Machine-learning-assisted self-improving object-identification system and method

A method of identifying and tracking objects, the method comprising the steps of:

• capturing an image of an object located in an area designated for holding objects;
• submitting a version of the image to a deep neural network to determine whether the deep neural network recognizes the object in the captured image;
• detect a location of the object while the object is in the area irrespective of whether the deep neural network recognizes the object;
• and
• retraining the deep neural network using the version of the image and object information associated with the detected object and the location of the object if the deep neural network is unable to recognize the object in the submitted image.

A sensor module comprising:

• an image sensor disposed in an area designated for holding objects and configured to capture an image within its field of view;
• a depth sensor disposed in the designated area for holding objects and having a field of view that overlaps the field of view of the image sensor, the depth sensor being configured to acquire depth data including estimated depth values for an image captured by the depth sensor;
• and
• a controller in communication with the image sensor and depth sensor to receive image data associated with the image captured by the image sensor and estimated depth values associated with the image captured by the depth sensor, the controller including one or more processors configured to:
• submit the image data associated with the image captured by the image sensor and the estimated depth values associated with the image captured by the depth sensor to a deep neural network to determine if the deep neural network recognizes an object in the image and the depth data, detect a location of the object in the image while the object is in the area irrespective of whether the deep neural network is able to recognize the object in the submitted image and the depth data, and retrain the deep neural network using the submitted image, the depth values associated with the image captured by the depth sensor, and object information about the detected object and the location of the object if the deep neural network is unable to recognize the object in the submitted image.