High density queue estimation and line management

• US 8,131,010 B2
• Filed: 07/30/2007
• Issued: 03/06/2012
• Est. Priority Date: 07/30/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Senior, Andrew W., Zhai, Yun
• Primary Examiner(s): Le, Vu
• Assistant Examiner(s): PARK, SOO JIN
• Application Number:
  US11/830,270
• Publication Number:
  US 20090034846A1
• Time in Patent Office: 1,681 Days
• Field of Search: 382/103, 382/190, 382/192, 348/150, 348/149
• US Class Current: 382/103
• CPC Class Codes: G06V 20/53 Recognition of crowd images, e.g. recognition of crowd congestion

Abstract

The present invention is directed to the measurement of attributes of a queue. A method for measuring an attribute of a queue in accordance with an embodiment includes: acquiring a plurality of images of a queue; extracting features from the images of the queue; analyzing the extracted features; and measuring the attribute based on the analysis of the extracted features; wherein the analyzing further comprises analyzing the extracted features at a plurality of successive time points to determine successive correspondences between the extracted features, and wherein the measuring further comprises measuring the attribute based on the successive correspondences

Claims

1. A method for measuring an attribute of a queue, comprising:

acquiring a plurality of images of a queue;

extracting features from the images of the queue;

analyzing, using at least one computer system, the extracted features; and

measuring the attribute based on the analysis of the extracted features, wherein the attribute comprises a fullness of the queue that is determined based on a waiting line length of the queue, a length of the queue, a number of objects in the queue, and a density of the queue, wherein the density of the queue is equal to a ratio of the number of objects in the queue to the waiting line length of the queue;

wherein the analyzing further comprises analyzing the extracted features at a plurality of successive time points to determine successive correspondences between the extracted features, wherein the measuring further comprises determining the fullness of the queue based on the successive correspondences, wherein the fullness of the queue is determined by comparing the waiting line length of the queue to the length of the queue, and wherein the queue is determined to be full if the waiting line length of the queue is about equal to the length of the queue and the density of the queue is above a predetermined threshold.

2. The method of claim 1, wherein the attribute further comprises an average moving speed at any point in the queue.

3. The method of claim 1, wherein the attribute further comprises an overall average moving speed of the entire queue.

4. The method of claim 1, wherein the attribute further comprises an average waiting time of an object to exit the queue at any point in the queue.

5. A system for measuring an attribute of a queue, comprising:

at least one computer system, including;

a system for acquiring a plurality of images of a queue;

a system for extracting features from the images of the queue;

a system for analyzing the extracted features; and

a system for measuring the attribute based on the analysis of the extracted features, wherein the attribute comprises a fullness of the queue that is determined based on a waiting line length of the queue, a length of the queue, a number of objects in the queue, and a density of the queue, wherein the density of the queue is equal to a ratio of the number of objects in the queue to the waiting line length of the queue;

wherein the system for analyzing is configured to analyze the extracted features at a plurality of successive time points to determine successive correspondences between the extracted features, wherein the system for measuring determines the fullness of the queue based on the successive correspondences, wherein the fullness of the queue is determined by comparing the waiting line length of the queue to the length of the queue, and wherein the queue is determined to be full if the waiting line length of the queue is about equal to the length of the queue and the density of the queue is above a predetermined threshold.

6. The system of claim 5, wherein the attribute further comprises an average moving speed at any point in the queue.

7. The system of claim 5, wherein the attribute further comprises an overall average moving speed of the entire queue.

8. The system of claim 5, wherein the attribute further comprises an average waiting time of an object to exit the queue at any point in the queue.

9. A program product stored on a non-transitory computer readable medium, which when executed, measures an attribute of a queue, the computer readable medium comprising program code for:

acquiring a plurality of images of a queue;

extracting features from the images of the queue;

analyzing the extracted features; and

measuring the attribute based on the analysis of the extracted features, wherein the attribute comprises a fullness of the queue that is determined based on a waiting line length of the queue, a length of the queue, a number of objects in the queue, and a density of the queue, wherein the density of the queue is equal to a ratio of the number of objects in the queue to the waiting line length of the queue;

wherein the analyzing further comprises analyzing the extracted features at a plurality of successive time points to determine successive correspondences between the extracted features, wherein the measuring further comprises determining the fullness of the queue based on the successive correspondences, wherein the fullness of the queue is determined by comparing the waiting line length of the queue to the length of the queue, and wherein the queue is determined to be full if the waiting line length of the queue is about equal to the length of the queue and the density of the queue is above a predetermined threshold.

10. A method for deploying an application for measuring an attribute of a queue, comprising:

providing a computer infrastructure being operable to;

acquire a plurality of images of a queue;

extract features from the images of the queue;

analyze the extracted features; and

measure the attribute based on the analysis of the extracted features, wherein the attribute comprises a fullness of the queue that is determined based on a waiting line length of the queue, a length of the queue, a number of objects in the queue, and a density of the queue, wherein the density of the queue is equal to a ratio of the number of objects in the queue to the waiting line length of the queue;

wherein the analyzing further comprises analyzing the extracted features at a plurality of successive time points to determine successive correspondences between the extracted features, wherein the measuring further comprises determining the fullness of the queue based on the successive correspondences, wherein the fullness of the queue is determined by comparing the waiting line length of the queue to the length of the queue, and wherein the queue is determined to be full if the waiting line length of the queue is about equal to the length of the queue and the density of the queue is above a predetermined threshold.