The Automated Drone Image Analysis Tool (ADIAT) provides a platform through which algorithms can be used to programmatically identify areas of interest in a set of images. The primary use case for this tool is to aid in the analysis of images taken by UAVs during Search and Rescue Operations.
This tool was developed by Texas Search and Rescue (TEXSAR) as an open source project for the SAR community. If you are interested in helping in developing the project please reach out to us at firstname.lastname@example.org
The color detection algorithm looks at each pixel in an image to determine if it is within a user-specified color range. If it find a section of pixels that is larger than the configured minimum size, it will add it to a list of areas of interest.
The color anomaly algorithm leverages the RX Anomaly Detector that is part of the Python Spectral library. This algorithm uses the squared Mahalanobis distance as a measure of how anomalous a pixel is with respect to an assumed background. The SPy rx function computes RX scores for an array of image pixels. The squared Mahalanobis distance is given byy.
These values are then compared against a configurable probability threshold (.999) by default.
ADIAT is written in Python and compiled for Windows. We have provided a Windows installer to make setup easier and are looking at providing a similar installer for macOS in the near future.