679400.38250207714736461639, 1.42291189813233498462
9, 962583.94103543518576771021, 4679425.56980424746870994568, 0.61039659415260505426
10, 962636.77780134743079543114, 4679438.55778922885656356812, 1.39495989422877175734
12, 962588.36662184179294854403, 4679491.91802635509520769119, 0.70217278682507833221
14, 962702.00457477755844593048, 4679515.78200196381658315659, 1.23893470155015839040
15, 962742.01065406959969550371, 4679517.84080024715512990952, 3.85121609779867846157
16, 962623.58481595665216445923, 4679586.68697658833116292953, 0.86834117065646931888
19, 962706.29357602179516106844, 4679574.35197771433740854263, 2.69089847670651050038
20, 962680.82856852817349135876, 4679608.10015064571052789688, 2.24415430981135832766

I hope it will help you.

Thanks

Alain

 

Thanks for sharing this information Alain. But could you please provide us with the initial GCPs that you used and not the ones that were generated after calibration? Also if you could please send us over the GCP and output coordinate system that you used? The images in your project are not geotagged.

 

Thanks and best regards

Amritha

Hi,

The initial  GCPs we used : (the output coordinate system is North UTM31 WGS84)

962528.736,4679332.377,5.57
962513.941,4679366.658,2.47
962491.593,4679365.105,6.61
962540.714,4679358.63,2.24
962572.369,4679404.312,1.18
962583.512,4679400.372,1.42
962584.873,4679401.41,1.25
962585.749,4679400.228,1.36
962584.349,4679399.173,1.39
962583.909,4679425.57,0.62
962636.789,4679438.565,1.39
962588.066,4679461.187,0.76
962588.374,4679491.923,0.72
962587.322,4679534.066,0.79
962701.987,4679515.778,1.25
962742.067,4679517.83,3.88
962623.614,4679586.685,0.93
962642.771,4679578.118,1.01
962687.749,4679576.823,0.96
962706.276,4679574.349,2.63
962680.821,4679608.087,2.27

Thanks and best regards

Bonjour, j’ai le même problème, avez-vous trouvé la solution?

Merci

Bonjour Tomas,

Pouvez-vous vérifier que votre carte graphique est bien à jour?

Bonjour,

oui ma carte est à jour.

Quand je géoreference avec le système de coordonnées de sortie arbitraire, cela fonctionne :

mais quand je bascule dans le système RGF / Lambert 93 cela me créé ceci :

donc cela génère ce nuage, ce n’est pas bien grave au final car quand je lis le fichier .las sur CloudCompare le nuage est bon.

Donc es-ce un bug d’affichage lié au système de coordonnée ? 

Merci 

 

Cher Thomas,

Merci pour ton partage.
Pour l’instant je ne peux pas vraiment t’indiquer l’orgine du bug.
Toutefois nos développeurs seraient intéressés par jeter un coup d’oeil à ton projet afin de tenter de reproduire l’erreur pour en comprendre l’origine.
Est-ce que tu pourrais partager avec nous le fichier *.p4d, toutes tes images et le fichier *.log?
Tu peux les charger ici:

One drive

Merci d’avance!
Cordialement

Bonjour

Il se trouve que j’ai le même problème sur différents projets.

Merci

Bonjour Jérémy,

Pouvez vous partager un projet avec nous avec le fichier *.p4d, toutes tes images et le fichier *.log?
Tu peux les charger ici:
one drive

Merci d’avance!
Cordialement