Order from Lab 6B384F

Aerial ground truth

Commercial aerial imagery (aircraft or drone, 30 cm per pixel or finer, RGB plus near infrared, georeferenced) where a person went on site and recorded what was there: crop type per field, damage grade per building, with the visit date, GPS point and a ground photo. Each label must come from the visit, not from reading the image, a model, or a public map, and must sit within 14 days of the flight (crop) or pair a before and after image of a dated event (damage).

How answering works

Any number of people fill one order. The lab receives one dataset, in its own columns.

Your agent reads the order
It looks at what you hold and says which columns you can fill, and what is missing.
Mora checks what you send
Every record: no copies, no personal data, not already public, not on Mora from someone else.
The lab pays per record it accepts
The lab named the price. You fill as much of the order as you can.

The columns the lab wants

You do not need every column. Mora says which ones each dataset fills.

Column Type What it must hold

scene_id id One orthorectified image tile

image GeoTIFF (cloud-optimized) The tile itself: RGB plus near infrared, original bit depth, 0.2 to 5 GB each

footprint polygon (WGS84) Where the tile sits on the ground, exact, not blurred

gsd_cm number Real ground resolution in centimetres per pixel

captured_at datetime (UTC) When the flight took the image

sensor / platform / flight_id text, id Camera model, aircraft or drone, and the flight the tile came from

obs_id id One on-site visit to one field or building

geometry polygon or GPS point The field parcel or building footprint the surveyor visited

observed_at datetime Date of the ground visit

method label How the label was made: walked the field, inspected inside, seen from the road, farmer declaration, claim payout

label label Crop type, or building damage grade

label_scheme text The code list or scale used (national crop codes, EMS-98, FEMA, insurer scale)

ground_photo image with its own GPS and time Photo taken on site, proving the visit

gps_accuracy_m number Accuracy of the surveyor's position

surveyor_id generated id Who made the visit, as an id, so a surveyor's day can be checked

days_between number Days between the flight and the visit

event_id / event_date / pre_scene_id / post_scene_id ids, date For damage: the event, and the image before and after it

second_label label A second surveyor's label on 2% of sites, to measure agreement

Column	Type	What it must hold
`scene_id`	id	One orthorectified image tile
`image`	GeoTIFF (cloud-optimized)	The tile itself: RGB plus near infrared, original bit depth, 0.2 to 5 GB each
`footprint`	polygon (WGS84)	Where the tile sits on the ground, exact, not blurred
`gsd_cm`	number	Real ground resolution in centimetres per pixel
`captured_at`	datetime (UTC)	When the flight took the image
`sensor / platform / flight_id`	text, id	Camera model, aircraft or drone, and the flight the tile came from
`obs_id`	id	One on-site visit to one field or building
`geometry`	polygon or GPS point	The field parcel or building footprint the surveyor visited
`observed_at`	datetime	Date of the ground visit
`method`	label	How the label was made: walked the field, inspected inside, seen from the road, farmer declaration, claim payout
`label`	label	Crop type, or building damage grade
`label_scheme`	text	The code list or scale used (national crop codes, EMS-98, FEMA, insurer scale)
`ground_photo`	image with its own GPS and time	Photo taken on site, proving the visit
`gps_accuracy_m`	number	Accuracy of the surveyor's position
`surveyor_id`	generated id	Who made the visit, as an id, so a surveyor's day can be checked
`days_between`	number	Days between the flight and the visit
`event_id / event_date / pre_scene_id / post_scene_id`	ids, date	For damage: the event, and the image before and after it
`second_label`	label	A second surveyor's label on 2% of sites, to measure agreement

How much, in the lab's words

150,000 fields visited on site over 3 growing seasons (40+ crops, 5+ regions) and 80,000 buildings inspected on site over 12+ damage events, with their imagery: about 40,000 km² of ground counted once, about 30 TB. I count in visited sites and km², never in tiles.

scenes
ground-truth
links
event-pairs
repeat-survey
rights