IP Library › Granted Patent US 12710285
Granted Patent B2
US 12710285 · App. 18/888,541 · Granted Aug 18, 2026

Conflation between target and source road polylines from GIS databases

Inventor: Joao Andre de Franca Ferreira Galinho (Jersey City, NJ)
Assignee: Trimble Inc.
G01C21/387G01C21/3815G01C21/3867G01C21/3896
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12710285
App. No.
18/888,541
Granted
Aug 18, 2026
Kind
B2
Abstract

Described herein are systems, methods, and other techniques for performing conflation between target road polylines and source road polylines stored in GIS databases. A pairing of a target road polyline and a source road polyline is obtained. The target road polyline and the source road polyline each include a sequence of connected 2D or 3D vertices that form a continuous line. A similarity score is computed between the target road polyline and the source road polyline based on one or more similarity metrics. A conflation classification is assigned to the target road polyline, the source road polyline, or the pairing based on the similarity score. The conflation classification is one of a set of possible conflation classifications. A post-classification action is performed based on the conflation classification.

Claims (75)

1 . A computer-implemented method of performing conflation between a set of target road polylines and a set of source road polylines, the computer-implemented method comprising:

obtaining a pairing of a target road polyline and a source road polyline, the target road polyline from the set of target road polylines and the source road polyline from the set of source road polylines, wherein the target road polyline and the source road polyline each include a sequence of connected vertices that form a continuous line;

computing a similarity score between the target road polyline and the source road polyline based on one or more similarity metrics;

assigning a conflation classification to the target road polyline, the source road polyline, or the pairing based on the similarity score, the conflation classification being one of a set of possible conflation classifications; and

performing a post-classification action on the set of target road polylines as stored in a target database based on the conflation classification, the post-classification action including at least one of:

adding the source road polyline to the set of target road polylines as stored in the target database;

modifying the sequence of connected vertices of the target road polyline as stored in the target database based on the sequence of connected vertices of the source road polyline; or

deleting the target road polyline from the target database.

2 . The computer-implemented method of claim 1 , wherein the one or more similarity metrics include at least one of:

a distance between the target road polyline and the source road polyline;

a distance delta between the target road polyline and the source road polyline;

an overlap between the target road polyline and the source road polyline; or

an angle between the target road polyline and the source road polyline.

3 . The computer-implemented method of claim 2 , wherein the distance between the target road polyline and the source road polyline is computed by:

projecting each of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections and identifying a maximum distance of the target-to-source projections; or

projecting each of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections and identifying a maximum distance of the source-to-target projections.

4 . The computer-implemented method of claim 2 , wherein the distance delta between the target road polyline and the source road polyline is computed by:

projecting each of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections, identifying a maximum distance and a minimum distance of the target-to-source projections, and computing a difference between the maximum distance and the minimum distance of the target-to-source projections; or

projecting each of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections, identifying a maximum distance and a minimum distance of the source-to-target projections, and computing a difference between the maximum distance and the minimum distance of the source-to-target projections.

5 . The computer-implemented method of claim 2 , wherein the overlap between the target road polyline and the source road polyline is computed by:

projecting two of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections that connect with the source road polyline at a set of connection points along the source road polyline and computing a distance between the set of connection points along the source road polyline; or

projecting two of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections that connect with the target road polyline at a set of connection points along the target road polyline and computing a distance between the set of connection points along the target road polyline.

6 . The computer-implemented method of claim 2 , wherein the angle between the target road polyline and the source road polyline is computed by:

projecting two of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections that connect with the source road polyline at a set of connection points along the source road polyline and computing an angle between a first vector formed by the two of the connected vertices of the target road polyline and a second vector formed by the set of connection points along the source road polyline; or

projecting two of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections that connect with the target road polyline at a set of connection points along the target road polyline and computing an angle between a first vector formed by the two of the connected vertices of the source road polyline and a second vector formed by the set of connection points along the target road polyline.

7 . The computer-implemented method of claim 2 , wherein computing the similarity score between the target road polyline and the source road polyline based on one or more similarity metrics includes setting the similarity score to zero if the one or more similarity metrics are greater than or less than one or more corresponding thresholds.

8 . The computer-implemented method of claim 1 , wherein the set of source road polylines are retrieved from a government database.

9 . The computer-implemented method of claim 1 , wherein performing the post-classification action includes modifying one or more of the set of target road polylines as stored in a target database.

10 . A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

obtaining a pairing of a target road polyline and a source road polyline, the target road polyline from a set of target road polylines and the source road polyline from a set of source road polylines, wherein the target road polyline and the source road polyline each include a sequence of connected vertices that form a continuous line;

computing a similarity score between the target road polyline and the source road polyline based on one or more similarity metrics;

assigning a conflation classification to the target road polyline, the source road polyline, or the pairing based on the similarity score, the conflation classification being one of a set of possible conflation classifications; and

performing a post-classification action on the set of target road polylines as stored in a target database based on the conflation classification, the post-classification action including at least one of:

adding the source road polyline to the set of target road polylines as stored in the target database;

modifying the sequence of connected vertices of the target road polyline as stored in the target database based on the sequence of connected vertices of the source road polyline; or

deleting the target road polyline from the target database.

11 . The non-transitory computer-readable medium of claim 10 , wherein the one or more similarity metrics include at least one of:

a distance between the target road polyline and the source road polyline;

a distance delta between the target road polyline and the source road polyline;

an overlap between the target road polyline and the source road polyline; or

an angle between the target road polyline and the source road polyline.

12 . The non-transitory computer-readable medium of claim 11 ,

wherein the distance between the target road polyline and the source road polyline is computed by:

projecting each of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections and identifying a maximum distance of the target-to-source projections; or

projecting each of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections and identifying a maximum distance of the source-to-target projections.

13 . The non-transitory computer-readable medium of claim 11 , wherein the distance delta between the target road polyline and the source road polyline is computed by:

projecting each of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections, identifying a maximum distance and a minimum distance of the target-to-source projections, and computing a difference between the maximum distance and the minimum distance of the target-to-source projections; or

projecting each of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections, identifying a maximum distance and a minimum distance of the source-to-target projections, and computing a difference between the maximum distance and the minimum distance of the source-to-target projections.

14 . The non-transitory computer-readable medium of claim 11 , wherein the overlap between the target road polyline and the source road polyline is computed by:

projecting two of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections that connect with the source road polyline at a set of connection points along the source road polyline and computing a distance between the set of connection points along the source road polyline; or

projecting two of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections that connect with the target road polyline at a set of connection points along the target road polyline and computing a distance between the set of connection points along the target road polyline.

15 . A system comprising:

one or more processors; and

a computer-readable medium comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

obtaining a pairing of a target road polyline and a source road polyline, the target road polyline from a set of target road polylines and the source road polyline from a set of source road polylines, wherein the target road polyline and the source road polyline each include a sequence of connected vertices that form a continuous line;

computing a similarity score between the target road polyline and the source road polyline based on one or more similarity metrics;

assigning a conflation classification to the target road polyline, the source road polyline, or the pairing based on the similarity score, the conflation classification being one of a set of possible conflation classifications; and

performing a post-classification action on the set of target road polylines as stored in a target database based on the conflation classification, the post-classification action including at least one of:

adding the source road polyline to the set of target road polylines as stored in the target database;

modifying the sequence of connected vertices of the target road polyline as stored in the target database based on the sequence of connected vertices of the source road polyline; or

deleting the target road polyline from the target database.

16 . The system of claim 15 , wherein the one or more similarity metrics include at least one of:

a distance between the target road polyline and the source road polyline;

a distance delta between the target road polyline and the source road polyline;

an overlap between the target road polyline and the source road polyline; or

an angle between the target road polyline and the source road polyline.

17 . The system of claim 16 , wherein the distance between the target road polyline and the source road polyline is computed by:

projecting each of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections and identifying a maximum distance of the target-to-source projections; or

projecting each of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections and identifying a maximum distance of the source-to-target projections.

18 . The system of claim 16 , wherein the distance delta between the target road polyline and the source road polyline is computed by:

projecting each of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections, identifying a maximum distance and a minimum distance of the target-to-source projections, and computing a difference between the maximum distance and the minimum distance of the target-to-source projections; or

projecting each of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections, identifying a maximum distance and a minimum distance of the source-to-target projections, and computing a difference between the maximum distance and the minimum distance of the source-to-target projections.

19 . The system of claim 16 , wherein the overlap between the target road polyline and the source road polyline is computed by:

projecting two of the connected vertices of the target road polyline onto the source road polyline to generate target-to-source projections that connect with the source road polyline at a set of connection points along the source road polyline and computing a distance between the set of connection points along the source road polyline; or

projecting two of the connected vertices of the source road polyline onto the target road polyline to generate source-to-target projections that connect with the target road polyline at a set of connection points along the target road polyline and computing a distance between the set of connection points along the target road polyline.