Menu

machine learning

Azavea publishes public cloud dataset for satellite imagery machine learning applications

2 Aug - 11:08 SAST
Detection and removal of clouds in satellite imagery is a natural application of machine learning, but freely available training datasets have been in short supply. To address this, Azavea has produced a public dataset...
+ Read more

Spot the Crop Data Challenge focuses on using satellite imagery and machine learning for automating South African crop identification

13 Jul - 15:14 SAST
Radiant Earth Foundation’s Spot the Crop Data Challenge seeks entrants to predict crop types in the Western Cape, South Africa using satellite image time-series.
+ Read more

Giga and Mapbox join forces with mapping gaming to locate schools across the globe

29 Jun - 22:55 SAST
To accelerate the mapping of schools globally, Giga and Mapbox created The Mapping Game, a crowdsourced app to identify schools using satellite imagery.
+ Read more

Report: Geospatial AI/ML applications and policies – A global perspective

13 Apr - 05:57 SAST
The report aims to inform governments and industry alike on the geospatial artificial intelligence (GeoAI) trends for the near-, mid- and long-term. Through the inputs from WGIC member companies and industry experts, areas of...
+ Read more

Paper: Bayesian statistics and modelling

3 Feb - 17:49 SAST
This paper describes the stages involved in Bayesian analysis, from specifying the prior and data models to deriving inference, model checking and refinement.
+ Read more

African contestants take top prizes at global satellite data labelling contest

25 Sep - 20:52 SAST
Participants were asked to identify cloudy pixels in Sentinel-2 scenes, to help develop a large-scale accurate cloud detection training dataset.
+ Read more

GroundWork adds validation, magic wand and other features to speed up image classification for machine learning

25 Aug - 11:11 SAST
Improvements in GroundWork includes validation, the introduction of a “magic wand” to speed up labelling, and a global label view and clickable task map.
+ Read more

Radiant Earth Foundation releases benchmark land cover training data for Africa

28 Jul - 15:14 SAST
Radiant Earth Foundation has released “LandCoverNet,” a human-labelled global land cover classification training dataset. This release contains data across Africa, which accounts for ~1/5 of the global dataset.
+ Read more

Picterra introduces a QGIS Plugin that allows users to access automated imagery processing from within their current workflows

12 Jun - 18:44 SAST
Picterra has released the beta version of its QGIS plugin to complement its online platform and to enable users to integrate Picterra into their workflow.
+ Read more

Aiding healthy bee populations with advanced analytics, machine learning and mapping

20 May - 14:55 SAST
With the number of bee colonies drastically declining around the world, SAS is using technology such as the Internet of Things (IoT), machine learning and visual analytics to help maintain and support healthy bee...
+ Read more

Mapillary introduces global dataset for lifelong place recognition

27 Apr - 15:42 SAST
Mapillary has released its Street-Level Sequences Dataset, a diverse publicly available dataset for lifelong place recognition. The dataset enables benchmarking and training machine learning models for large-scale lifelong place recognition. The dataset is available...
+ Read more

Crop detection in Africa competition winners announced

16 Apr - 23:33 SAST
Five data scientists emerged as winners of Radiant Earth Foundation’s competition, in partnership with Zindi Africa, to create a machine learning model that classifies farm fields in Kenya by crop type using time series...
+ Read more

Microsoft is building a planetary computer

15 Apr - 15:36 SAST
Microsoft is entering the next phase of its AI for Earth program, dedicated to building a Planetary Computer platform. This will provide the AI for Earth community (which includes nearly 500 organisations in 81...
+ Read more

Image segmentation and classification labelling for geospatial imagery

7 Apr - 14:49 SAST
GroundWork is a tool that creates training data for machine learning models run on geospatial imagery. The company has been using it internally for six months already, and is now opening access to everyone...
+ Read more

Picterra incorporates testing areas into workflow

12 Feb - 18:53 SAST
Picterra has released a new feature, Testing Areas, which is now an integral part of the workflow for training a detector. Its purpose is to visually assess the results while building the detector.
+ Read more

Automatically detect and count animals on aerial imagery

11 Feb - 17:28 SAST
Information from Picterra Measuring animal populations is often the first step in conservation efforts for taking well-prepared, effective action. The problem researchers face is how to detect and count animals...
+ Read more

Enhancing Pléiades and SPOT imagery for better object detection

7 Feb - 12:42 SAST
The Super-resolution algorithm uses machine learning (convolutional neural network) to increase the sharpness of low-resolution images, making it easier to extract information from. This is an important pre-processing step before using machine learning object...
+ Read more

Open geospatial training data released for crop type machine learning

9 Dec - 19:04 SAST
Radiant MLHub debuted its cloud-based open library of earth observation training data with “crop type” training data for major crops in Kenya, Tanzania and Uganda.
+ Read more