Blackshark.ai's Orca is a series of containers run as a service through the Microsoft Azure Synapse Analytics Marketplace. This computer vision service is designed to receive aerial images as input, detect one or many desired classes of object, and deliver output in either raster or polygonal vector data.
If this is the first time using Orca, please see initial setup for a step-by-step tutorial on getting started with Orca.
Orca's footprint detection model is trained to detect the footprint of buildings. Model BSHRK-W-BFD-11 is capable of detecting buildings from aerial image input and producing a high-quality probability heatmap of buildings. The image can then be vectorized for an accurate representation of building footprints.
Input imagery by Vexcel Imaging GmbH
Orca's vegetation detection model is trained to detect vegetation coverage. Model BSHRK-W-VGD-21 is capable of detecting vegetation coverage from aerial image input and producing a high-quality probability heatmap of vegetation coverage.
Input imagery by Airbus - UP42 - Airbus Pléiades