UFO.track

giant.ufo.ufo_class:

UFO.track()

This method packages the possible UFO detections for the Tracker, passes them to the tracker, and then attempts to track the UFOs from image to image.

The packaging of the detections essentially boils down to building a scipy.spatial cKDTree on the (x_raw, y_raw) pixel locations for each detection, and retrieving the id for each UFO detection (the index of the Detector.detection_data_frame). This data is then passed to the tracker, and the Tracker.track() method is called. The results are then stored in the Tracker.confirmed_filters and Tracker.confirmed_standard_deviations attributes.

You can filter which possible detections are fed to the tracker using the tracking_quality_code_minimum attribute.

Note that this method can take a while to run, will use multi-processing, and will likely take up a lot of memory (depending on the number of images/number of detections per image).