| CPC G06V 40/23 (2022.01) [B64D 47/08 (2013.01); G06V 10/462 (2022.01); G06V 10/751 (2022.01)] | 40 Claims |
|
1. A gesture recognition system, the gesture recognition system comprising:
a computer system; and
a machine learning model manager in the computer system, wherein the machine learning model manager is configured to:
identify temporal color images for a set of gestures used for ground operations for an aircraft;
generate optical flow data identifying a distribution of visual velocities of a movement of a set of brightness patterns in the temporal color images on a per image basis from the temporal color images;
generate saliency maps identifying movement in the temporal color images using image segmentation in which a saliency map in the saliency maps is generated on the per image basis, wherein the temporal color images, the optical flow data and the saliency maps form training data;
train a set of feature machine learning models to recognize features using the training data; and
train a set of classifier machine learning models to recognize gestures using the features recognized by the set of feature machine learning models trained using the training data, wherein the set of feature machine learning models and the set of classifier machine learning models form a gesture recognition machine learning model system.
|