Electronic article surveillance system and method
Patent Information
- Application Number
- CN202311846645.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-04-17
- Filing Date
- 2020-04-08
- Publication Date
- 2026-10-27
- Estimated Expiration
- 2040-04-08
Smart Images

Figure CN117789384B_ABST
Abstract
Description
[0001] This application is a divisional application of the Chinese national phase application of PCT application filed on April 8, 2020, with international application number PCT / EP2020 / 059983 and entitled "Electronic Anti-theft System and Method". The entire contents of the Chinese national phase application, which entered the Chinese national phase on October 15, 2021, with application number 202080029353.8, are incorporated herein by reference. Technical Field
[0002] Theft, also known as shoplifting, is a problem for many retailers, especially those that sell consumer goods such as clothing, garments that are relatively easy to hide under coats, and handbags, particularly when fitting rooms are available.
[0003] Electronic Article Surveillance (EAS) is known in the art to trigger alarms and potentially prevent goods from being removed from a store or shopping area in an unauthorized manner.
[0004] In a typical EAS system, sales staff affix electromagnetic tags to merchandise, such as more expensive items. Antennas are placed near the entrance / exit of the store or shopping area and coupled to circuitry that detects the passing of the tags affixed to the merchandise. Typically, the tags are removed when the merchandise is paid for at the checkout. Therefore, when a tag is detected passing between the antennas, it is usually a theft-related incident.
[0005] Despite the widespread installation of such systems, theft remains a major problem for retailers in almost every store, such as those that sell clothing or even those that sell food.
[0006] It was discovered that individuals intending to commit theft entered stores or shopping areas carrying magnets configured to unlock locks that kept the aforementioned tags affixed to merchandise. Inside the store, they then removed the tags from the merchandise, leaving the labels behind. They then carried the merchandise out of the store without triggering any alarms through the regular EAS alarm system.
[0007] Such a magnet, configured to unlock locks on items bearing the aforementioned tags, is called a separator, separating magnet, or unlocking magnet. However, this separating magnet is difficult to detect because it is easily confused with other magnetic objects present and can even move within and around the shopping area. Magnets can be used for locks on bags, and metal parts, such as those in shoes or bags, can behave as magnets.
[0008] One problem is that automated detection is prone to generating false alarms, or failing to detect a magnet when it should. In this regard, it should be noted that salespeople and customers who might be wrongly accused of theft strongly dislike false alarms. Background Technology
[0009] EP 2997557 B1 relates to an electronic anti-theft system that automatically detects when a separating magnet enters a store or shopping area and describes an alarm system that sounds when a strong magnet used in a separator enters the shopping area. The electronic anti-theft system includes: a first multi-axis magnetometer and a second multi-axis magnetometer, arranged in a first station and a second station and configured to output a first vector signal and a second vector signal, respectively representing the movement of a first magnetic field vector and a second magnetic field vector; and a signal processor, coupled to receive the first vector signal and the second vector signal and configured to: estimate a first rotation of the first magnetic field vector and a second rotation of the second magnetic field vector; generate an indication signal including an indication of reverse rotation or same-direction rotation; and determine whether to issue or suppress an alarm signal regarding a possible theft-related event, at least in response to the indication signal. The system issues an alarm if a decryption magnet used for anti-theft tags passes between stations, for example, when the stations are located on each side of the shopping area entrance.
[0010] However, there is a desire to further improve the reliability related to detecting theft-related incidents. Summary of the Invention
[0011] It has been observed that reliably detecting theft-related incidents in some places, and sometimes with existing systems (e.g., systems for detecting separated magnets), is problematic. Therefore, existing systems either generate false alarms when they should or fail to issue an alarm. It has also been observed that conventional time-domain filtering may be sufficient in some cases, but not in all. Therefore, the inventors designed:
[0012] An electronic anti-theft system includes:
[0013] A first multi-axis magnetometer (101) is arranged in a first station at a first position and configured to output a first vector signal (vs0) representing the motion of a first magnetic field vector;
[0014] A second multi-axis magnetometer (102, 104) is arranged in a second station at a second position and configured to output a second vector signal (vs1, vs3) representing the motion of the second magnetic field vector; and
[0015] The signal processor (501), coupled to receive a first vector signal (vs0) and a second vector signal (vs1), is configured to:
[0016] The first value of the parameter of the first multidimensional transformation (C1) is determined based on the optimization of the difference between the second vector signal (vs1) and the first compensation signal (cs1); wherein the first compensation signal (cs1) is generated from the transformation of the first vector signal (vs0) according to the first multidimensional transformation (C1);
[0017] A compensated second vector signal (cvs1) is generated from the second vector signal (vs1) and the first compensated signal (cs1);
[0018] The detector signal (D) responding to the compensated second vector signal is determined to meet a predetermined criterion; and
[0019] At least in response to determining that the detector signal meets predetermined criteria, issue or abstain from issuing an initial alert regarding a possible theft-related incident.
[0020] Therefore, theft-related incidents can be detected more reliably, and at least the risk of generating false alarms or failing to issue alarms when they should be issued is reduced.
[0021] Specifically, but not limited to, theft-related incidents can be reliably detected despite the presence of interfering magnetic fields emitted by high-power electromagnetic devices. These high-power electromagnetic devices can be associated with overhead contact lines, such as those used in railways, subway lines, and trams.
[0022] Interference from, for example, high-power electromagnetic devices can be suppressed in a compensated second vector signal using a first compensation signal, obtained via a first transformation and information from the first vector signal. The first transformation can be adapted to, for example, one or both of rotation and scaling transformations. The first transformation can represent the difference between a first magnetic field vector and a second magnetic field vector.
[0023] The first multi-axis magnetometer can be placed at a distance from the second multi-axis magnetometer, within a range of several meters (e.g., 1-20 meters). Depending on the system to be protected, spatial information related to the interfering magnetic field is included in the first transformation because the first magnetometer senses the magnetic field at the location of the first station, which differs from the location of the second station. The first vector signal is acquired at a location different from the second vector signal. The second multi-axis magnetometer can be placed near the area where theft-related events are expected to be detected (e.g., entrance area, fitting room area, or passageway). The first multi-axis magnetometer can be placed closer to a magnetic field source (e.g., overhead contact line of a train, subway, or bus line) and / or in a location where customers are not desired to pass through (at least not frequently).
[0024] Therefore, compared to conventional time-domain filtering, signal processors enable more effective filtering of interference from electromagnetic devices. It has been found that, despite the presence of electromagnetic devices emitting strong electromagnetic fields, the systems requiring protection can adequately suppress the effects of interfering magnetic fields to more reliably detect theft-related incidents.
[0025] The second multi-axis magnetometer can be used alone or in combination with one or more additional multi-axis magnetometers (e.g., in combination with the first multi-axis magnetometer) to sense theft-related events.
[0026] The second vector signal used for compensation is obtained using information from the first multi-axis magnetometer. Specifically, a first multidimensional transformation is applied so that the first vector signal can be used to compensate the second vector signal.
[0027] In some embodiments, a first value of the parameter of the first multidimensional transformation (C1) is determined on a cyclic basis according to a first timing (T1, T2, T3).
[0028] Therefore, despite the presence of time-varying interference magnetic fields emitted by high-power electromagnetic devices, theft-related incidents can be reliably detected. For example, a shift in the level of essentially DC current has been observed in the time-varying interference magnetic fields emitted by overhead contact lines associated with railways, subway lines, trams, etc. These time-varying interference magnetic fields alternate regularly or irregularly and may occur at frequencies associated with theft-related incidents occurring near a second station.
[0029] In some aspects, in response to a first value of a parameter re-determined on a cyclic basis according to a first timing, a compensated second vector signal is generated from the second vector signal based on the first value re-determined on a cyclic basis. Therefore, the most recent first value is used for compensation.
[0030] The first value of the parameter of the first multidimensional transformation can be determined at regular intervals (e.g., every 30 seconds, every 60 seconds, every 3 minutes) or at other regular or irregular intervals.
[0031] In some embodiments, the difference between the second vector signal (vs1) and the first compensation signal (cs1) is determined at a parallel time period or a portion of the parallel time period between the first vector signal (vs0) and the second vector signal (vs1); and a compensated second vector signal (cvs1) is generated from the second vector signal (vs2) at a time after the parallel time period and based on a first value re-determined on a cyclic basis.
[0032] Therefore, the compensation is based on the most recent first value of the parameters of the first transformation, and the compensated second vector signal adapts to the changing disturbance magnetic field more quickly.
[0033] Parallel time periods (i.e., the current time period) may overlap with or not overlap with previous parallel time periods, or may be continuous or discontinuous.
[0034] In some embodiments, the electronic anti-theft system includes:
[0035] The third multi-axis magnetometer (103, 105) is arranged in the third station and is configured to output a third vector (vs2) signal representing the motion of the third magnetic field vector;
[0036] The signal processor (501) is further configured as follows:
[0037] The second value of the parameters of the second multidimensional transformation (T2) is determined based on the optimization of the difference between the third vector signal (A3) and the second compensation signal (cs2); wherein the second compensation signal (cs2) is generated from the transformation of the first vector signal (vs0) according to the second multidimensional transformation (C2);
[0038] A compensated third vector signal (cvs2) is generated from the third vector signal (vs2) and the second compensated signal (cs2);
[0039] The detector signal (D) responds to the compensated third vector signal (cvs2).
[0040] The second and third multi-axis magnetometers can be located on each side of the passageway, for example, the passageway leading to the shopping area or the passageway leading to the fitting room.
[0041] Therefore, the second and third stations can be located on opposite sides of the passage. The first station can be located at a first distance from either the second or the third station, wherein the first distance is greater than (e.g., at least twice) the distance between the second and the third stations.
[0042] Although the distance between the second station and the first station is relatively small, the corresponding transformation values of the second and third magnetometers can be different. For example, the first magnetometer may have an orientation different from one or both of the second and third magnetometers. It is also possible for the orientation of one or more magnetometers to be changed intentionally or unintentionally.
[0043] It can process one or both of the compensated second vector signal and the compensated third vector signal to issue or abort an alarm warning about a possible theft-related incident. This is described in more detail in EP 2997557 B2 related to passageways and application PCT / EP2018 / 077148 related to fitting rooms.
[0044] In some embodiments, the electronic anti-theft system includes:
[0045] The fourth multi-axis magnetometer (104, 106) is arranged in the fourth station and is configured to output a fourth vector (vs3, vs5) signal representing the motion of the fourth magnetic field vector;
[0046] The signal processor is also configured as follows:
[0047] The third value of the parameters of the third multidimensional transformation (C3) is determined based on the optimization of the difference between the fourth vector signal (vs3) and the third compensation signal; wherein, the second compensation signal is generated from the transformation of the first vector signal (vs1) based on the third multidimensional transformation (C3);
[0048] A compensated third vector signal is generated from the third vector signal and the second compensated signal;
[0049] The detector signal (D) responds to the compensated third vector signal.
[0050] It can process one or more of the compensated second vector signal, compensated third vector signal, and compensated fourth vector signal to issue or waive an alarm warning about a possible theft-related incident. This is also described in more detail in EP 2997557 B2 related to passageways and application PCT / EP2018 / 077148 related to fitting rooms.
[0051] In some embodiments, the signal processor is further configured to perform bandpass filtering on one or more of the first vector signal, the second vector signal, the third vector signal, and the fourth vector signal through a corresponding bandpass filter; wherein the corresponding bandpass filter has a lower cutoff frequency below 1.0 Hz and an upper cutoff frequency above 4 Hz and below 50 Hz.
[0052] Bandpass filters can effectively eliminate offsets that occur at frequencies above 4 Hz to above 50 Hz, corresponding to the Earth's magnetic field and AC noise such as that from electrical appliances, motors, etc.
[0053] In some respects, bandpass filtering is applied to provide a vector signal as a bandpass-filtered vector signal. Therefore, the aforementioned vector signal can be a bandpass-filtered vector signal. This improves the effectiveness of the compensation because the transformation can be estimated more accurately when offsets corresponding to the Earth's magnetic field and, for example, AC noise from electrical appliances are removed beforehand.
[0054] The bandpass filter can be implemented by a low-pass filter and a high-pass filter, or by a first low-pass filter and a second low-pass filter coupled via a summing unit to output a difference signal known in the art.
[0055] In some embodiments, one or more of the first multidimensional transformation (C1), the second multidimensional transformation (C2), and the third multidimensional transformation (C3) are estimated based on regularization applied during the iterative estimation of the transformation parameters, and the regularization penalizes the relatively large parameters of the transformation parameters compared to the relatively small parameters of the transformation parameters.
[0056] Regularization prevents or suppresses overfitting. This is advantageous because, typically, the magnetic field vector in three-dimensional space is much stronger in one direction than in others. This helps suppress overfitting in other directions. Regularization can be L1 regularization, L2 regularization, or another type of regularization. Regularization constrains the coefficient estimation (regularization) to zero. In other words, this technique discourages learning more complex or flexible models to avoid the risk of overfitting.
[0057] In some embodiments, a first vector signal (vs0) is acquired during a first time period (TS1) and a second time period (TS2), and a second vector signal (vs1) is acquired during the first time period (TS1) and the second time period (TS2); a first parameter (C1) is estimated from the first vector signal (vs0) and the second vector signal (vs1) at a first time (T1) during the first time period (TS1); a first parameter (C1') is estimated from the first vector signal and the second vector signal at a second time (T2) during the second time period; a compensated second vector signal (cvs1) is generated after the second time (T2) based on the first parameter (C1) estimated at the first time (T1) according to a first criterion; and a compensated second vector signal is generated after the second time (T2) based on the first parameter (C1') estimated at the second time (T2) according to a second criterion.
[0058] In this way, the system can be adapted to improved parameters that can be estimated on a continuous basis. The first and second time periods can be consecutive time periods, for example, with a duration of 30-120 seconds or shorter or longer. The first and second time periods can overlap or be spaced out in time to occur at regular or irregular times.
[0059] In some embodiments, the first criterion is satisfied when the compensated second vector signal (cvs1) generated from the second vector signal (vs1) in the first time period (TS2) based on the first parameter (C1') estimated in the second time period (T2) has a lower intensity than the compensated second vector signal (csv1) generated from the second vector signal (vc1) in the first time period based on the first parameter (C1) estimated in the first time period (T1).
[0060] Therefore, this intensity provides a metric and mutual threshold for estimating whether parameters are updated or maintained over time. Consequently, the system can adapt to time-varying magnetic fields and / or repositioning and / or rotation of the station and / or magnetometer relative to each other. This significantly reduces the system's service uptime and further reduces the frequency of false alarms or failures to issue alarms when they should.
[0061] In some embodiments,
[0062] At the first time: a first parameter (C1) is estimated based on a first vector signal (vs0) of a first time period (TS1), a second vector signal (vs1) of the first time period (TS1), and a first compensation signal of the first time period (TS1); wherein, the first compensation signal of the first time period (TS1) is generated from the first parameter (C1) estimated at the first time (T1) and the first vector signal (vs0) of the first time period (TS1);
[0063] At a second time: a first parameter (C1') is estimated based on a first vector signal (vs0) of a second time period (TS2), a second vector signal (vs1) of a second time period (T2), and a first compensation signal of a second time period; and a first compensation signal is generated from the first parameter (C1') estimated at the second time and the first vector signal of the second time period; a second vector signal (cvs1) of the first compensation is generated from the second vector signal (vs1) of the second time period (TS2), and a first compensation signal is generated from the first parameter (C1) estimated at the first time and the first vector signal of the second time period; a second vector signal (cvs1') of the second compensation is generated from the second vector signal (vs1) of the second time period, and a first compensation signal is generated from the first parameter (C1') estimated at the second time and the first vector signal of the second time period; the signal processor is further configured to:
[0064] Estimate the second vector signal (cvs1) with the first compensation and the second vector signal with the second compensation, and
[0065] The second vector signal (cvs1) with first compensation is determined to be superior to the second vector signal (cvs1') with second compensation, and the second vector signal (cvs1) with compensation is generated based on the first parameters estimated at the first time, while the second vector signal with compensation generated based on the first parameters estimated at the second time is abandoned.
[0066] In some embodiments, the signal processor (501) is further configured to perform the detection of corresponding motions of the first magnetic field vector and the second magnetic field vector through the following steps:
[0067] - Estimate the first rotation of the first magnetic field vector and the second rotation of the second magnetic field vector;
[0068] - Generate an indication signal that includes a reverse rotation or a rotation in the same direction;
[0069] - Determine whether the first alarm should be activated at least in response to the indication signal.
[0070] This is also described in more detail in EP 2997557 B2, which relates to the channel.
[0071] In some embodiments, the signal processor is further configured to:
[0072] - Detect the corresponding motion of the first magnetic field vector and the second magnetic field vector;
[0073] - After detecting the corresponding motion of the magnetic field vector, detect the start and duration of the fluctuation of at least the first magnetic field vector or the second magnetic field vector; wherein the duration of the fluctuation is determined according to a first timing criterion;
[0074] -At least in response to determining the start and duration of fluctuations in at least the first or second magnetic field vector, determine whether to issue or abandon the first alert regarding a possible theft-related incident.
[0075] In some embodiments, detecting the corresponding motion of the first magnetic field vector and the second magnetic field vector includes:
[0076] - Determine whether the motion of the first magnetic field vector and the second magnetic field vector corresponds to the basic horizontal motion of the magnet between the first and second stations.
[0077] In some embodiments, detecting the persistence of fluctuations includes:
[0078] - Determine whether the motion of one or both of the first and second magnetic field vectors corresponds to the oscillating motion of the magnet near one or both of the first and second stations.
[0079] This is also described in more detail in application PCT / EP2018 / 077148 related to fitting rooms.
[0080] A method for detecting theft-related incidents is also provided, comprising: a first multi-axis magnetometer disposed at a first station and configured to output a first vector signal representing the motion of a first magnetic field vector; a second multi-axis magnetometer disposed at a second station and configured to output a second vector signal representing the motion of a second magnetic field vector; and a signal processor coupled to receive the first and second vector signals, including:
[0081] Estimate the first multidimensional transformation (C n The first multidimensional transformation represents the difference between the first magnetic field vector and the second magnetic field vector and is estimated over a period of time based on the optimization of the difference between the first vector signal and the second vector signal;
[0082] In response to the first multidimensional transformation (C) n The first compensation signal is generated by transforming the second vector signal as defined in the second vector signal, thereby compensating the second vector signal;
[0083] The detector signal (D) responding to the compensated second vector signal is determined to meet a predetermined criterion; and
[0084] At least in response to determining that the detector signal meets predetermined criteria, issue or abstain from issuing an initial alert regarding a possible theft-related incident. Attached Figure Description
[0085] The following description is provided in more detail with reference to the accompanying drawings, in which:
[0086] Figure 1The magnetometer is shown as an example of an anti-theft system installed in the entrance area and fitting room area of a shopping area;
[0087] Figure 2 A magnetometer is shown, for example, in an anti-theft system installed in a fitting room area, comprising a first magnetometer and a second magnetometer;
[0088] Figure 3 A magnetometer is shown, for example, in an anti-theft system installed in an entrance area, including a first magnetometer, a second magnetometer, and a third magnetometer;
[0089] Figure 4 A magnetometer is shown, for example, in an anti-theft system installed in a fitting room area, comprising a first magnetometer and a second magnetometer;
[0090] Figure 5 A first block diagram of the signal processor of the anti-theft system is shown;
[0091] Figure 6 A second block diagram of the signal processor of the anti-theft system is shown; and
[0092] Figure 7 A timing diagram is shown for estimating and using the estimated parameters of the transformation. Detailed Implementation
[0093] The following describes an electronic anti-theft system with respect to different embodiments, including at least a first multi-axis magnetometer 101 and second multi-axis magnetometers 102 and 104. Typically, the first multi-axis magnetometer 101 is arranged in a first station at a first location and configured to output a first vector signal vs0 representing the motion of a first magnetic field vector. For example, the second multi-axis magnetometers, represented by 102 and 104, are arranged in a second station at a second location and configured to output second vector signals vs1 and vs3 representing the motion of a second magnetic field vector. A magnetic field vector refers to a representation of the physical magnetic field sensed by the respective magnetometer. Typically, the magnetometers herein are shown in a Cartesian coordinate system with x, y, and z axes. Magnetometers may be tilted relative to each other, although this is not shown herein. Magnetometer components may include markings or symbols printed on their surfaces to indicate the orientation of their axes.
[0094] A signal processor (described further below) is coupled to receive a first vector signal vs0 and a second vector signal vs1. The signal processor is, for example, configured with one or more processors running a program to generate at least one compensated vector signal and determine that a detector signal responding to the compensated vector signal meets a predetermined criterion; and at least in response to determining that the detector signal meets the predetermined criterion, to issue or abstain from issuing a first alarm regarding a possible theft-related event. In some embodiments, the signal processor is coupled to receive one or more additional vector signals, such as vs2 and vs3. In some embodiments, multiple signal processors are used, each coupled to receive two or more vector signals. The vector signals can be transmitted from a corresponding station to the signal processor via a wireless or wired connection. Furthermore, one or more signal processors can be coupled to an alarm transmitter via a wireless or wired connection, such as to a mobile alarm transmitter.
[0095] In this paper, a vector signal is a digital vector signal, comprising three sample values (e.g., specified as x). i y i and z i The magnetometer outputs a set of values, one sample value for each of three mutually orthogonal dimensions (e.g., x, y, and z). The magnetometer can be a digital type outputting digital values or an analog type outputting an analog signal, which is then converted into a digital vector signal by an analog-to-digital converter. The digital signal can communicate according to the I2C standard, the Bluetooth standard, or another protocol.
[0096] Before detailing the steering signal processor, the system configuration of the multi-axis magnetometer is described, which is arranged in the corresponding station at the corresponding location and configured to output the corresponding vector signal.
[0097] Figure 1 An example of a magnetometer system is shown, for example, installed in the entrance area and fitting room area of a shopping area as part of an anti-theft system. The magnetometer system 100 includes:
[0098] i) The first multi-axis magnetometer 101 outputs the first vector signal vs0;
[0099] ii) The first group 115 of magnetometers 102 and 103 outputs corresponding vector signals vs1 and vs2; and
[0100] iii) The second group of magnetometers 104, 105, 106 and 107, 108.
[0101] The first set 115 of magnetometers 102 and 103 can be arranged on opposite sides of the passage indicated by arrow 117, which can be the entrance to the shopping area, allowing people entering the shopping area to pass between magnetometers 102 and 103. The passage between the magnetometers can also be referred to as a "door". People who do not need to enter the shopping area via a door can pass along arrow 116. As described in more detail in EP 2997557 B2, it can be determined whether a person carrying a separate magnet (or a magnetically similar object) passes through the door along arrow 117 or whether a person carrying a separate magnet passes through the door along arrow 116. One or more "doors" can be installed in this manner. One or more magnetometers can be used for two adjacent doors to reduce the number of magnetometers required.
[0102] A second set 108 of magnetometers 104, 105, 106, and 107 can be arranged in, for example, the fitting room area of a shopping area. In some embodiments, in the event of a theft-related incident, each fitting room requires at least one magnetometer to distinguish which fitting room triggers an alarm. The fitting rooms are designated 109, 110, and 111 and can be accessed via corresponding passages indicated by arrows 112, 113, and 114. Here, magnetometers 104, 105, 106, and 107 are arranged to form a door at the entrance of each fitting room. As described in more detail in EP 2997557 B2, it is thus possible to determine if a person is carrying a separating magnet into the fitting room. As described in more detail in patent application PCT / EP2018 / 077148, it is possible to determine whether a predetermined and likely theft-related movement of a separating magnet has occurred in a given fitting room.
[0103] Importantly, the magnetometer system includes a first multi-axis magnetometer 101. The first multi-axis magnetometer 101 is positioned at a distance from a second multi-axis magnetometer, for example, within a few meters (e.g., 1-20 meters). Here, the second multi-axis magnetometer can be any of magnetometers 102, 103, 104, 105, 106, and 107. The magnetometers of the first group 108 can be positioned, for example, at a distance of approximately 0.5 meters to 2 meters or more, depending on the size of the fitting room. The magnetometers of the second group 115 can be positioned, for example, at a distance of approximately 1 meter to 4 meters or more, depending on the location of other alarm stations relative to the entrance. The first multi-axis magnetometer 101 can be placed closer to a magnetic field source (such as an overhead contact line of a train, subway, or bus line) and / or in a location where customers are not expected to pass through (at least not frequently).
[0104] A magnetometer system may include fewer or more groups of magnetometers to obtain one or more desired detection areas or gates.
[0105] Figure 2An example of a magnetometer system (including a first magnetometer and a second magnetometer) is shown, for example, installed in an anti-theft system in a fitting room area. The magnetometer system 200 includes a first magnetometer 101 and a second magnetometer 104. System 200 can be used, for example, in conjunction with fitting room 109. Therefore, a simpler system is provided. This system can be implemented as described in patent application PCT / EP2018 / 077148.
[0106] As will be described in more detail below, for the magnetometer system 200, the signal processor is configured to calculate the values of the parameters of the first multidimensional transformation C1, and generate a compensated second vector signal from the second vector signal vs1, the first vector signal vs0, and the first transformation C1. Transformation C1 is shown by a dashed line denoted as C1.
[0107] Figure 3 An example of a magnetometer system (including a first magnetometer, a second magnetometer, and a third magnetometer) is shown, for example, installed in an anti-theft system in an entrance area. The magnetometer system 300 includes a first magnetometer 101, a second magnetometer 102, and a third magnetometer 103. System 300 can be used, for example, in conjunction with one or both of the entrance area and the fitting room.
[0108] For the magnetometer system 300, the signal processor is configured to calculate the values of the parameters of the first multidimensional transform C1 and the parameters of the second multidimensional transform C2. Furthermore, the signal processor is configured to generate:
[0109] i) The compensated second vector signal derived from the second vector signal vs1, the first vector signal vs0, and the first transform C1; and
[0110] ii) The third vector signal compensated from the third vector signal vs2, the first vector signal vs0, and the second transformation C2.
[0111] Transformations C1 and C2 are shown by dashed lines representing C1 and C2, respectively.
[0112] Figure 4 An example of a system is shown, for example, a magnetometer (including a first magnetometer and a second magnetometer) installed in an anti-theft system in a fitting room area. The magnetometer system 400 includes a first magnetometer 102 and a second magnetometer 103. System 400 can be used, for example, in conjunction with one or both of an entrance area and a fitting room.
[0113] For the magnetometer system 400, the signal processor is configured to calculate the values of the parameters of the first multidimensional transformation C1 and generate a compensated second vector signal from the second vector signal vs2, the first vector signal vs0, and the first transformation C1. Transformation C1 is shown by a dashed line denoted as C1. Therefore, the first magnetometer 102 itself can be used as a magnetometer for a "door" or a fitting room.
[0114] Figure 5 A first block diagram of an example signal processor for an anti-theft system is shown. The first block diagram can be implemented as part of the hardware and / or software of the signal processor. The signal processor 501 is coupled to receive a first vector signal vs0, a second vector signal vs1, and a third vector signal vs3.
[0115] Bandpass filter 502 filters one or more of the first vector signal, the second vector signal, and the third vector signal using corresponding bandpass filters. The corresponding bandpass filters have a lower cutoff frequency below about 1.0 Hz and an upper cutoff frequency above about 4 Hz and below about 50 Hz, for example, at -3 dB. Bandpass filters can effectively eliminate offsets corresponding to the Earth's magnetic field and, for example, AC noise from appliances, motors, etc., with switching, rotating, or reciprocating electromagnetic circuits. For simplicity, the vector signals input to and output from the bandpass filter are represented by the same reference numerals. In some embodiments, the bandpass filter may be omitted.
[0116] The time intervals of the first vector signal vs0, the second vector signal vs1, and the third vector signal vs2 are stored in buffers denoted as [vs0], [vs1], and [vs2]. The buffers can be rewritten with the most recent time interval at regular time intervals (e.g., every 30 seconds).
[0117] The signal processor has a first branch configured to calculate a first transform C1 and a compensated second vector signal cvs1. Furthermore, the signal processor has a second branch configured to calculate a second transform C2 and a compensated third vector signal cvs2.
[0118] The first branch is based on an estimator 'Est'503, which is configured to determine a first value of the parameters of the first multidimensional transform C1 based on the optimization of the difference between the second vector signal vs1 and the first compensation signal cs1; wherein the first compensation signal is generated from the transform of the first vector signal vs0 according to the first multidimensional transform C1. More specifically, the estimator is configured to optimize the following expression according to an optimization algorithm (e.g., L-BFGS (Low memory Broyden-Fletcher-Goldfarb-Shanno) algorithm).
[0119]
[0120] Among them, vs n It is, for example, a vector signal from the second vector signal vs1 (an N×3 matrix; where N is the number of samples in the buffer); vs n =0 (N×3 matrix) is the first vector signal vs0; k is a constant; C n It is a 3×3 transformation matrix; i and j are summation variables; and |..| denotes the 1-norm.
[0121] This transformation converts the 3D representation of a first vector to the 3D representation of a second vector. The transformation may have parameters representing one or both of rotation and scaling. The transformation parameters may be stored in one or more variables, such as in an array known in the art. The signal processor may discard storage operations associated with all elements of the 3×3 matrix, for example, if the transformation includes five non-zero parameters.
[0122] Here, optimization can be the minimization of e. The above expression is iteratively optimized to minimize e as the values stored in the buffer are summed. n (The summation of values in the buffer is not shown in the expression above). The last term in the expression above is the so-called L1 regularization. Regularization penalizes a relatively large transform parameter value C compared to a relatively small transform parameter value. n Regularization prevents or suppresses overfitting. Other types of regularization can be applied instead of L1 regularization.
[0123] This can be done after a predetermined number of iterations, after a predetermined time period, or when a threshold e is reached. n When this is done, stop criteria known in the art are applied to obtain the value of the transformation.
[0124] The transformation (e.g., C1) can be computed in other ways (e.g., using other optimization algorithms selected from the steepest descent algorithm category).
[0125] In response to the value of the transformation C1 available after the above iterative calculation, the signal processor can generate a compensated second vector signal cvs1 from the second vector signal vs1 and the first compensated signal cs1, and generate the first compensated signal from the transformation of the first vector signal vs0 according to the first multidimensional transformation C1 available after the iterative calculation. The compensated second vector signal cvs1 can be generated by calculating the difference between the second vector signal vs1 and the first compensated signal cs1 through the summing unit 505. This difference can be calculated as a conventional difference or in another way.
[0126] The second branch is based on the estimator 'Est'504 and operates as described above.
[0127] Compensated second vector signal CVS1 and compensated third vector signal CVS2, generated from the first and second branches respectively, are input to vector processor 'VP' 707. Vector processor 507 receives the vector signals, processes them, and generates detector signals (D). Vector processor 507 may operate as described in more detail in EP 2997557 B2 or PCT / EP2018 / 077148. Therefore, the compensated vector signals described herein are input to the vector processor, instead of receiving uncompensated vector signals as described in EP 2997557 B2 and PCT / EP2018 / 077148.
[0128] Detector signal D is input to an alarm unit, which determines that the detector signal meets a predetermined criterion. The predetermined criterion may be that an alarm is activated by an enable signal and the detector signal undergoes a predetermined transformation, or that a predetermined threshold is reached. At least in response to determining that the detector signal meets the predetermined criterion, alarm unit 508 issues or cancels an alarm regarding a possible theft-related event.
[0129] The alarm can be transmitted via a wireless transmission device (such as radio circuit 508) to a mobile device, for example, carried by a store clerk.
[0130] In another example, signal processor 501 is coupled to receive a first vector signal vs0, a second vector signal vs3, and a third vector signal vs4. In other examples, signal processor is coupled to receive a first vector signal vs0, a second vector signal vs3, a third vector signal vs4, a fourth vector signal vs5, and a fifth vector signal vs6. The signal processor is configured to process vector signals with necessary modifications. In addition to the first vector signal, one or more vector signals may be processed to generate compensated vector signals.
[0131] Figure 6 A second block diagram of a signal processor for an anti-theft system is shown. This second block diagram can be implemented as part of the signal processor's hardware and / or software. Signal processor 601 can be part of or interconnected with signal processor 501. Signal processor 601 is configured to estimate the value of a first transformation over time.
[0132] The time intervals of the first vector signal vs0 and the second vector signal vs1 are stored in buffers denoted as [vs0] and [vs1]. The buffers can be rewritten with the most recent time interval at regular time intervals (e.g., every 30 seconds or every 180 seconds) or at other time intervals.
[0133] As described above, the values of the parameters of the first transformation C1 can be calculated using an iterative algorithm. First, C1 has already been calculated during the time period TS1 and is available at the first time point T1 (see...). Figure 7 As described above, a compensated second vector signal cvs1 can be generated via summing units 603 and 604. The compensated second vector signal cvs1 is input to the estimator “Eval” 602. Next, at a later time point T2, the values of the parameters of the first transformation are recalculated, as represented by C1’, which has already been calculated during the time period TS2 and is available at the second time point T2 (see [link to original text]). Figure 7 A compensated second vector signal cvs1' based on the transform C1' can be generated. The compensated second vector signal cvs1' is also input to the estimator "Eval" 602. Therefore, the estimator 602 receives the compensated second vector signal cvs1 and the compensated second vector signal cvs1'.
[0134] Estimator 602 can estimate two versions of the compensated second vector signal to determine which first transform C1 or C1' to use to calculate the compensated second vector signal for at least some future time periods.
[0135] Estimator 602 can estimate two versions of the compensated second vector signal, for example, according to the following expressions calculated for cvs1 and cvs1':
[0136]
[0137] Where Sig represents the signal strength metric; |..| represents the 1-norm; M (e.g., M = 600) represents the number of samples in a time period, x i y i and z i These represent three sample values at time or sample instance i, one sample value for each of the three mutually orthogonal dimensions (e.g., x, y, and z).
[0138] Estimator 602 can determine that C1' results in a lower signal strength by comparing the corresponding value of |Sig|, and thus determine to replace C1 with C1' at least for some future time periods. Alternatively, estimator 602 can determine that C1 results in a lower signal strength by comparing the corresponding value of |Sig|, and thus determine to keep C1 for at least some future time periods instead of replacing C1 with C1' to calculate the compensated vector signal.
[0139] The above estimation can be performed on a cyclical basis according to a predetermined timing (e.g., at times T1, T2, T3, etc.). To save memory, C1 can contain the transformation of the vector signal currently used to calculate the compensation, while C1' can represent the most recent candidate transformation. Therefore, although C1 is calculated from the time period preceding the time periods stored in buffers [vs0] and [vs1], C1 is compared with C1' calculated from the time periods stored in buffers [vs0] and [vs1]. Buffers [vs0] and [vs1] contain parallel time periods of the first vector signal vs0 and the second vector signal vs1.
[0140] Figure 7 A timing diagram is shown for estimating and using the estimated parameters of the transformation. The timing diagram is shown as a function of time t. The first vector signal vs0 and the second vector signal vs1 are shown with time, and especially with the time intervals TS1, TS2 and TS3 that have elapsed at times T1, T2 and T3, respectively.
[0141] The calculation of the first value of the first parameter of the first transformation, 'Comp.', occurs from T1 to T1a, as indicated by the pointer box denoted as C1, and T1a is available at the first time after T1. Subsequently, the recalculation of the first value of the first parameter of the first transformation occurs from T2 to T2a, as indicated by the pointer box denoted as C1', and T2a is available at the first time after T2.
[0142] As described above, the signal processor can determine that C1 is superior to C1' and continue to use C1. This is shown with respect to label "C_A". Alternatively, as described above, the signal processor can determine that C1' is superior to C1 and use C1' instead of C1, as shown with respect to label "C_B".
[0143] The signal processor can be configured to process a pair of vector signals, such as vs0 and vs1, using the disclosures provided above, or to process multiple vector signals simultaneously. For example, signal processor 501 can omit the second branch to include the first branch.
[0144] An electronic anti-theft system is also provided, including:
[0145] A first multi-axis magnetometer (101) is arranged in a first station at a first position and configured to output a first vector signal (vs0) representing the motion of a first magnetic field vector;
[0146] A second multi-axis magnetometer (102, 104) is arranged in a second station at a second position and configured to output a second vector signal (vs1, vs3) representing the motion of the second magnetic field vector; and
[0147] The signal processor (501), coupled to receive a first vector signal (vs0) and a second vector signal (vs1), is configured to:
[0148] The first value of the parameter of the first multidimensional transformation (C1) is determined based on the optimization of the difference between the second vector signal (vs1) and the first compensation signal (cs1); wherein the first compensation signal is generated from the transformation of the first vector signal (vs0) according to the first multidimensional transformation (C1);
[0149] A compensated second vector signal (cvs1) is generated from the second vector signal (vs1) and the first compensated signal (cs1);
[0150] The detector signal (D) responding to the compensated second vector signal is determined to meet a predetermined criterion; and
[0151] At least in response to determining that the detector signal meets predetermined criteria, issue or abstain from issuing an initial alert regarding a possible theft-related incident.
[0152] The above electronic security system can be configured with a one-dimensional magnetometer, and the first transformation can be a one-dimensional transformation, such as multiplication or summation. This electronic security system can be equipped with a first and second magnetometer arranged in substantially the same orientation. The electronic security system can also be equipped with a first and second magnetometer arranged in an orientation different from mutually orthogonal orientations. The magnetometers can be arranged with a mutual orientation of less than about 60° (e.g., less than about 45°). In some embodiments, the second station includes a plurality of one-dimensional magnetometers arranged along a substantially vertical axis (e.g., arranged in an elongated, vertical, or upright body or bracket mount for fixation to a wall). Embodiments of the above electronic security system including one-dimensional magnetometers are defined in the dependent claims and the summary portion, wherein the multi-axis magnetometers can be replaced by one-dimensional magnetometers and / or one-dimensional transformations.
[0153] In some embodiments, the multi-axis magnetometer is a two-dimensional magnetometer. The electronic anti-theft system may be equipped with a first magnetometer and a second magnetometer arranged in an orientation different from their mutually orthogonal orientations. The magnetometers may be arranged in a mutual orientation of less than about 60° (e.g., less than about 45°). Each of the two-dimensional magnetometers may have substantially orthogonal axes.
[0154] As can be clearly seen from the above, it can more reliably detect theft-related incidents and at least reduce the risk of generating false alarms or failing to issue alarms when they should.
Claims
1. An electronic anti-theft system, comprising: A first multi-axis magnetometer is arranged in a first station at a first position and configured to output a first vector signal representing the motion of a first magnetic field vector. A second multi-axis magnetometer is arranged in a second station at a second position and is configured to output a second vector signal representing the motion of a second magnetic field vector. as well as A signal processor, coupled to receive the first vector signal and the second vector signal, and configured to: Estimate the first multidimensional transformation, which represents the difference between the first magnetic field vector and the second magnetic field vector; A first compensation signal is generated from the transformation of the first vector signal according to the first multidimensional transformation; A compensated second vector signal is generated from the second vector signal and the first compensation signal; The detector signal responding to the compensated second vector signal is determined to meet a predetermined criterion; and At least in response to determining that the detector signal meets the predetermined criteria, issue or abort issuing a first alarm regarding a possible theft-related incident.
2. The electronic anti-theft system according to claim 1, wherein, The first multidimensional transformation is estimated on a cyclical basis based on a first timing, wherein the first timing includes at least a first time and a second time.
3. The electronic anti-theft system according to claim 2, in, The difference between the second vector signal and the first compensation signal is determined during a parallel time period or a portion of the parallel time period between the first vector signal and the second vector signal. and Wherein, after the parallel time period, the compensated second vector signal is generated from the second vector signal based on the first value re-determined on the basis of the cycle.
4. The electronic anti-theft system according to claim 3 further includes: A third multi-axis magnetometer is arranged in the third station and configured to output a third vector signal representing the motion of the third magnetic field vector; The signal processor is further configured to: Estimate the second multidimensional transformation, which represents the difference between the first magnetic field vector and the third magnetic field vector; A second compensation signal is generated from the transformation of the first vector signal according to the second multidimensional transformation; A compensated third vector signal is generated from the third vector signal and the second compensation signal; The detector signal is in response to the compensated third vector signal.
5. The electronic anti-theft system according to claim 4 further includes: A fourth multi-axis magnetometer is arranged in the fourth station and configured to output a fourth vector signal representing the motion of the fourth magnetic field vector; The signal processor is further configured to: Estimate the third multidimensional transformation, which represents the difference between the first magnetic field vector and the fourth magnetic field vector; A third compensation signal is generated from the transformation of the first vector signal according to the third multidimensional transformation; A compensated fourth vector signal is generated from the fourth vector signal and the third compensation signal; The detector signal is responsive to the third compensation vector signal.
6. The electronic anti-theft system according to claim 5, wherein, The signal processor is also configured to: One or more of the first vector signal, the second vector signal, the third vector signal, and the fourth vector signal are bandpass filtered by a corresponding bandpass filter; wherein the corresponding bandpass filter has a lower cutoff frequency of less than 1.0 Hz and an upper cutoff frequency of more than 4 Hz and less than 50 Hz.
7. The electronic anti-theft system according to claim 6, wherein, One or more of the first multidimensional transformation, the second multidimensional transformation, and the third multidimensional transformation are estimated based on regularization applied during the iterative estimation of the parameters of the transformation, wherein the regularization is L1 regularization.
8. The electronic anti-theft system according to claim 7, wherein: The first vector signal is acquired in a first time period and a second time period, and the second vector signal is acquired in the first time period and the second time period; Estimate the first parameter of the first multidimensional transformation from the first vector signal and the second vector signal during the first time period; Estimate the first parameter of the first multidimensional transformation from the first vector signal and the second vector signal in the second time period; According to a first standard, the compensated second vector signal is generated based on the time after the second time, according to the first parameter estimated at the first time. According to the second standard, the compensated second vector signal is generated based on the time after the second time, according to the first parameter estimated at the second time.
9. The electronic anti-theft system according to claim 8, wherein, The first criterion is satisfied when the compensated second vector signal generated from the second vector signal in the first time period according to the first parameter estimated in the second time has a lower intensity than the compensated second vector signal generated from the second vector signal in the first time period according to the first parameter estimated in the first time.
10. The electronic anti-theft system according to claim 9, wherein: At the first time: a first parameter is estimated based on the first vector signal of the first time period, the second vector signal of the first time period, and the first compensation signal of the first time period; wherein, the first compensation signal of the first time period is generated from the first parameter estimated at the first time and the first vector signal of the first time period. At the second time: a first parameter is estimated based on the first vector signal of the second time period, the second vector signal of the second time period, and the first compensation signal of the second time period; and the first compensation signal is generated from the first parameter estimated at the second time and the first vector signal of the second time period. A first compensated second vector signal is generated from the second vector signal in the second time period, and the first compensated signal is generated from the first parameter estimated in the first time period and the first vector signal in the second time period. A second compensated second vector signal is generated from the second vector signal in the second time period, and the first compensated signal is generated from the first parameter estimated in the second time period and the first vector signal in the second time period; The signal processor is also configured to: Estimate the second vector signal of the first compensation and the second vector signal of the second compensation, and The system determines that the second vector signal of the first compensation is superior to the second vector signal of the second compensation, and generates the second vector signal of the compensation based on the first parameter estimated at the first time, while abandoning the generation of the second vector signal of the compensation based on the first parameter estimated at the second time.
11. The electronic anti-theft system according to claim 10, wherein, The signal processor is also configured to: The detection of the corresponding motions of the first magnetic field vector and the second magnetic field vector is performed through the following steps: - Estimate the first rotation of the first magnetic field vector and the second rotation of the second magnetic field vector; - Generate an indication signal that includes a reverse rotation or a rotation in the same direction; - Determine whether the first alarm is activated at least in response to the indication signal.
12. The electronic anti-theft system according to claim 11, wherein, The signal processor is also configured to: - Detect the corresponding motion of the first magnetic field vector and the second magnetic field vector; - After detecting the corresponding motion of the magnetic field vector, detect the start and duration of a fluctuation of at least the first magnetic field vector or the second magnetic field vector; wherein the duration of the fluctuation is determined according to a first timing criterion; -At least in response to determining the start and duration of fluctuations in at least the first magnetic field vector or the second magnetic field vector, determine whether to issue or abandon issuing a first alarm regarding a possible theft-related incident.
13. The electronic anti-theft system according to any one of claims 11 to 12, wherein, Detecting the corresponding motion of the first magnetic field vector and the second magnetic field vector includes: - Determine whether the motion of the first magnetic field vector and the second magnetic field vector corresponds to the basic horizontal motion of the magnet between the first station and the second station.
14. The electronic anti-theft system according to claim 12, wherein, Detecting the duration of the fluctuation includes: - Determine whether the motion of one or both of the first magnetic field vector and the second magnetic field vector corresponds to the oscillating motion of the magnet near one or both of the first station and the second station.
15. A method for detecting theft-related incidents, The system includes: A first multi-axis magnetometer is arranged in a first station at a first position and configured to output a first vector signal representing the motion of a first magnetic field vector. A second multi-axis magnetometer is arranged in a second station at a second position and is configured to output a second vector signal representing the motion of a second magnetic field vector. And a signal processor, coupled to receive the first vector signal and the second vector signal: Estimate the first multidimensional transformation, which represents the difference between the first magnetic field vector and the second magnetic field vector; A first compensation signal is generated from the transformation of the first vector signal according to the first multidimensional transformation; A compensated second vector signal is generated from the second vector signal and the first compensation signal; The detector signal responding to the compensated second vector signal is determined to meet a predetermined criterion; and At least in response to determining that the detector signal meets the predetermined criteria, issue or abort issuing a first alarm regarding a possible theft-related incident.
Citation Information
Patent Citations
Theft-preventing system and method with magnetic field detection
EP2997557B1
Magnetic sensor
CN106781144A
Alarm system for anti-theft magnetic snap
CN109147247A