Kite Mapping
Pascaline Alexandre, Julien Barde , Sylvain Poulain, Sylvain Bonhommeau,
Emmanuel Chassot, Beenesh Motah
Summary: different steps
Expected products:
Scientific aspects
Goals, expected products & needs:
Citizen science ?
Different sensors:
Different platforms:
Tools and bravery:
Material
Sensors
Paddle
Kite board
Kite (aerial survey)
Citizen science
300€ < Cost < 1000 €:
Data :
Products :
Partnerships:
Citizen science => connected watch (or phone)
A few numbers and comments
50 sessions per year for one person (~ 1TB of photos)
Computing
Using infrastructures:
Material: good spots for good time data collection
Material: remote sensing waiting for ground truth
GPS tracks with a watch (or a phone..)
Material: kite GPS tracks (multiple sessions)
Material: kite GPS tracks (one session)
Material: surf GPS tracks
Material: paddle GPS tracks
Material: on the kite (aerial view)
Material: on the kite (aerial view)
Material: on the kite (aerial view)
Scientific aspects: image recognition
Paddle (corals)
Kite (corals)
Scientific aspects: image recognition
Kite (seagrass)
Kite (seagrass)
Scientific aspects: image recognition
Kite (algae)
Kite (sea cucumber)
Scientific aspects: image recognition
Kite
Snorkeling
Scientific aspects: image recognition
Kite
Paddle
Products: infer geolocation of photos (Postgis)
Products: image geolocation (google map example)
Deep Learning => need for scientific neural networks
Product: image recognition (Azure deep learning)
Product: image recognition (AWS deep learning)
Product: image recognition (AWS deep learning)
Product: photogrammetry (One Eye, Mauritius)
Summary: different steps
Expected products:
Fieldwork
Expected products:
Fieldwork : set up action cameras
Important points:
Summary: different steps
Expected products:
Phase 2 => Back at work
Metadata of sessions / deployments:
Phase 2 => Back at work: files structure & naming
Phase 2 => Back at work
All codes on a Github repository :
Geolocated photos => metadata storage in Postgis
Photos metadata stored in a “materialized view” of the postgis database:
Geolocated photos => metadata storage in Postgis
Phase 2 => Back at work
GPS tracks and photos processing
Main goals:
Postgres / Postgis database exploitation
Once metadata of photos are stored in a Postgis database, it is possible:
Summary: different steps
Expected products:
Post processing : example of products
=> Processes derived from aerial drones
Photogrammetry
Reminder =>
Photogrammetry
REAL WORLD 3D
2D IMAGES
DIGITAL 3D
Photogrammetry - processes
Tie points calculation and Bundle Adjustment
Photogrammetry - processes
Images calibration
Photogrammetry - processes
Sparse point cloud
Photogrammetry - processes
Dense point cloud
Photogrammetry - processes
Mesh + Texture generation (optional)
+
Photogrammetry - processes
DEM computation
Photogrammetry - processes
Orthophoto computation (from dense point cloud or DEM)
Photogrammetry - How to process ?
Main steps:
Quick Overview from user: open source softwares
�OpenDroneMap (ODM) WebODM
� MicMac
Full process from images to final product : Point Cloud + Mesh + DEM + Orthophoto
Accessing to WebODM:
Micmac interfaces
Micmac interfaces
From Lia Duarte
Product: photogrammetry
One Eye, Mauritius
Conclusion
Outlooks
Contacts
For more information, please contact us:
Github repository : https://github.com/juldebar/Deep_mapping