Back to results
US 12,183,201 B1Granted
Machine learning parking determination
OwnerKrish Arorainventor
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, methods, and computer program products for machine learning parking determination. A method includes receiving an image of a parking spot taken by an image sensor. A method includes processing an image of a parking spot using a machine learning model to determine a likelihood that the parking spot is occupied. A method includes communicating to a user, on an electronic display screen of a hardware computing device, whether a parking spot is occupied or available based on a determined likelihood that the parking spot is occupied.
Claims14
An apparatus, comprising:
- multiple image sensors disposed at different parking lots, the different parking lots comprise ng multiple parking spots each and each having different distances from a destination;
- an electronic hardware parking device in communication with the multiple image sensors, the electronic hardware parking device comprising a processor and a memory, the memory storing computer program code executable by the processor to perform operations, the operations comprising:
- receiving images taken by the image sensors, each of the images comprising multiple parking spots at one of the different parking lots;
- presenting a graphical user interface to administrators for each of the different parking lots allowing the administrators to graphically define, in the images of the multiple parking spots, locations of the multiple parking spots by selecting at least two diagonal corners for each of the multiple parking spots within the graphical user interface;
- processing the images of the multiple parking spots using a machine learning model to determine likelihoods that each of the multiple parking spots are occupied;
- predicting, using a second machine learning model, whether one or more of the parking spots are likely to be occupied or available at a future time based on the likelihoods that the parking spots are occupied and on a plurality of previous determinations of likelihoods that the parking spots are occupied;
- and
- communicating to a user, on a hardware computing device, a recommendation of one or more of the different parking lots for an unoccupied parking spot, based on the determined likelihoods that the parking spots are occupied, the predictions of whether one or more of the parking spots are likely to be occupied, and on a distance away from the destination that a user is willing to park.
5 more elements
A computer program product comprising a non-transitory computer readable storage medium storing computer program code executable to perform operations, the operations comprising:
- receiving images taken by the image sensors, each of the images comprising multiple parking spots at one of the different parking lots;
- presenting a graphical user interface to administrators for each of the different parking lots allowing the administrators to graphically define, in the images of the multiple parking spots, locations of the multiple parking spots by selecting at least two diagonal corners for each of the multiple parking spots within the graphical user interface;
- processing the images of the multiple parking spots using a machine learning model to determine likelihoods that each of the multiple parking spots are occupied;
- predicting, using a second machine learning model, whether one or more of the parking spots are likely to be occupied or available at a future time based on the likelihoods that the parking spots are occupied and on a plurality of previous determinations of likelihoods that the parking spots are occupied;
- and
- communicating to a user, on a hardware computing device, a recommendation of one or more of the different parking lots for an unoccupied parking spot, based on the determined likelihoods that the parking spots are occupied, the predictions of whether one or more of the parking spots are likely to be occupied, and on a distance away from the destination that a user is willing to park.
3 more elements
A method comprising:
- receiving images taken by the image sensors, each of the images comprising multiple parking spots at one of the different parking lots;
- presenting a graphical user interface to administrators for each of the different parking lots allowing the administrators to graphically define, in the images of the multiple parking spots, locations of the multiple parking spots by selecting at least two diagonal corners for each of the multiple parking spots within the graphical user interface;
- processing the images of the multiple parking spots using a machine learning model to determine likelihoods that each of the multiple parking spots are occupied;
- predicting, using a second machine learning model, whether one or more of the parking spots are likely to be occupied or available at a future time based on the likelihoods that the parking spots are occupied and on a plurality of previous determinations of likelihoods that the parking spots are occupied;
- and
- communicating to a user, on a hardware computing device, a recommendation of one or more of the different parking lots for an unoccupied parking spot, based on the determined likelihoods that the parking spots are occupied, the predictions of whether one or more of the parking spots are likely to be occupied, and on a distance away from the destination that a user is willing to park.
3 more elements
Dependencies are inferred from the claim text.