Point Cloud
2 Likes
Log files, first few runs it would not process my original project that wanted to share with over 7k images. So I decided to try it with the smaller data set collected with ebee and SODA sensor. Can you please add WALDO XCAM sensor to the database. We use these for very large data sets collected with the Cessna aircraft.
Can you expand the log file attachment size its capped at 4k?
This project was to support agriculture research at Colorado State University. We flew this corn field full of test plots almost every week for months to look for small changes and progression within each plot to see how different conditions and seed types handles. The data after years of trials is published within the agriculture industry to help change the way crops are grown in the future. Data was captured in RGB, multispectral, and thermal images - this is an example of one RGB flight. In addition to Pix4D we use specialized test plot software to analyze the field and that software requires very high resolution and high overlap images. This field has over 1,100 photos with a resolution of ~0.9cm/pixel.
4 Likes
Hi @daniel.helc, good idea for the WALDO XCAM. We could continue the conversation in this thread dedicated to adding more camera support to Pix4Dmatic if you’re interested: Camera for adding to supported list - #20 by Pierangelo_Rothenbuhler
As for the log file attachment, I’ll ping @Rodrigo_Goncalves.
Svínoy, Faroe Islands.
Svínoy is a remote island, in the north-east part of the Faroes, with a population of around 30 people.
This map is part of an ongoing project to map all urban areas in The Faroe Islands. With about 120 urban areas, one pilot, a flying season from about late april to mid. september, it takes about 3 years to get around to all the islands and their respective populated areas. And when we’ve been all the way around, we start over. The short season is due to the north atlantic weather conditions and lighting conditions in the winter half.
The end products are collected in an ArcGIS setup and displayed here.
This project was flown with an eBee X RTK in three flights, collecting 1996 images with an Aeria X camera.
SvinoyLog.zip (475.0 KB)
5 Likes
Good afternoon everyone, so this project was a good challenge, the end files were mainly used for hydrological modelling and calculations on a pipeline project (about 25km long).
It took two LONG days on the field, with a WingtraOne PPK VTOL, 5 PPK base stations as well as a rover, and after 9 400m. AGL flights. The result was 3152 42Mp photos, supplemented with 30 GCPs and control points for accuracy assessment and fine georeferencing.
For a final 5.75 cm GSD and around 25 sq. km project, I was more than satisfied with the results and accuracy obtained on Pix4Dmatic, also it was about 6 hours faster than on Pix4Dmapper, quite impressive!!
2020-11-11_05-30-21-PART1.txt (3.1 MB) 2020-11-11_05-30-21-PART2.txt (3.2 MB) 2020-11-11_05-30-21-PART3.txt (2.6 MB) 2020-11-11_05-30-21-PART4.txt (2.9 MB) 2020-11-11_05-30-21-PART5.txt (1.5 MB) 2020-11-11_05-30-21-PART6.txt (1.8 MB) 2020-11-09_16-45-22.txt (16.9 KB) 2020-11-10_07-41-22.txt (65.8 KB) 2020-11-11_14-21-29.txt (1.5 KB) quality_report.txt (3.4 KB) !
4 Likes
This is a complete drone scan for University of Sussex.
The scan took 1815 photos and around 40 GCP on ground to insure highest accuracy. Avg GSD is around 2.33cm.
The drone used for mapping is DJI Phantom 4 Pro.
Total scanned area is around 176 Acres (714,000 sqm)
Couldn’t include the log file in this post so I will attach it on the following one.
7 Likes
Here is my submission for the Community Challenge. The context was the refugee influx into Bangladesh from Myanmar in Sept 2018. And I was deployed with United Nations High Commissioner for Refugees UNHCR as part of their shelter team. We eventually received nearly 1 million refugees essentially in one month. The result was large camps the largest being Kutupalong (KTP) Camp of 1.1 million people. This is the 4th largest city in Bangladesh. We needed new tools to manage, plan and design such a camp-city and thus we developed the “drone today, map tomorrow” workflow. This map of KTP camp was achieved using a Mavic pro and PIX4D Capture. Authority to fly was through the RRR the Bangladesh Govt Agency responsible for refugees. At that time 10 km2 was seen as the upper limit for rotary UAV drones. But with planning and field trials 16km2 of KTP could be mapped in a day and then crunched using PIX4D Mapper. I note that PIX4Dmatic was much quicker at the crunching taking about half the time it took with PIX4D Mapper
.2020-11-13_09-03-59.txt (1.3 KB)
regards
Regan
6 Likes
This is a map of the Squire Course at PGA National as part of a $100MM renovation of the entire resort. The project consisted of 4,160 images captured with a Phantom 4 RTK connected to CORS. The project took 2.5 days of data capture at 200’ AGL creating a GSD of 1.8cm/.06ft. 36 ground targets were measured with 9 GCPs utilized in Pix4Dmapper processing and 27 as Checkpoints with a Z RMSE of .13’. The project was independently “ground-truthed” by a survey firm which sealed the dataset for project permitting requirements.
The architect used the dataset to document current conditions and topography giving him the knowledge and information to redesign the golf course.
Pix4Dmatic processed the dataset significantly faster at 8h to generate dense point cloud.
6 Likes
project log file is too large to upload
1 Like
This project was for mapping the properties of a small town in the rural area of the state of Minas Gerais here in Brazil. Orthomosaic and point cloud were used to generate polygons for each property to update the city’s cadastral base. I used a Phantom 4 Pro to capture 8,842 images with Pix4DCapture.With 26 Ground Control Points and GDS = 3.27 cm. When it was done, it was necessary to divide it into 8 projects to generate the maps. And now it was done in just 4 days with the Pix4Dmatic the project all 8,842 images at once with a small machine.
Platform: Windows
Operating System: Windows 7 Professional, 64-bit
CPU: Intel (R) Core ™ i7-4790 CPU @ 3.60GHz, cpus = 1, threads = 8
RAM: 15.91 GB
GPU renderer: GeForce GTX 760 / PCIe / SSE2
GPU vendor: NVIDIA Corporation
GL version: 4.1.0 NVIDIA 390.65
Quality_report
Version: 1.5.0
Project name: DDI-FULL-011
Date: 05 November 2020
Project details
Average GSD: 3.27054 cm
Project coordinate system: SIRGAS 2000 / UTM zone 23S + EGM96 height [EGM96]
Camera name: DJI FC6310
Number of images: 8842
Calibration
- Image Scale: 1/1
- Maximum extracted key points: Automatic
- Reoptimized: true
- Duration: 2h 20m 2s
Completed: 99.9435%
Calibrated images: 8837
Uncalibrated images: 5
Camera optimization: 0.228771%
Median number of matches per camera: 2,405
Ground Control Points (GCPs)
Number of GCPs: 26
GCP coordinate system: SIRGAS 2000 / UTM zone 23S + EGM96 height [EGM96]
Units: meter
GCP error [cm]:
mean position error x, mean position error y, mean position error z
0.00757413, 0.00783585, 0.0024229
sigma position error x, sigma position error y, sigma position error z
0.00630797, 0.00688342, 0.00186903
RMS position error x, RMS position error y, RMS position error z
0.00985688, 0.0104299, 0.00306002
Densification
- Density: Optimal
- Minimum matches: 3
- Duration: 19h 53m 14s
Number of points: 619,886,226
5 Likes
A perfect Butterfly Effect: Small changes in emissions can have big impact on future
Abstract:
The city named Valsad in Gujarat state of India is under substantial erosion of beaches and coastal wetlands. This city coastal region is sandwiched between Arabian sea and Wanki River, due to which it is highly vulnerable to coastal flooding by extreme events (storm, wave) and longterm sea level rise. The DEM, 3D view and orthomosaic is required for further coastal protection measures. Here i shared the output corresponding to 360 acres.
A hash tag for the village #be CALM (Sea) 
# be CALM (River) 
# be COOL (Urban) 
Climate change (Global warming > provoke Nature)
- Sea level rise, Extreme waves and storms (# be CALM)
- flash floods (# be CALM)
- Urban (# be COOL)
Application:
- Impact of climate change on coastal environment
- Coastal zone management
- Potential source of information for coastal policy planner and decision makers
Similar to this a lot more studies are required over an 8000km coastline of India.
Acknowledgement:
Grateful 
to all PIX4D team for changing the survey industry much simpler 
, lightening 
and affordable 
.
logfile link: 2020-11-13_01-00-06.txt - Google Drive
5 Likes
This is a 5-mile corridor flown with a Phantom 4 Pro v2. We flew it at 200ft AGL and collected 2282 photos. Initially processed in Pix4DMapper using multiple sub-projects and merging. Very impressed with Pix4DMatic which was able to process the entire data set in a fraction of the time with great accuracy.
2020-11-11_14-58-19.zip (434.5 KB)
4 Likes
Open pit mine project originally processed with Pix4D Mapper now with Matic. There was problem with one GCP but I did not care because I was just testing Matic. 4916 images captured with P4PV2 flying mostly automatically with waypoints generated by our own 3D flight route design program. Average GSD 2.5 cm, thats abot 0.08 ft. We have done same size open pit mine with 1 cm GSD that means that we are flying typically closer than 40 meters from the walls.
1.231 km2 / 123.1378 ha / 0.48 sq. mi. / 304.4377 acres
CPU: Intel(R) Core™ i7-9700K CPU @ 3.60GHz
RAM: 64GB
GPU: NVIDIA GeForce RTX 2070 SUPER (Driver: 26.21.14.4187)
Project details
Average GSD: 2.54978 cm
Project coordinate system: KKJ / Finland Uniform Coordinate System + EGM84 height [geoid height ~ -2.55 m]
Camera name: DJI FC6310, DJI FC6310S
Number of images: 4916
Calibration
- Image Scale: 1/1
- Maximum extracted key points: Automatic
- Reoptimized: false
- Duration: 6h 5m 8s
Densification
- Density: Optimal
- Minimum matches: 3
- Duration: 9h 2m 14s
Number of points: 479,076,804
Digital Surface Model (DSM)
- Resolution(cm/px): 1 GSD 2.55 cm
- Surface smoothing: 2
- Interpolation: true
- Duration: 41m 54s
Image resolution: 57822x61538 px
Best Regards!
Matti Hytola
Hytola ENgineering Oy
2 Likes
The preservation of the environment is of vital importance for future generations, in the same way the ecosystems that fight every day to maintain their natural balance.
The Valley of the Moon, located in the City of La Paz - Bolivia, is one of the most important and best preserved natural ecosystems in the city, Neil Armstrong was the one who baptized this place as Valley of the Moon, seeing the similarity that it existed with the landscapes of the moon.
However, by not having a Conservation Project, the probability of losing it is growing.
Therefore, I saw the opportunity through this contest to carry out the Photogrammetric Survey of this area, which will allow that, in the future, I hope not too distant, the collected data will be used either for an analysis of deterioration or for the elaboration of an Ecotourism Conservation Project.
Thanks.
Jose
5 Likes