Method of and device for acquiring information on a traffic line of persons

• US 6,654,047 B2
• Filed: 10/20/1999
• Issued: 11/25/2003
• Est. Priority Date: 10/27/1998
• Status: Expired due to Fees

• Current Assignee: Toshiba Tec Corporation (Toshiba Corporation)
• Original Assignee: Toshiba Tec Corporation (Toshiba Corporation)
• Inventors: Iizaka, Hitoshi
• Primary Examiner(s): Kelley, Chris
• Assistant Examiner(s): Bugg, George A
• Application Number:
  US09/422,003
• Publication Number:
  US 20030164878A1
• Time in Patent Office: 1,497 Days
• Field of Search: 348/143, 382/103, 382/116, 382/118
• US Class Current: 348/143
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A feature vector creating section creates a feature vector unique to a customer on the basis of the images of a person picked up by cameras in a store. An entrance information storage section stores entrance information obtained by adding the identification code, entrance time, etc. to the feature vector of the incoming person. A customer collating section collates the feature vector of a person leaving and moving around the store with the feature vectors stored in the entrance information storage section. When the collation has found the same customer, the customer'"'"'s identification code, pickup time, pickup place, and others are stored as traffic-line information into a traffic-line storage section. The route that a customer took is found by picking up pieces of the traffic-line information with the same identification code.

Claims

1. A method of acquiring information on a traffic line of persons representing routes taken by the persons, comprising:

sensing existence of respective persons and extracting images of the respectively sensed persons from images picked up by a plurality of image pickup devices which are provided in respective predetermined positions in a facility and which respectively pick up images of persons entering the facility, leaving the facility, and moving within the facility;

creating feature vectors unique to the respectively sensed persons by image-processing the extracted images of the persons entering the facility, leaving the facility, and moving within the facility;

adding information including an identification code and an entrance time to each of the feature vectors created with respect to the persons entering the facility to thereby create entrance information, and then storing the entrance information;

collating the feature vectors created with respect to the persons leaving the facility and moving within the facility with the feature vectors contained in said stored entrance information to thereby discriminate respective same persons;

storing traffic-line information including the identification codes of the respectively discriminated same persons, positions of the image pickup devices picking up images of the same persons, and times at which the image pickup devices have picked up the images of the same persons; and

acquiring traffic-line information with a same identification code from the stored traffic-line information, determining respective positions of the same persons based on the positions of the image pickup devices picking up the images of the same persons, and providing a route of each person moving within said facility.

2. The method according to claim 1, wherein the extracted images are face images.

3. The method according to claim 1, wherein said feature vectors are lower-dimensional vectors created by principal component analysis.

4. The method according to claim 1, further comprising:

judging attributes of the respectively sensed persons,

wherein said feature vectors include said attributes and the collating is performed based on said attributes.

5. The method according to claim 4, wherein said attributes include the sex of said respectively sensed persons.

6. The method according to claim 4, wherein the route of each person is provided by attribute.

7. The method according to claim 1, further comprising:

preserving past visit information obtained by adding a visit date to said entrance information; and

collating subsequently created feature vectors with respect to persons entering the facility with the feature vectors included in said past visit information to discriminate respective same persons,

wherein when a same person is discriminated, the past visit date and a preceding identification code of said same person are added to the past visit information and preserved.

8. The method according to claim 7, further comprising:

providing a visit pattern including tendency data on frequency of visits and visit times of respective persons based on said past visit information.

9. The method according to claim 1, further comprising:

preserving past traffic-line information obtained by adding a visit date to said traffic-line information; and

referring to said past traffic-line information and providing a past route of each person moving around said facility.

10. The method according to claim 4, further comprising:

preserving past visit information obtained by adding a visit date to said entrance information.

11. The method according to claim 10, further comprising:

providing a visit pattern including tendency data by attribute on frequency of visits and visit times of respective persons based on said past visit information.

12. The method according to claim 10, further comprising:

preserving past traffic-line information obtained by adding a visit date to said traffic-line information; and

referring to said past traffic-line information and past visit information and providing a past route by attribute of each person moving around said facility.

13. A device for acquiring information on a traffic line of persons representing routes taken by the persons, comprising:

a plurality of image pickup devices which are provided in respective predetermined positions in a facility and which respectively pick up images of persons entering the facility, leaving the facility, and moving within the facility;

means for sensing existence of respective persons and extracting images of the respectively sensed persons from the images picked up by said image pickup devices;

means for creating feature vectors unique to the respectively sensed persons by image-processing the extracted images of the persons entering the facility, leaving the facility, and moving within the facility;

means for adding an identification code and an entrance time to each of the feature vectors created with respect to the persons entering the facility to thereby create entrance information, and for then storing the entrance information;

means for collating the feature vectors created with respect to the persons leaving the facility and moving within the facility with the feature vectors contained in said stored entrance information to thereby discriminate respective same persons;

means for storing traffic-line information including the identification codes of the respectively discriminated same persons, positions of the image pickup devices picking up images of the same persons, and times at which the image pickup devices have picked up the images of the same persons; and

means for acquiring traffic-line information with a same identification code from the stored traffic-line information, determining respective positions of the same persons based on the positions of the image pickup devices picking up the images of the same persons, and providing a route of each person moving within said facility.

14. The device according to claim 13, wherein the extracted images are face images.

15. The device according to claim 13, wherein said feature vectors are lower-dimensional vectors created by principal component analysis.

16. The device according to claim 13, further comprising:

means for judging attributes of the respectively sensed persons,

wherein said feature vectors include said attributes and the collating is performed based on said attributes.

17. The device according to claim 16, wherein said attributes include the sex of said respectively sensed persons.

18. The device according to claim 16, wherein the route of each person is provided by attribute.

19. The device according to claim 13, further comprising:

means for preserving past visit information obtained by adding a visit date to said entrance information; and

means for collating subsequently created feature vectors with respect to persons entering the facility with the feature vectors included in said past visit information to discriminate respective same persons,

wherein when a same person is discriminated, the past visit date and a preceding identification code of said same person are added to the past visit information and preserved.

20. The device according to claim 13, further comprising:

means for preserving past traffic-line information obtained by adding a visit date to said traffic-line information; and

means for referring to said past traffic-line information and providing a past route of each person moving around said facility.