Methods and systems for vehicle classification from laser scans using global alignment

Patent metadata

Field	Value
Publication number	US-9519060-B2
Application number	US-201414287538-A
Country	US
Kind code	B2
Filing date	May 27, 2014
Priority date	May 27, 2014
Publication date	Dec 13, 2016
Grant date	Dec 13, 2016

Abstract

A system and method for classifying vehicles from laser scan data by receiving laser scan data corresponding to multiple vehicles from a laser scanner; extracting vehicle shapes corresponding to the multiple vehicles based on the laser scan data; aligning the vehicle shapes; and generating vehicle profiles based on the aligned vehicle shapes. The system and method can further include aligning the vehicle shapes using sequence kernels, such as global alignment kernels, and constraining the sequence kernels based on determined weights.

First claim

What is claimed is:

1. A method for classifying vehicles comprising:
receiving laser scan data corresponding to a plurality of vehicles;
extracting, based on the laser scan data, a plurality of vehicle shapes corresponding to the plurality of vehicles;
aligning the plurality of vehicle shapes to generate a plurality of aligned vehicle shapes, wherein aligning the plurality of vehicle shapes comprises aligning the plurality of vehicle shapes using sequence kernels; and
generating one or more vehicle profiles based on the aligned vehicle shapes.

2. The method of claim 1 , wherein the sequence kernels are global alignment kernels.

3. The method of claim 2 , wherein sets of alignments corresponding to the global alignment kernels are constrained based on determined weights.

4. The method of claim 1 , further comprising classifying the plurality of vehicles by comparing the plurality of aligned vehicle shapes to the one or more vehicle profiles.

5. The method of claim 1 , further comprising:
receiving additional laser scan data corresponding to at least one additional vehicle;
extracting, based on the additional laser scan data, at least one additional vehicle shape corresponding to the at least one additional vehicle; and
classifying the at least one additional vehicle by comparing the at least one additional vehicle shape to the one or more vehicle profiles.

6. The method of claim 1 , wherein the plurality of vehicle shapes are two-dimensional shapes.

7. The method of claim 1 , wherein the plurality of vehicle shapes are three-dimensional shapes.

8. The method of claim 1 , further comprising, prior to extracting the plurality of vehicle shapes, replacing missing data in the laser scan data using interpolation based on the missing data's neighbor in the laser scan data.

9. The method of claim 1 , further comprising identifying and extracting data corresponding to vehicles of interest in the laser scan data, wherein extracting, based on the laser scan data, the plurality of vehicle shapes comprises extracting the plurality of vehicle shapes from the extracted data.

10. A system for classifying vehicles comprising:
a processing system comprising one or more processors; and
a memory system comprising one or more computer-readable media, wherein the one or more computer-readable media contain instructions that, when executed by the processing system, cause the processing system to perform operations comprising:
receiving laser scan data corresponding to a plurality of vehicles;
extracting, based on the laser scan data, a plurality of vehicle shapes corresponding to the plurality of vehicles;
aligning the plurality of vehicle shapes to generate a plurality of aligned vehicle shapes, wherein aligning the plurality of vehicle shapes comprises aligning the plurality of vehicle shapes using sequence kernels; and
generating one or more vehicle profiles based on the aligned vehicle shapes.

11. The system of claim 10 , wherein the sequence kernels are global alignment kernels.

12. The system of claim 11 , wherein sets of alignments corresponding to the global alignment kernels are constrained based on determined weights.

13. The system of claim 10 , the operations further comprising classifying the plurality of vehicles by comparing the plurality of aligned vehicle shapes to the one or more vehicle profiles.

14. The system of claim 10 , the operations further comprising:
receiving additional laser scan data corresponding to at least one additional vehicle;
extracting, based on the additional laser scan data, at least one additional vehicle shape corresponding to the at least one additional vehicle; and
classifying the at least one additional vehicle by comparing the at least one additional vehicle shape to the one or more vehicle profiles.

15. The system of claim 10 , wherein the plurality of vehicle shapes are two-dimensional shapes.

16. The system of claim 10 , wherein the plurality of vehicle shapes are three-dimensional shapes.

17. The system of claim 10 , the operations further comprising, prior to extracting the plurality of vehicle shapes, replacing missing data in the laser scan data using interpolation based on the missing data's neighbor in the laser scan data.

18. The system of claim 10 , the operations further comprising identifying and extracting data corresponding to vehicles of interest in the laser scan data, wherein extracting, based on the laser scan data, the plurality of vehicle shapes comprises extracting the plurality of vehicle shapes from the extracted data.

Assignees

• Xerox Corp

Inventors

• Chidlovskii Boris
• Csurka Gabriela
• Rodriguez Serrano Jose Antonio

Classifications

• G06V10/754

  involving a deformation of the sample pattern or of the reference pattern; Elastic matching · CPC title

• G01S17/42

  Simultaneous measurement of distance and other co-ordinates (indirect measurement G01S17/46) · CPC title

• G01S17/88 Primary

  Lidar systems specially adapted for specific applications · CPC title

• G06K9/00805

  Physics · mapped topic

• G01S7/4802

  using analysis of echo signal for target characterisation; Target signature; Target cross-section · CPC title