MANDALA - EarthInsight
An automated change detection toolkit for processing and generating observed change information products from myriad electro-optical earth observation sensors.
Introducing our flagship software for our EarthInsight product line, Mandala-EI Change Detection.
Change Detection is an arduous process; manually selecting the best image pairs to be used for a period of observation, pre-processing those images to increase the probability of positive change detection, and manual human interpretation of change results. Mandala spares its users all of those time and resource expenses.
Mandala-EI is fully automated change detection perfected. Mandala was designed from the ground up to contend with the most seemingly insoluble problems associated with image to image change detection. Mandala can be depended upon to deliver only real change results, with no false positives normally caused by a wide variety of external influences. Mandala delivers standardized and customizable change products inclusive of a wide variety of software formats and visualization engines.
Key Features & Advantages
- Accepts raw images, without calibration
- Accepts image pairs with up to ten-year time deltas between observations
- Accepts image pairs collected by different sensors and platforms
- Accepts image pairs collected by different imaging modalities
- Accepts image pairs collected from airborne versus satellite platforms
- Accepts image pairs with different times of collection, look angle, solar azimuth, seasons parallax, adjacency and imaging geometry
- Mandala change results are impervious to the effects of imaging geometry, look angle, and image artifacts
- Mandala change results are impervious to instrument and environmental noise
- Mandala change results can be masked or labelled by thematic classes
- Mandala requires no human interventions
- Mandala is invariant to scaling and can run on massive datasets
- Mandala can be integrated into many cloud or other large data platforms and service engines
End Products & Method of Delivery
- A binary change/no change raster image in any standard raster format
- Vectorized points and polygons of changes with thematic labels in any standard vector format
- Heat maps overlain on color imagery where the heat gradient corresponds to the most rapid and intense changes across multiple observations
- A chi-square error accuracy analysis image
- Machine readable meta-data for use in statistical analysis, data visualization and related functions
- Mandala can run and deliver products integrated within external platforms or within Exogenesis AWS infrastructure. Products can be delivered through a wide variety of web services and APIs.
This exemplar observes refugee camp activity by comparing RapidEye-3 to RapidEye-4 Pan-Sharpened with high resolution panchromatic imagery. The required image pair consists of a t0 (t-zero) image from August of 2015 and a t1 (t-one) image from August of 2016. The below images (Figures 1 – 5) illustrate change detected via point density, heat maps, and binary change/no change polygons respectively.
How TO PURCHASE
If you would like to have Exogenesis perform change detection for your project or if you would like to inquire about software licensing options please contact us below.
Mandala-EI is also embedded on some of our partner's platforms and can be purchased through them. Click the participating partner links below to learn more.