System and method for identifying parking spaces and parking occupancy based on satellite and/or aerial images

1. A method for identifying and evaluating on-street parking spots of an area based on at least one of satellite and aerial images of said area, the method comprising

retrieving and processing a plurality of at least one of satellite and aerial images by at least quality benchmarking and georeferencing the images; and

for each image, detecting objects of interest comprising at least detected vehicles, computing street sections based on map data corresponding to the images, and assigning the detected objects to at least street sections; and

for each street section, identifying on-street vehicle lanes based on the detected objects of interest; and

combining street sections based on the on-street vehicle lanes from the plurality of at least one of satellite and aerial images; and

in the combined sections, identifying parking lanes and deriving individual parking spots comprised thereon, and

computing availability of parking spaces in a given neighborhood at a given time.

2. The method according to claim 1 further comprising consolidating and interpreting data related to the identified individual parking spots.

3. The method according to claim 1 further comprising computing parking space occupancy of the identified parking lanes.

4. The method according to claim 1 further comprising using a plurality of at least one of satellite and aerial images of an area taken over a certain period of time to identify at least time-dependent parking space availability.

5. The method according to claim 1 further comprising identifying parking rules based on the interpreted identified individual parking spots.

6. A method for identifying on-street parking spots of an area based on at least one of satellite and aerial images of said area, the method comprising

retrieving and processing a plurality of at least one of satellite and aerial images by at least quality benchmarking and georeferencing the images, and

for each image, detecting objects of interest comprising at least detected vehicles, computing street sections based on map data corresponding to the images, and assigning the detected objects to at least street sections, and

for each street section, identifying on-street vehicle lanes based on the detected objects of interest, and

combining street sections based on the on-street vehicle lanes from the plurality of at least one of satellite and aerial images, and

in the combined sections, identifying parking lanes and deriving individual parking spots comprised thereon,

wherein identifying on-street vehicle lanes further comprises computing a closest street section for each detected vehicle and recursively identifying on-street vehicle lanes based on a plurality of vehicles present in each street section.

7. The method according to claim 6 wherein the recursively identified on-street vehicle lanes are quality controlled by at least comparing their slope with that of the respective street section.

8. The method according to claim 7 wherein the recursively identified on-street vehicle lanes not compliant with the quality control are further compared to similar lanes compliant with the quality control and adjusted to comply as well by recursively adapting their slope.

9. A method for identifying on-street parking spots of an area based on at least one of satellite and aerial images of said area, the method comprising

retrieving and processing a plurality of at least one of satellite and aerial images by at least quality benchmarking and georeferencing the images, and

for each image, detecting objects of interest comprising at least detected vehicles, computing street sections based on map data corresponding to the images, and assigning the detected objects to at least street sections, and

for each street section, identifying on-street vehicle lanes based on the detected objects of interest, and

combining street sections based on the on-street vehicle lanes from the plurality of at least one of satellite and aerial images, and

in the combined sections, identifying parking lanes and deriving individual parking spots comprised thereon,

wherein combining street sections comprises inputting a plurality of intersecting images, removing vehicles likely located off-street, and consolidating on-street vehicle lanes between the street sections and wherein identifying parking lanes further comprises computing the distance between the vehicles located in each on-street vehicle lane and wherein the method further comprises assigning an identification parameter to each on-street vehicle lane.

10. A method for identifying on-street parking spots of an area based on at least one of satellite and aerial images of said area, the method comprising

retrieving and processing a plurality of at least one of satellite and aerial images by at least quality benchmarking and georeferencing the images, and

for each image, detecting objects of interest comprising at least detected vehicles, computing street sections based on map data corresponding to the images, and assigning the detected objects to at least street sections, and

for each street section, identifying on-street vehicle lanes based on the detected objects of interest, and

combining street sections based on the on-street vehicle lanes from the plurality of at least one of satellite and aerial images, and

in the combined sections, identifying parking lanes and deriving individual parking spots comprised thereon,

wherein identifying individual parking spots comprises computing a mean and minimal distance between neighboring vehicles and determining orientation of parked vehicles with respect to their respective parking lane.

11. A method for identifying on-street parking spots of an area based on at least one of satellite and aerial images of said area, the method comprising

retrieving and processing a plurality of at least one of satellite and aerial images by at least quality benchmarking and georeferencing the images; and

detecting vehicles and street sections in the georeferenced images; and

identifying on-street vehicle lanes based on the detected street sections and vehicles; and

identifying parking lanes and deriving individual parking spots on them; and

consolidating data from the plurality of at least one of satellite and aerial images of the area to compute an average parking occupancy in said area,

wherein identifying on-street vehicle lanes comprises recursively assigning vehicles on a given street section to possible on-street vehicle lanes until an optimal solution yielding at least one lane is obtained.

12. A method for identifying on-street parking spots of an area based on at least one of satellite and aerial images of said area, the method comprising

retrieving and processing a plurality of at least one of satellite and aerial images by at least quality benchmarking and georeferencing the images, and

detecting vehicles and street sections in the georeferenced images, and

identifying on-street vehicle lanes based on the detected street sections and vehicles, and

identifying parking lanes and deriving individual parking spots on them; and

consolidating data from the plurality of at least one of satellite and aerial images of the area to compute an average parking occupancy in said area,

wherein the parking spot derivation comprises at least computing distance between nearest neighbor vehicles on each parking lane, determining types of vehicles and parking orientation and obtaining an average parking spot with a corresponding size based on the above.

13. A computer-implemented system for identifying on-street parking spots based on at least one of satellite and aerial images, the system comprising

a storage component configured to store a plurality of at least one of satellite and aerial images and map data; and

a processing component configured for retrieving and processing a plurality of at least one of satellite and aerial images from the storage component by at least quality controlling and georeferencing the images; and

for each image, detecting objects of interest comprising at least detected vehicles, computing street sections based on map data corresponding to the images, and assigning each detected object to a street section for each street section, identifying on-street vehicle lanes based on the detected objects of interest; and

combining street sections based on the on-street vehicle lanes from the plurality of at least one of satellite and aerial images; and

in the combined sections, identifying parking lanes and individual parking spots comprised thereon;

an output component configured to output the individual parking spots determined by the processing component,

wherein the processing component is further configured for computing availability of parking spaces in a given neighborhood at a given time.

14. The method according to claim 11 further comprising deriving rules related to parking from the consolidated data.

15. The method according to claim 11 further comprising consolidating the detected street sections with the respective identified lanes and identifying the parking lanes as the outermost lanes in the resulting consolidated street sections.

16. The system according to claim 13 wherein the processing component is further configured for identifying parking rules based on the identified individual parking spots.

17. The system according to claim 13 wherein the output component comprises at least one of

an application for a user's personal computing device that assists a user with parking spot finding; and

an interface, for third parties to obtain access to known parking spots in a given area.