Correction of misaligned map data from different sources

• US 9,836,813 B1
• Filed: 03/31/2014
• Issued: 12/05/2017
• Est. Priority Date: 04/04/2011
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: Brewington, Brian Edmond
• Primary Examiner(s): Thompson, James A
• Application Number:
  US14/230,064
• Time in Patent Office: 1,345 Days
• Field of Search: 345629-643, 345682
• CPC Class Codes: G06T 3/14 Transformations for image registration, e.g. adjusting or mapping for alignment of images
• CPC Class Codes: G06T 3/147 using affine transformations

Abstract

Misaligned map data received from different sources is corrected to generate a map that includes aligned features. Each data source is associated with a reliability value that identifies the likelihood that the map data received from the corresponding source is aligned with a particular map location. A corrected version of the map data is generated based on the reliability values of the data sources. Generally, map data from unreliable sources is adjusted toward map data from more reliable sources until the map data from the different sources is aligned.

Claims

1. A computer-implemented method comprising:

receiving sets of map data from a plurality of sources, each source having a reliability value associated therewith,

wherein the sets of map data from each of the plurality of sources identify all distinct features from each other and correspond to a same map location;

determining, by one or more processors, whether the distinct features of the sets of map data from the plurality of sources are substantially aligned by evaluating whether a deviation corresponding to a visibility gap between the distinct features of the sets of map data satisfies a threshold value;

adjusting, using the one or more processors, at least one of the distinct features of the sets of map data based on the determination; and

generating a map using the adjusted at least one distinct feature of the set of map data.

2. The method of claim 1, wherein the reliability values associated with each source identifies a likelihood that map data received from that source is aligned with a corresponding map location.

3. The method of claim 1, wherein the adjusting comprises:

when the reliability value associated with a first one of the plurality of sources exceeds the threshold value and the reliability value associated with a second one of the plurality of sources does not exceed the threshold value, adjusting the distinct feature of the set of map data from the second one of the plurality of sources to be aligned with the distinct feature of the set of map data from the first one of the plurality of sources.

4. The method of claim 1, wherein one or more distinct features of the sets of map data from one or more of the plurality of sources that have a lowest reliability value are adjusted to be aligned with the one or more distinct features of the sets of map data from the one or more of the plurality of sources having a highest reliability values.

5. The method of claim 1, wherein:

the sets of map data include a first set of map data from a first one of the plurality of sources, a second set of map data from a second one of the plurality of sources, and a third set of map data from a third one of the plurality of sources;

the first one of the plurality of sources has a first reliability value, the second one of the plurality of sources has a second reliability value, and the third one of the plurality of sources has a third reliability value, the first reliability value being greater than the second reliability value and less than the third reliability value; and

the adjusting includes;

adjusting distinct features of the first set of map data towards distinct features of the third set of map data by a first amount, and

adjusting distinct features of the second set of map data towards distinct features of the third set of map data by a second amount, the second amount being greater than the first amount.

6. The method of claim 5, wherein the distinct features of the first set of image data includes outlines of roadways, the distinct features of the second set of image data include street level views and the distinct features of the third set of image data includes outlines of buildings.

7. The method of claim 1, wherein the reliability value of each source is an aggregate based on a statistically weighted mean value associated with a map location.

8. The method of claim 1, wherein:

the sets of map data include a first set of map data from a first one of the plurality of sources having a first reliability value and a second set of map data from a second one of the plurality of sources having a second reliability value; and

when the first reliability value is substantially the same as the second reliability value, a final version of the map is generated with first and second distinct features of the sets of map data being adjusted toward each other until the first and second distinct features of the sets of map data are aligned.

9. The method of claim 1, wherein the distinct features of the sets of map data from the plurality of sources include outlines of buildings, outlines of roadways, outlines of parcels and street level views.

10. A computer-implemented method comprising:

receiving sets of map data from a plurality of sources, wherein the sets of map data from each of the plurality of sources identify all distinct features from each other and correspond to a same map location, each source having a reliability value associated therewith, each reliability value being determined based on one or more of collection dates of the sets of map data and one or more physical device characteristics of devices that collected the sets of map data;

determining, by one or more processors, whether the distinct features of the sets of map data from the plurality of sources are substantially aligned by determining a misalignment score between selected distinct features of the sets of map data;

adjusting, using the one or more processors based on the misalignment score and the reliability values, at least one distinct feature of the sets of map data based on the determination, the adjusting including adjusting a first distinct feature of the sets of map data having a lower reliability value to be aligned to a second distinct feature of the sets of map data having a higher reliability value; and

generating a map using the adjusted at least one distinct feature of the set of map data.

11. The method of claim 10, wherein the reliability value of each of the plurality of sources is formed using an aggregate based on a statistically weighted mean value associated with a map location.

12. The method of claim 10, wherein each reliability value is further determined in accordance with how often each set of map data is updated.

13. The method of claim 10, wherein the distinct features of the sets of map data from the plurality of sources include outlines of buildings, outlines of roadways, outlines of parcels and street level views.

14. A system, comprising one or more processors configured to:

receive sets of map data from a plurality of sources, wherein the sets of map data from each of the plurality of sources identify all distinct features from each other and correspond to a same map location, each source having a reliability value associated therewith, each reliability value being determined based on one or more of collection dates of the sets of map data and one or more physical device characteristics of devices that collected the sets of map data;

determine whether the distinct features of the sets of map data from the plurality of sources are substantially aligned by determining a misalignment score between selected distinct features of the sets of map data;

adjust, based on the misalignment score and the reliability values, at least one distinct feature of the sets of map data based on the determination, and adjust a first distinct feature of the sets of map data having a lower reliability value to be aligned to a second distinct feature of the sets of map data having a higher reliability value; and

generate a map using the adjusted at least one distinct feature of the set of map data.

15. The system of claim 14, wherein the reliability value of each of the plurality of sources is formed using an aggregate based on a statistically weighted mean value associated with a map location.

16. The system of claim 14, wherein the reliability values associated with each source identifies a likelihood that map data received from that source is aligned with a corresponding map location.

17. The system of claim 14, wherein one or more distinct features of the sets of map data from one or more of the plurality of sources that have a lowest reliability value are adjusted to be aligned with one or more distinct features of the sets of map data from the one or more of the plurality of sources having a highest reliability values.

18. The system of claim 14, wherein:

the sets of map data include a first set of map data from a first one of the plurality of sources, a second set of map data from a second one of the plurality of sources, and a third set of map data from a third one of the plurality of sources;

the first one of the plurality of sources has a first reliability value, the second one of the plurality of sources has a second reliability value, and the third one of the plurality of sources has a third reliability value, the first reliability value being greater than the second reliability value and less than the third reliability value; and

the one or more processors are configured to perform the adjusting by;

adjusting distinct features of the first set of map data towards distinct features of the third set of map data by a first amount, and

adjusting distinct features of the second set of map data towards distinct features of the third set of map data by a second amount, the second amount being greater than the first amount.

19. The system of claim 18, wherein the distinct features of the first set of image data includes outlines of roadways, the distinct features of the second set of image data include street level views and the distinct features of the third set of image data includes outlines of buildings.

20. The system of claim 14, wherein the distinct features of the sets of map data from the plurality of sources include outlines of buildings, outlines of roadways, outlines of parcels and street level views.