Method and system for determining the impact of crowding on retail performance

• US 8,812,344 B1
• Filed: 06/29/2009
• Issued: 08/19/2014
• Est. Priority Date: 06/29/2009
• Status: Active Grant

• Current Assignee: Videomining Corporation
• Original Assignee: Videomining Corporation
• Inventors: Saurabh, Varij, Jung, Namsoon, Sharma, Rajeev
• Primary Examiner(s): Pats, Justin M
• Assistant Examiner(s): GURSKI, AMANDA KAREN
• Application Number:
  US12/459,283
• Time in Patent Office: 1,877 Days
• Field of Search: 705/7.29
• US Class Current: 705/7.29
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06V 20/53 Recognition of crowd images, e.g. recognition of crowd congestion

Abstract

The present invention is a system, method, and apparatus for determining the impact of crowding on retail performance based on a measurement for behavior patterns of people in a store area. The present invention captures a plurality of input images of the people by at least a means for capturing images, such as cameras, in the store area. In the captured plurality of input images, each person'"'"'s shopping path is detected by a video analytics-based tracking algorithm. A subset of the people is identified as a crowd in the store area. In relation to the crowd, the behavior patterns of the target person are measured. After aggregating the measurements for the behavior patterns over a predefined window of time, the present invention can calculate a crowd index and a crowd impact index for the store area based on the measurements. A crowd index shows the level of crowd density in the store area caused by a crowd, including traffic count of the crowd in the store area. A crowd impact index comprises a traffic count of the target people who make trips to the store area and a shopping time index, such as average shopping time changes of the target people, in relation to a crowd in the measured store area.

Claims

1. A method for determining impact of crowding on retail performance based on a measurement for behavior patterns of people in a store area using computer vision-based behavior analysis and segmentation measurement, comprising the following steps of:

a) processing a plurality of input images in order to track each person among the people using a computer, by applying a computer vision-based tracking algorithm to the plurality of input images that are captured by a means for capturing images in the store area, wherein the plurality of input images are transferred to the computer via a means for video interface,

b) identifying a subset of the people as a crowd based on a first path analysis of tracks by tracking each person among the people,

c) measuring the behavior patterns of a person based on a second path analysis of tracks by tracking the person in relation to the crowd,

d) measuring segmentation of the person in relation to the crowd,

e) aggregating the measurements for the behavior patterns and segmentation over a predefined window of time, using the computer, and

f) calculating a crowd index and a crowd impact index for the store area based on the measurements, using the computer,

g) measuring elasticity of behavior of the people with respect to crowding,

wherein the elasticity is defined as a change in behavioral response, the elasticity changes, depending on season, occasion, time-of-day, or trip type, and the elasticity is measured per segment that includes a demographic group or a group of people with a same trip type, and

h) calculating an average density of sections in the store area over a predefined period of time,

wherein the density is measured based on traffic counts using the computer vision-based tracking of each person,

wherein the first path analysis comprises an application of a proximity rule among the tracks,

wherein the crowd impact index comprises a traffic count and a shopping time index of people outside the crowd and whose shopping activity is impacted by the crowd,

wherein the segmentation includes classification of demographic groups and trip types of the people, and

wherein the trip types include stock-up trip, fill-in trip, quick trip, and occasion-based trip.

2. The method according to claim 1, wherein the method further comprises a step of counting the number of people within a given radius of a person, as a relative measure among the people,

wherein a starting point of the radius is a center point of each person, and

wherein the size of the radius is adjusted to change granularity for calculating the crowd index and the crowd impact index.

3. The method according to claim 1, wherein the method further comprises a step of measuring different behaviors impacted by crowding, including u-turns, shopping time, traffic to shopper conversion rate, basket size, and sales,

wherein the traffic to shopper conversion rate is measured by calculating a percentage of shoppers among people who form traffic in the store area during the predefined window of time.

4. The method according to claim 1, wherein the method further comprises a step of measuring the impact of crowding by relating the incidence of u-turns with the number of people in the store area and measuring a loss caused by the incidence, wherein the loss comprises loss of sales, loss of dollar value, and loss of shopper traffic count in the store area.

5. The method according to claim 1, wherein the method further comprises a step of measuring the relationship of the crowd index and the crowd impact index with the performance of product categories in the store area,

wherein the relationship is analyzed according to the characteristics of the product categories,

including product category distribution and product category allocation.

6. The method according to claim 1, wherein the method further comprises a step of measuring the relationship of the crowd index and the crowd impact index with store layout in the store area,

whereby the measurement is used to optimize the size and shape of the aisle and to control crowd navigation.

7. The method according to claim 1, wherein the method further comprises a step of measuring the relationship of the crowd index and the crowd impact index with purchase movement between premeditated purchase movement and impulse purchase movement.

8. The method according to claim 1, wherein the method further comprises a step of calculating optimal shopper distance among the shoppers by measuring the distance between tracks in the crowd,

wherein the optimal shopper distance provides a level of crowding in the store area at which total sales are highest or sales per shopper are highest.

9. An apparatus for determining impact of crowding on retail performance based on a measurement for behavior patterns of people in a store area using computer vision-based behavior analysis and segmentation measurement, comprising:

a) means for capturing a plurality of input images of the people by at lust a means for capturing images in the store area,

b) a means for video interface that transfers the plurality of input images to a computer, and

c) the computer that is programmed to perform the following steps of;

processing the plurality of input images in order to track each person among the people, by applying a computer vision-based tracking algorithm to the plurality of input images that are captured by the means for capturing images,

identifying a subset of the people as a crowd based on a first path analysis of tracks by tracking each person among the people,

measuring the behavior patterns of a person based on a second path analysis of tracks by tracking the person in relation to the crowd,

measuring segmentation of the person in relation to the crowd,

aggregating the measurements for the behavior patterns and segmentation over a predefined window of time, using the computer,

calculating a crowd index and a crowd impact index for the store area based on the measurements,

measuring elasticity of behavior of the people with respect to crowding,

wherein the elasticity is defined as a change in behavioral response, the elasticity changes, depending on season, occasion, time-of-day, or trip type, and the elasticity is measured per segment that includes a demographic group or a group of people with a same trip type, and

calculating an average density of sections in the store area over a predefined period of time, wherein the density is measured based on traffic counts using the computer vision-based tracking of each person,

wherein the first path analysis comprises an application of a proximity rule among the tracks,

wherein the crowd impact index comprises a traffic count and a shopping time index of people outside the crowd and whose shopping activity is impacted by the crowd,

wherein the segmentation includes classification of demographic groups and trip types of the people, and

wherein the trip types include stock-up trip, fill-in trip, quick trip, and occasion-based trip.

10. The apparatus according to claim 9, wherein the apparatus further comprises a computer for counting the number of people within a given radius of a person, as a relative measure among the people,

wherein a starting point of the radius is a center point of each person, and

wherein the size of the radius is adjusted to change granularity for calculating the crowd index and the crowd impact index.

11. The apparatus according to claim 9, wherein the apparatus further comprises a computer for measuring different behaviors impacted by crowding, including u-turns, shopping time, traffic to shopper conversion rate, basket size, and sales,

wherein the traffic to shopper conversion rate is measured by calculating a percentage of shoppers among people who form traffic in the store area during the predefined window of time.

12. The apparatus according to claim 9, wherein the apparatus further comprises a computer for measuring the impact of crowding by relating the incidence of u-turns with the number of people in the store area and measuring a loss caused by the incidence,

wherein the loss comprises loss of sales, loss of dollar value, and loss of shopper traffic count in the store area.

13. The apparatus according to claim 9, wherein the apparatus further comprises a computer for measuring the relationship of the crowd index and the crowd impact index with the performance of product categories in the store area,

wherein the relationship is analyzed according to the characteristics of the product categories,

including product category distribution and product category allocation.

14. The apparatus according to claim 9, wherein the apparatus further comprises a computer for measuring the relationship of the crowd index and the crowd impact index with store layout in the store area,

whereby the measurement is used to optimize the size and shape of the aisle and to control crowd navigation.

15. The apparatus according to claim 9, wherein the apparatus further comprises a computer for measuring the relationship of the crowd index and the crowd impact index with purchase movement between premeditated purchase movement and impulse purchase movement.

16. The apparatus according to claim 9, wherein the apparatus further comprises a computer for calculating optimal shopper distance among the shoppers by measuring the distance between tracks in the crowd,

wherein the optimal shopper distance provides a level of crowding in the store area at which total sales are highest or sales per shopper are highest.