NZ605651B - Tracking system - Google Patents

Abstract

605651 Disclosed is a system for tracking people within a customer purchasing facility. The system includes a plurality of sensors (1) arranged so that between them they cover overlapping areas (2) within the facility, and each sensor (1) can sense the direction of movement of a person within the area (2) that it covers. A processing means (3) is arranged to receive and process information from the sensors (1) indicative of the direction of movement. The processing means (3) is adapted to determine when a matching direction of movement has been detected by way of and between more than one of the sensors (1), to attribute such movement to the same person and, based on such determination, to progressively track movement of the person through the facility in the same manner but using other combinations of the sensors (1). The processing means (3) is adapted to determine shopping characteristics of the person based on the tracking. area (2) that it covers. A processing means (3) is arranged to receive and process information from the sensors (1) indicative of the direction of movement. The processing means (3) is adapted to determine when a matching direction of movement has been detected by way of and between more than one of the sensors (1), to attribute such movement to the same person and, based on such determination, to progressively track movement of the person through the facility in the same manner but using other combinations of the sensors (1). The processing means (3) is adapted to determine shopping characteristics of the person based on the tracking.

Description

Patents Form No. 5 THE PATENTS ACT 1953 COMPLETE SPECIFICATION TRACKING SYSTEM We, REVEAL LIMITED, a New Zealand company, of care of NSA Ltd Level 6, 3 City Road Grafton Auckland 1010, New Zealand, hereby declare this invention for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement: TITLE Tracking System.

FIELD OF INVENTION This invention relates to means for tracking movement.. A preferred form of the invention relates to a system for tracking the movement of consumers within a retail area or the like. However it should be appreciated that the invention is not limited to this.

BACKGROUND It is known to monitor the movement of consumers in retail or other areas. This has been done by physically following them and recording their path of travel and activities.

The consumer behaviour is subsequently analysed and used to determine optimal parameters for setting up and/or running a retail or other environment. A problem with this is that it can be time consuming and expensive to acquire sufficient data for viable statistically analysis. It is accordingly an object of a preferred form of the invention to go at least some way towards addressing this problem or to at least provide the public with a useful alternative.

SUMMARY OF THE INVENTION According to a first aspect of the invention there is provided a system for tracking people within a customer purchasing facility, comprising: a plurality of sensors arranged so that between them they cover overlapping areas within the facility and each sensor can sense the direction of movement of a person within the area that it covers; processing means arranged to receive and process information from the sensors indicative of the direction of movement; the processing means being adapted to determine when a matching direction of movement has been detected by way of and between more than one of the sensors, to attribute such movement to the same person and, based on such determination, to progressively track movement of the person through the facility in the same manner but using other combinations of the sensors; and the system being such that the processing means is adapted to determine shopping characteristics of the person based on the tracking.

Preferably the system is adapted to track a plurality of people in the manner set out above, to process data indicative of each of their shopping characteristics, and to determine and report shopping characteristics for such people as a group.

Preferably each sensor comprises a camera.

Optionally the processing means comprises a central processing module and a plurality of tracking modules.

Optionally each camera is associated with a tracking module adapted to receive information from the camera and process it to determine the trajectories of people moving within scope of the camera.

Preferably data representative of each trajectory is sent from each respective tracking module to the central processing module, and wherein the central processing module determines matching directions of travel indicative of the same person.

Preferably the processing means comprises a database of electronic information processed by the central processing module.

Preferably the system comprises a report generator adapted to obtain data from the database and generate a report based on the direction of movement of people within the facility.

Optionally the system provides an on-screen report which identifies one or more of:  favoured entry points to the facility;  favoured exit points to the facility;  the amount of time a person visited the facility;  the average amount of time customers visited the facility;  favoured display aisles;  areas of high foot traffic;  areas of low foot traffic;  the number of people that leave the facility without making a purchase;  interactions between service staff and customers;  how long customers waited for service from a staff member;  customers who left without being attended to by a staff member; and  the number of customers approached by a staff member.

Optionally the processing means identifies individual people by way of one or more of:  clothing colour;  clothing texture;  height;  gender; and  facial features.

Optionally the processing means identifies service staff by way of areas within the facility that they move to and distinguishes such staff from other people.

Optionally the processing means identifies service staff by way of markers worn by staff and distinguishes such staff from other people.

Optionally the customer purchasing facility comprises at least one of:  a retail store;  a mall;  a wholesale store;  a bank;  a casino; and  an entertainment facility.

DETAILED DESCRIPTION Some preferred embodiments of the invention will now be described by way of example and with reference to the accompanying drawing, of which: Figure 1 is schematic view of a system for tracking customers within a customer purchasing facility.

The system comprises a series of video cameras 1 arranged in a customer purchasing facility, for example in a mall, retail store, bank, casino, etc, to cover essentially the entire customer floor area between them. Only two cameras are shown in figure 1 although in practice here may be many more. The cameras 1 are arranged so that they overlap in terms of the part of the floor that they cover. Put another way, two or more cameras are set to capture the area of overlap 2 between adjacent parts of floor space.

Each camera 1 is arranged to continuously track the trajectory of each person moving within its field of view. Electronic information indicative of the trajectory is ultimately transmitted, wirelessly or by hardwire, to a central processing module 3. This module processes the information from each camera to determine whether more than one camera has detected movement in the same trajectory. Based on a trajectory match the central processing module 3 determines that the same person has been detected and tracks movement of that person. At the processing level the identity of the person is maintained as he or she moves from one camera field to the other. In this regard other pairs of the cameras operate in the same way to track the person moving from one area to another.

Data indicative of the person’s movement is recorded electronically. The process is repeated simultaneously and/or continuously for a large number of people within the area.. The central processing module 3 processes this data to determine trends in movement and therefore shopping characteristics across the group of people. For example the central processing module 3 identifies favored retail store entry and exit points, popular display aisles, areas of high customer foot traffic, areas of low customer foot traffic, the number of people that leave the store without making a purchase (eg without going through a sales checkout), customer penetration, etc.

The manner in which the trajectory information is obtained prior to transmission to the central processing module 3 will now be described. Each camera 1 is associated with a processing tracking module 4 which runs a trajectory software algorithm. In a preferred embodiment this algorithm is in the form of a Kalman Filter which continuously tracks all people in the camera’s field of view. Each tracking module determines each trajectory of movement and reports this as electronic information to the central processing module 3. The trajectory information includes data indicative of:  an identifier for each trajectory;  a time-stamp for each trajectory;  the location of a person corresponding to the trajectory in terms of a 3D coordinate frame; and  auxiliary information as to one or more of: o clothing colour; o clothing texture; o height; o gender; and o facial feature biometrics.

The central processing module 3 accepts the trajectory information from the tracking modules 4 and then maps it into a virtual coordinate frame representative of the customer purchasing facility, eg a retail store. The parameters used for such mapping are determined by an offline calibration step performed prior to deployment of the system.

As will be appreciated, a person moving through the fields of multiple cameras will have multiple trajectories, ie one for each camera that detects the person. Trajectories identified as belonging to the same person are combined at the processing level into a single trajectory. The central processing module 3 decides whether to combine trajectories based on the proximity of each trajectory and also on the features of vectors provided by the tracking modules 4. Preferably synchronisation of the cameras 1 is not required, for example information from the tracking modules 4 need not be provided to the central processing module 3 at the same moment in time. The central processing module 3 may also employ a software algorithm based on a Kalman Filter for making predictions and combining trajectory information from the tracking modules Trajectories that have been identified by the central processing module 3 as belonging to the same person are monitored (ie after trajectories have already been combined).

However if subsequent information indicates that the matching of trajectories was wrong, for example if a person is found to not have been tracked accurately, then the trajectories are separated..

The output of each tracking module 4 comprises a stream of trajectory information relevant to each individual in the combined field of view of the cameras 1. After combining of trajectory information , location and feature information is stored for each trajectory for each instant in time. Each trajectory that is exported is given a unique trajectory identifier.

The system may be arranged to track customers in a retail store, bank, casino or other customer service facility and to process the tracking information to determine:  the time each person entered the facility;  the time each person left the facility;  the duration of each person’s visit;  the average amount of time spent for each visit, optionally calculated for different time periods;  typical paths of travel used by people through the store (this can be used to determine overall floor space utilisation and customer penetration into the facility); and  interactions between trajectories and defined areas of interest within the facility to determine whether a person visits a particular area and if so then for how long they dwell within that area..

In preferred forms of the invention the system is adapted to differentiate between staff and customers. This may be achieved by monitoring trajectories for people who move behind a sales counter or in other staff only zones and assigning them a staff member identifier. The central processing module 3 preferably excludes staff from calculations which determine and lead to reports on characteristics of customers. In some embodiments of the invention staff may be identified by other parameters, for example the tracking modules 4 may identify a staff uniform by its colour or may identify a transmitting or non-transmitting tracking object worn by only staff. The system preferably records information relevant to staff movements and provides a report on staff performance based on such movements.

In some embodiments of the invention the system collects and processes data related to interactions between customers and staff, such as whether a customer was approached by staff to ask whether assistance was needed, how long the customer was attended to, the percentage of customers who were attended to, how much time passed between when a customer entered the facility and was engaged by staff, the percentage of customers approached within a specified time frame (eg after entering), the percentage of customers who left without being approached by staff, the number of customers each staff member approached, etc.

In some embodiments of the invention the system ‘profiles’ customers based on information extracted from the cameras or other sensing devices. For example the system may be adapted to detect the presence of a customer with a mobile phone and to determine, based on a signal from the phone, which phone service provider the customer is using. In some embodiments the other sensing devices may be adapted to receive and process other information, such as the identity of the customer holding the phone, whether the person is a VIP, the person’s gender and/or ethnicity, etc.

In further embodiments the system is adapted to identify groups of customers by the trajectory of each person in the group, for example whether they enter the facility together, whether they stay in close proximity while moving, etc. Data indicative of groups is recorded and used by the central processing module 3 to generate information related to the number and composition of groups and their shopping patterns.

In still further embodiments of the invention the system is adapted and has sensors to identify, record and report on more specific movement characteristics of customers.

These include hand movements, head movements, eye movements and/or facial expressions. The central processing module 3 associates all of these characteristics with the identifier for the customer concerned.

Preferably the system is adapted to provide a suite of easy to interpret on-screen reports based on the information received, processed and recorded. These enable users to readily analyse and profile customer activity and store area utilisation, including but not limited to store plan visualisation with a filterable heat map overlay.

They also allow trajectory data to be viewed and played back for different periods and for various classifications of both staff and customers, and to capture both customer visitation and dwell time. The heat map identifies hot spots in terms of customer activity and trends based on large volumes of data for large numbers of customers.

Because the system enables users to identify key or popular entry points, routes of travel and exit points for the facility, staff and management are able to organise product displays to maximise exposure of key products offerings. It also enables users to identify areas of low floor space utilisation and potential limitations in the design or layout of the facility.

In some embodiments the system is adapted to process trajectory information with data filters or other algorithms suitable to determine and report on:  customers who viewed a particular product;  customers who purchased a particular product;  customers who were approached by a specific sales staff member;  customers who were not approached by a or the staff member;  customers who were engaged for a predetermined period of time (eg 5 minutes);  number of items per sale;  average sales value;  sales conversion rate;  number of items per sale; and/or  customer staff ratio.

The results of all of the data processing described in this document may be reported on screen, as a printout, or otherwise. Preferably the central processing module 3 sends processed data to a database 5. The database may also receive and store data from other relevant sources as indicated at 6. The database is accessed by a software report generator 7 which creates reports that are displayed on a PC screen 8 on demand, for example via an internet or other link.

The central processing module 3 may comprise one or more computerised device, eg a PC or mainframe computer, etc. The tracking modules 4 may also each comprise one or more computerised device (not necessarily a computer per se). In some embodiments the central processing module 3, the tracking modules 4, the database 5, the report generator 7 and the display screen 8, may together form a ‘processing means’. However the term ‘processing means’ should not be seen as restricted to only this type of combination as in alternative embodiments it may comprise a completely different arrangement of parts. Further, in some embodiments each camera 1 may comprise at least part of a ‘sensor’ which can sense the direction of movement of a person. However the term ‘sensor’ should not be seen as restricted to only this type of device.

While some preferred forms of the invention have been described by way of example and with reference to the accompanying drawing, it should be appreciated that modifications and improvements can occur without departing from the scope of the following claims.

Claims (15)

WHAT WE CLAIM IS:

1. A system for tracking people within a customer purchasing facility, comprising: a plurality of sensors arranged so that between them they cover overlapping 5 areas within the facility and each sensor can sense the direction of movement of a person within the area that it covers; processing means arranged to receive and process information from the sensors indicative of the direction of movement; the processing means being adapted to determine when a matching direction of 10 movement has been detected by way of and between more than one of the sensors, to attribute such movement to the same person and, based on such determination, to progressively track movement of the person through the facility in the same manner but using other combinations of the sensors; and the system being such that the processing means is adapted to determine 15 shopping characteristics of the person based on the tracking.

2. A system according to claim 1, adapted to track a plurality of people in the manner set out above, to process data indicative of each of their shopping characteristics, and to determine and report shopping characteristics for such people as a group.

3. A system according to claim 1 or 2, wherein each sensor comprises a camera.

4. A system according to claim 1, 2 or 3, wherein the processing means comprises a central processing module and a plurality of tracking modules.

5. A system according to claim 4 when read on claim 3, wherein each camera is associated with a tracking module adapted to receive information from the camera and process it to determine the trajectories of people moving within scope of the camera.

6. A system according to claim 5, wherein data representative of each trajectory is sent from each respective tracking module to the central processing module, and wherein the central processing module determines matching directions of travel indicative of the same person.

7. A system according to claim 6, wherein the processing means comprises a database of electronic information processed by the central processing module. 5

8. A system according to any one of the preceding claims, comprising a report generator adapted to obtain data from the/a database and generate a report based on the direction of movement within the facility.

9. A system according to claim 8, which provides an on-screen report which identifies 10 one or more of: a) favoured entry points to the facility; b) favoured exit points to the facility; c) the amount of time a person visited the facility; d) the average amount of time customers visited the facility; 15 e) favoured display aisles; f) areas of high foot traffic; g) areas of low foot traffic; h) the number of people that leave the facility without making a purchase; i) interactions between service staff and customers; 20 j) how long customers waited for service from a staff member; k) customers who left without being attended to by a staff member; and l) the number of customers approached by a staff member.

10. A system according to any one of the preceding claims, wherein the processing 25 means identifies individual people by way of one or more of: a) clothing colour; b) clothing texture; c) height; d) gender; and 30 e) facial features.

11. A system according to anyone of the preceding claims, wherein the processing means identifies service staff by way of areas within the facility that they move to and distinguishes such staff from other people.

12. A system according to anyone of the preceding claims, wherein the processing means identifies service staff by way of markers worn by staff and distinguishes such staff from other people.

13. A system according to any one of the preceding claims, wherein the customer purchasing facility comprises at least one of: a) a retail store; b) a mall; 10 c) a wholesale store; d) a bank; e) a casino; and f) an entertainment facility. 15

14. A system according to claim 1, substantially as herein described.

15. A system substantially as herein described with reference to the accompanying drawing. McCABE & CO Attorney for the Applicant