Lateral sign placement determination

• US 9,460,355 B2
• Filed: 10/14/2014
• Issued: 10/04/2016
• Est. Priority Date: 10/14/2014
• Status: Active Grant

• Current Assignee: HERE Global B.V. (Nokia Corporation)
• Original Assignee: HERE Global B.V. (Nokia Corporation)
• Inventors: Stenneth, Leon, Modica, Leo
• Primary Examiner(s): Chawan, Sheela C
• Application Number:
  US14/513,956
• Publication Number:
  US 20160104049A1
• Time in Patent Office: 721 Days
• Field of Search: 382/155, 382/103, 382/104, 382/156, 382/159, 382/165, 382/291, 382/224, 382/190, 700/48, 700/51, 706/20
• US Class Current: 1/1
• CPC Class Codes: G06F 18/217 Validation; Performance evaluation; Active pattern learning techniques
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/20 Ensemble learning
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]
• CPC Class Codes: G06V 10/768 using context analysis, e.g. recognition aided by known co-occurring patterns
• CPC Class Codes: G06V 20/582 of traffic signs
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

Systems, methods, and apparatuses are described for predicting the placement of road signs. A device receives data depicting road signs from multiple vehicles. The device analyzes a detected placement of the road signs and at least one characteristic of a collection of the data. The characteristic describes the road upon which the data was collected, an operation of the vehicle from which the data was collected, or an environment in which the data was collected. The device generates a model that associates values for the detected placement of the road signs with values for the at least one characteristic. The model may be later accessed to interpret subsequent sets of data describing one or more road signs.

Claims

1. A method comprising:

receiving image data depicting road signs from a plurality of vehicles;

analyzing, by a processor, a detected lateral placement of the road signs and at least one characteristic of a collection of the image data;

generating, by the processor, a model that associates values for the lateral placement of the road signs with values for the at least one characteristic;

receiving a subsequent image including a road sign collected under a subsequent characteristic;

identifying a plurality of road links in proximity to the road sign of the subsequent image; and

determining, based on the model that associates values for the lateral placement of the road signs with values for the at least one characteristic, a selected road link from the plurality of road links based on a subsequent detected placement of the road sign and the subsequent characteristic.

2. The method of claim 1, further comprising:

accessing the model that associates values for the lateral placement of the road signs with values for the at least one characteristic;

applying the subsequent characteristic to the model; and

receiving a predicted placement of the road sign from the model.

3. The method of claim 2, further comprising:

comparing the subsequent detected placement to the predicted placement.

4. The method of claim 3, further comprising:

modifying the subsequent detected placement when the subsequent detected placement and the predicted placement are mismatched.

5. The method of claim 3, further comprising:

identifying values from the road sign in the subsequent image when the subsequent detected placement matches the predicted placement.

6. The method of claim 1, wherein the at least one characteristic includes one or more of characteristics of a road link from which the image data depicting road signs is collected.

7. The method of claim 1, wherein the one or more characteristics of the road link includes whether a ramp, merge, specialized lane, or intersection is associated with the road link.

8. The method of claim 1, wherein the at least one characteristic includes environment conditions.

9. The method of claim 1, wherein the values for the detected lateral placement of the road signs include angle ranges, angles, or predetermined lateral placement positions.

10. The method of claim 9, wherein the predetermined lateral placement positions include any combination of overhead, right, and left.

11. The method of claim 1, wherein the model includes a machine learned model.

12. An apparatus comprising:

at least one processor; and

at least one memory including computer program code for one or more programs;

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least;

receive data describing road signs from a plurality of vehicles;

analyze a detected placement of the road signs with respect to a road, and at least one characteristic of a collection of the data;

generate a model that associates values for the detected placement of the road signs with values for the at least one characteristic;

receive a subsequent image including a road sign;

identify a plurality of road links in proximity to the road sign of the subsequent image; and

determine a selected road link from the plurality of road links based on the model.

13. The apparatus of claim 12, wherein the at least one characteristic includes one or more of properties of a road link from which the data describing road signs is collected or includes one or more environment conditions under which the data describing road signs was collected.

14. A method comprising:

receiving data describing at least one road sign;

accessing, by a processor, a historical learned model that associates values for placement of road signs with at least one characteristic of a collection of the data;

applying, by the processor, the at least one characteristic to the historical learned model;

receiving a predicted placement of the at least one road sign from the historical learned model;

analyzing, by the processor, the data based on the predicted placement of the at least one road sign;

identifying a road link associated with the collection of the data; and

assigning, to the road link, at least one value from the at least one road sign in the data when the detected placement matches the predicted placement.

15. The method of claim 14, wherein analyzing the data comprises:

identifying a detected placement of the at least one road sign in the data; and

comparing the detected placement to the predicted placement.

16. The method of claim 15, further comprising:

modifying the detected placement when the detected placement and the predicted placement are mismatched.

17. The method of claim 15, further comprising:

identifying values from the at least one road sign in the data when the detected placement matches the predicted placement.

18. An apparatus comprising:

at least one processor; and

at least one memory including computer program code for one or more programs;

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least;

receive data describing at least one road sign including a detected placement;

access a historical learned model that associates values for lateral placement of road signs with at least one characteristic of a collection of the data;

apply the at least one characteristic to the historical learned model;

receive a predicted lateral placement of the at least one road sign from the historical learned model;

analyze the data based on the lateral predicted placement of the at least one road sign;

identify a road link associated with the collection of the data; and

assign, to the road link, at least one value from the at least one road sign in the data when the detected placement matches the lateral predicted placement.