Window length determination and security check method and device of sliding window

Abstract

The invention discloses a window length determination and security check method and device of a sliding window, and relates to the technical field of security check. The security check method comprises the following steps: acquiring a detection signal acquired by a coil sensor arranged on a security check door in real time; performing orthogonal demodulation processing on the detection signal to obtain an orthogonal demodulation signal; based on a set sliding window, performing sliding window processing with a set length on the orthogonal demodulation signal to extract orthogonal demodulation data corresponding to the orthogonal demodulation signal, the set length being consistent with the window length of the sliding window; and generating a security check result according to the orthogonal demodulation data. According to the invention, the installation and maintenance cost is reduced, the problem of false alarm is avoided, the detection of articles carried by passing personnel can be rapidly and accurately realized in real time, the problem of missing alarm is avoided, and the potential safety hazard is reduced.

Description

Technical Field

[0001] This application belongs to the technical field of security inspection, and particularly relates to a method and device for determining the window length of a sliding window and a security inspection method and device. Background Art

[0002] With the development of society, security gates are increasingly used in various places, mainly in crowded public places such as airports and railway stations, to detect whether people passing through are carrying prohibited items such as metals and liquids.

[0003] Currently, most security gates use infrared technology. Infrared emitters are installed on the security gate. When a person is detected passing through, the detection of the items carried by the person is triggered. However, infrared emitters are easily affected by environmental factors such as sunlight, lighting, temperature changes, and airflows, resulting in false alarms. In addition, the detection range of infrared emitters is limited, making it easy to miss detections, and they have high requirements for installation and calibration, prone to false alarms or missed detections. False alarms mean that even if no one enters the security gate area, the security gate will still start the detection and display the detection result, which not only wastes the time and resources of security personnel but also wrongly increases the count of the number of people passing through. Missed detections increase security risks. In addition, additional infrared emitters and corresponding control systems need to be installed on the security gate, increasing the installation and maintenance costs. Summary of the Invention

[0004] The purpose of the embodiments of this application is to provide a method and device for determining the window length of a sliding window, a security inspection method, device, electronic device, and storage medium to solve the problems in the related art, such as being prone to false alarms, which not only wastes the time and resources of security personnel but also wrongly increases the count of the number of people passing through, being prone to missed detections and increasing security risks, and the need to additionally install infrared emitters and corresponding control systems on the security gate, increasing the installation and maintenance costs.

[0005] To achieve the above purpose, the embodiments of this application adopt the following technical solutions:

[0006] In a first aspect, the embodiments of this application provide a method for determining the window length of a sliding window, including: obtaining the sampling frequency of a coil sensor installed on a security gate; determining the window length of the sliding window according to the sampling frequency of the coil sensor; the sliding window is used for: the security inspection device performs a sliding window process of a set length on the quadrature demodulation signal based on the sliding window to extract the quadrature demodulation data corresponding to the quadrature demodulation signal, the quadrature demodulation signal is obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor in real time, and the set length is the same as the window length; the quadrature demodulation data is used for: the security inspection device generating a security inspection result according to the quadrature demodulation data.

[0007] Second aspect, an embodiment of the present application provides a security inspection method, including: obtaining in real time a detection signal collected by a coil sensor disposed on a security inspection gate; performing quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal; based on a set sliding window, performing a sliding window process of a set length on the quadrature demodulation signal to extract quadrature demodulation data corresponding to the quadrature demodulation signal, the set length being consistent with the window length of the sliding window, and the window length being determined according to the window length determination method of the sliding window described in the first aspect embodiment of the present application; generating a security inspection result according to the quadrature demodulation data.

[0008] Third aspect, an embodiment of the present application provides a device for determining the window length of a sliding window, including: a first acquisition module for acquiring the sampling frequency of a coil sensor disposed on a security inspection gate; a determination module for determining the window length of the sliding window according to the sampling frequency of the coil sensor; the sliding window is used for: the security inspection device performing a sliding window process of a set length on the quadrature demodulation signal based on the sliding window to extract quadrature demodulation data corresponding to the quadrature demodulation signal, the quadrature demodulation signal being obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor obtained in real time, and the set length being consistent with the window length; the quadrature demodulation data is used for: the security inspection device generating a security inspection result according to the quadrature demodulation data.

[0009] Fourth aspect, an embodiment of the present application provides a security inspection device, including: a second acquisition module for obtaining in real time a detection signal collected by a coil sensor disposed on a security inspection gate; a first processing module for performing quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal; a second processing module for performing a sliding window process of a set length on the quadrature demodulation signal based on a set sliding window to extract quadrature demodulation data corresponding to the quadrature demodulation signal, the set length being consistent with the window length of the sliding window, and the window length being determined according to the device for determining the window length of the sliding window described in the third aspect embodiment of the present application; a generation module for generating a security inspection result according to the quadrature demodulation data.

[0010] Fifth aspect, an embodiment of the present application provides an electronic device, including: a processor, a memory, and a program or instruction stored on the memory and executable on the processor, and when the program or instruction is executed by the processor, it implements the steps of the method described in the first aspect embodiment of the present application, or implements the steps of the method described in the second aspect embodiment of the present application.

[0011] Sixth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, it implements the steps of the method described in the embodiment of the first aspect of the present application, or implements the steps of the method described in the embodiment of the second aspect of the present application.

[0012] The above at least one technical solution adopted in the embodiment of the present application can achieve the following beneficial effects:

[0013] The embodiment of the present application does not require additional installation of infrared pair emitters and corresponding control systems, reducing the installation and maintenance costs, and avoiding the false alarm problems caused by the infrared pair emitters being vulnerable to environmental factors and having high requirements for installation and calibration, as well as the waste of the time and resources of security personnel caused by false alarms, and the problem of incorrect increase in the counted number of passing people. In addition, since the detection signals collected by the coil sensors on the security door are obtained in real time, and the quadrature demodulation signals are processed by using a sliding window method to generate detection results, as long as a person passes through, the collected signals will definitely change compared with when no one passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection results can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And by using the sliding window method, the amount of data processed each time is small, so the detection of items carried by passing people can be realized in real time, quickly and accurately, avoiding the missed alarm problems caused by the limited detection range of infrared pair emitters and their high requirements for installation and calibration, and reducing the security risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0015] Figure 1 It is a schematic flowchart of a security inspection method provided by an embodiment of the present application;

[0016] Figure 2 It is a schematic diagram of a security door provided by an embodiment of the present application;

[0017] Figure 3 It is a schematic diagram of a receiving coil provided by an embodiment of the present application;

[0018] Figure 4 It is a schematic diagram of a sliding window processing provided by an embodiment of the present application;

[0019] Figure 5 It is a schematic flowchart of a security inspection method provided by another embodiment of the present application;

[0020] Figure 6 Flow chart of a method for determining the window length of a sliding window provided by an embodiment of the present application;

[0021] Figure 7 Flow chart of a method for determining the window length of a sliding window provided by another embodiment of the present application;

[0022] Figure 8 Flow chart of a method for determining the window length of a sliding window provided by another embodiment of the present application;

[0023] Figure 9 Flow chart of a method for determining the window length of a sliding window provided by another embodiment of the present application;

[0024] Figure 10 Structural diagram of a device for determining the window length of a sliding window provided by an embodiment of the present application;

[0025] Figure 11 Structural diagram of a device for determining the window length of a sliding window provided by another embodiment of the present application;

[0026] Figure 12 Structural diagram of a security inspection device provided by an embodiment of the present application;

[0027] Figure 13 Structural diagram of a security inspection device provided by another embodiment of the present application;

[0028] Figure 14 Structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0029] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0030] The terms "first", "second", etc. in this application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application can be implemented in an order other than those illustrated or described herein. In addition, "and / or" in this application means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after. It should be noted that all the data involved in this application are obtained under the premise of obtaining user authorization.

[0031] With the development of society, the application scenarios of security gates are increasing day by day. They are mainly used in crowded public places such as airports and stations to detect whether the passing personnel are carrying prohibited items such as metals and liquids. In the related art, most security gates use infrared technology. An infrared emitter is set on the security gate. When it detects that a person passes by, it triggers the detection of the items carried by the person. However, the infrared emitter is easily affected by environmental factors, such as sunlight, lighting, temperature changes, airflows, etc., resulting in false alarms. In addition, the detection range of the infrared emitter is limited, usually defined according to the height and width of the security gate. In some cases, such as when a person is below the detection range of the infrared emitter or enters at a special angle, they cannot be accurately detected and are prone to missed alarms. Moreover, the infrared emitter has high requirements for installation and calibration. Adjustments such as the installation position, angle, and sensitivity all require experience and professional knowledge, which are prone to false alarms or missed alarms. False alarms mean that even if no one enters the security gate area, the security gate will still start the detection and display the detection result, which not only wastes the time and resources of security personnel but also wrongly increases the statistics of the number of people passing through. Missed alarms mean that people carrying prohibited items cannot be accurately detected, increasing the security risk. In addition, an additional infrared emitter and the corresponding control system need to be set on the security gate, increasing the purchase, installation, and maintenance costs after a failure. Therefore, this application proposes a method for determining the window length of a sliding window, a security inspection method, device, electronic device, and storage medium.

[0032] The following will detail the technical solutions provided by each embodiment of this application in conjunction with the accompanying drawings.

[0033] Figure 1 It is a flowchart of a security inspection method provided for an embodiment of this application. As Figure 1 shown, the security inspection method of the embodiment of this application may specifically include the following steps:

[0034] S101, obtain in real time the detection signal collected by the coil sensor set on the security gate.

[0035] In the embodiments of the present application, the execution subject of the security inspection method of the present application is a security inspection device, which can be arranged in an electronic device. The electronic device can be a terminal device or a server. Among them, the terminal device can be a mobile phone, a tablet computer, a desktop computer, a portable notebook, a vehicle-mounted device, etc.; the server can be an independent server or a server cluster composed of multiple servers. Specifically, the electronic device can be arranged in a security inspection terminal used by security personnel.

[0036] As Figure 2 shown, the security inspection door can be composed of two left and right door panels 201 and a top box body 202. The coil sensor includes a coil 203 and a corresponding signal processing module. The coil 203 is arranged in the two left and right door panels. One is a transmitting door panel with a transmitting coil arranged inside, and the other is a receiving door panel with a receiving coil arranged inside. Key electronic devices such as a board card and a power supply are arranged in the top box body 202.

[0037] Since the acquisition of the detection signal is mainly related to the receiving coil, the receiving coil will be mainly described below. Figure 3 A simple example designed for the receiving coil of the receiving door panel. As Figure 3 shown, the entire receiving door panel 301 is evenly divided into 8 detection zones according to height. Each detection zone uses an independent receiving coil 302 to collect detection signals, that is, each detection zone has an independent receiving coil responsible for detection. Therefore, the signal processing module in the coil sensor can collect the detection signals of the receiving coils at eight places on the receiving door panel.

[0038] Among them, dividing the receiving coil into multiple independent detection zones has the following advantages:

[0039] 1) Improve signal strength: By dividing the receiving coil into multiple detection zones, the intensity of the received detection signal can be increased. The receiving coil of each detection zone only needs to receive signals at specific positions, and the mutual interference with other receiving coils is reduced. This separation of detection zones can improve the intensity and stability of the detection signal and reduce signal attenuation caused by interference.

[0040] 2) Reduce noise interference: When each detection zone uses an independent receiving coil for detection, the signal interference caused by environmental noise and interference can be reduced. Because each receiving coil only focuses on signals at specific positions and does not have to process interference sources at other positions, the signal-to-noise ratio between the signal and the noise can be improved, making it easier to distinguish and analyze the target signal.

[0041] 3) Improve anti-interference ability: Using independent receiving coils for detection in different zones can enhance the anti-interference ability of the system. When a certain receiving coil is interfered with, the other receiving coils can still operate normally. This redundant design can reduce the failure of the entire system caused by the malfunction or interference of a single receiving coil, improving safety and reliability.

[0042] 4) Optimize signal demodulation: Using independent receiving coils for detection can more easily demodulate and analyze the detection signals of each security zone. This can improve the accuracy and efficiency of the signal processing algorithm, thereby enhancing the performance and response speed of the system.

[0043] The receiving coil can be set in an inverted eight shape, and the inverted eight shape design has the following advantages:

[0044] 1) Cancel noise interference: The inverted eight-shaped coil design can form a symmetric layout in space, where the current direction of one coil is opposite to that of the adjacent coil. This symmetry helps to cancel the influence of the external magnetic field on each other and reduce the impact of noise interference. By eliminating or reducing the interference signal, the inverted eight-shaped coil can improve the signal-to-noise ratio of the system, making the detection more accurate.

[0045] 2) Modular design: The layout of the inverted eight-shaped coil usually adopts a modular design, that is, the receiving coils of the entire security door are divided into multiple identical modules. This modular design makes the manufacturing, maintenance, and upgrade of the security door more convenient. If a certain module in the receiving coil fails, the module can be replaced separately without replacing the entire security door system, thus saving labor, material, and time costs.

[0046] The working principle of the coil sensor is as follows:

[0047] 1) When a person passes through the security door, the charged particles (such as electrons) in the body and the carried metal items will cut the magnetic field around the receiving coil.

[0048] 2) This magnetic field change will cause an induced electromotive force (Electro Magnetic Fields, abbreviated as EMF) to be generated in the receiving coil. The magnitude of the induced electromotive force is related to the rate of magnetic field change and the material of the coil. Since the movement speed of the human body is relatively slow and the number of charged particles in the body is very small, the electromotive force caused by the human body cutting the magnetic field induced in the receiving coil is relatively weak. Most of the induced electromotive force is caused by the carried metal items cutting the magnetic field.

[0049] 3) To convert the induced electromotive force into a processable electrical signal, a signal processing module in the coil sensor needs to perform a series of amplification, filtering, and conversion operations on the induced electromotive force. First, the induced electromotive force is converted into a differential voltage signal by a differential amplifier, and then the differential voltage signal is input into a pre-amplifier for further amplification and filtering. Additionally, it can also go through multiple-stage amplifiers for further gain and filtering to improve the signal-to-noise ratio and sensitivity.

[0050] 4) After the above signal processing, the signal is input into an analog-to-digital converter (ADC) in the signal processing module for digital processing. The ADC samples the input signal and converts it into a digital signal through analog-to-digital conversion. Finally, the detection signal obtained by the security inspection device is 8 consecutive digital signals of the receiving coils.

[0051] It should be noted here that different from the infrared technology where the detection of the detection signal collected by the receiving coils between a person entering and leaving is only started after the infrared is triggered, in the embodiment of the present application, the detection signal collected by the receiving coils is obtained in real time and subsequent detections are performed. Therefore, as long as a person passes through the security door, the signal collected by the receiving coils will definitely change compared to when no person passes through. Based on this part of the change, the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. Thus, the detection of items carried by passing personnel can be achieved in real time, quickly, and accurately, avoiding the false alarm problem caused by the limited detection range of the infrared emitter and the high requirements for installation and calibration, and reducing the safety hazard.

[0052] S102, perform quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal.

[0053] In the embodiment of the present application, quadrature demodulation is also called IQ (In-Phase, Quadrature) modulation. The quadrature demodulation signal includes an in-phase component signal, a quadrature component signal, an amplitude signal, and a phase signal. The specific process of quadrature demodulation processing is as follows:

[0054] The digital signal of the transmitting coil serves as a reference signal. Performing a Hilbert transform on the reference signal is equivalent to generating a 90-degree phase difference in the frequency domain, and the resulting signal is called an envelope signal. The reference signal and the envelope signal correspond to two orthogonal local oscillators (commonly known as the I-channel and the Q-channel) in quadrature demodulation, where the I-channel is the reference signal and the Q-channel is the envelope signal. Mix (i.e., multiply) the I-channel and the Q-channel with the detection signal of the receiving coil respectively, and perform low-pass filtering on the signals after mixing these two channels to remove high-frequency components and only retain the baseband signal (i.e., remove the influence of the transmitting coil and only retain the influence caused by people and objects passing through the security gate), obtaining the in-phase component (corresponding to the I-channel) signal I and the quadrature component (corresponding to the Q-channel) signal Q, and calculating the amplitude signal as The phase signal is

[0055] S103, based on the set sliding window, perform a sliding window process of a set length on the quadrature demodulation signal to extract the quadrature demodulation data corresponding to the quadrature demodulation signal, and the set length is the same as the window length of the sliding window.

[0056] In the embodiments of the present application, the sliding window, abbreviated as the sliding window, the window length m and the sliding step n of the sliding window are pre-set. The window length is also the window size. The sliding step refers to the distance that the sliding window slides each time. Among them, the window length m of the sliding window can be determined with reference to the embodiments of the window length determination method of the following sliding window, which will not be elaborated here.

[0057] In the embodiments of the present application, a smaller step can provide higher time resolution, corresponding to higher detection sensitivity, but the calculation cost will also increase. Therefore, comprehensive tests and verifications can be pre-conducted for a specific application environment to obtain the optimal sliding step. The sliding step can be fixed or adjusted according to needs.

[0058] Such as Figure 4 shown, perform a sliding window process of a set length on the in-phase component signal, the quadrature component signal, the amplitude signal, and the phase signal respectively to extract the quadrature demodulation data corresponding to the quadrature demodulation signal: the in-phase component data, the quadrature component data, the amplitude data, and the phase data.

[0059] Specifically, create a sliding window, and move the sliding window on the quadrature demodulation signal according to the window length m and the sliding step n of the sliding window. Usually, the starting position of the sliding window starts from the beginning of the quadrature demodulation signal and moves backward with a fixed sliding step. The quadrature demodulation signal does not stop, and the sliding window does not stop. Specifically, when the length of the received quadrature demodulation data is equal to the window length m, output the data for subsequent processing, slide the sliding window, discard the first n data, then receive n data to restore the data length to m, output the data for subsequent processing, and repeat the above process.

[0060] S104, generate the security inspection result based on the quadrature demodulation data.

[0061] In the embodiment of the present application, based on the quadrature demodulation data output in step S103, it is determined whether there are prohibited items, and the corresponding security inspection result is generated. The security inspection result may specifically include the existence of prohibited items, the non-existence of prohibited items, etc. For example, it can be determined whether there are prohibited items by means of database comparison or model judgment, etc.

[0062] In summary, the security inspection method of the embodiment of the present application does not require additional installation of infrared pairs and corresponding control systems, reducing the installation and maintenance costs, and avoiding the false alarm problems caused by the fact that infrared pairs are vulnerable to environmental factors and have high requirements for installation and calibration, as well as the waste of the time and resources of security personnel caused by false alarms, and the problem of incorrect increase in the statistics of the number of people passing through. In addition, since the detection signal collected by the coil sensor on the security door is obtained in real time, and the sliding window method is used to process the quadrature demodulation signal and generate the detection result, as long as a person passes through, the collected signal will definitely change compared with when no one passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And through the sliding window method, the amount of data processed each time is small, so the detection of items carried by passing personnel can be realized in real time, quickly, and accurately, avoiding the missed alarm problems caused by the limited detection range of infrared pairs and the high requirements for installation and calibration, and reducing the security risks.

[0063] Figure 5 It is a schematic flowchart of a security inspection method provided by another embodiment of the present application. As Figure 5 shown, on the basis of the embodiment shown in Figure 1 the security inspection method of the embodiment of the present application may specifically include the following steps:

[0064] S501, obtain in real time the detection signal collected by the coil sensor provided on the security door.

[0065] In the embodiment of the present application, step S501 is the same as step S101 in the above embodiment, and will not be elaborated here.

[0066] S502, perform band-pass filtering and / or downsampling processing on the detection signal.

[0067] In the embodiment of the present application, before performing quadrature demodulation processing on the measurement signal, band-pass filtering processing and / or downsampling processing can be performed on the reference signal of the transmitting coil and the detection signal of the receiving coil respectively.

[0068] Band-pass filtering processing, that is, only allowing waves in a specific frequency band to pass through while blocking waves in other frequency bands. Through band-pass filtering processing, the noise interference in the detection signal and the reference signal can be reduced, and the noise components that do not match the frequency of the target signal can be filtered out, thereby improving the quality of the detection signal and the reference signal.

[0069] Downsampling processing, also known as decimation processing, is a technique of multi-rate digital signal processing or a process of reducing the signal sampling rate, usually used to reduce the data transmission rate or the data size. The downsampling in the embodiments of the present application refers to reducing the data transmission rate. For example, the sampling frequency of the receiving coil is 200 KHz (kilohertz), that is, 200,000 points are sampled per second. After downsampling processing, 1000 points are sampled per second. Through downsampling processing, the amount of data can be reduced, the data processing efficiency can be improved, and the storage resources can be saved.

[0070] S503, perform quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal.

[0071] In the embodiments of the present application, step S503 is the same as step S102 in the above embodiments, and will not be elaborated here.

[0072] S504, based on a set sliding window, perform a sliding window process of a set length on the quadrature demodulation signal to extract the quadrature demodulation data corresponding to the quadrature demodulation signal, and the set length is the same as the window length of the sliding window.

[0073] In the embodiments of the present application, step S504 is the same as step S103 in the above embodiments, and will not be elaborated here.

[0074] Step S104 "generate a security inspection result according to the quadrature demodulation data" in the above embodiments specifically may include the following step S505:

[0075] S505, input the quadrature demodulation data into the target detection model to obtain the security inspection result output by the target detection model.

[0076] In the embodiments of the present application, a powerful target detection model can be pre-trained using deep learning to enable it to have the ability to distinguish different types of contraband. The target detection model can be trained for various scenarios such as when no one passes through, when a person carrying daily necessities passes through, and when a person carrying contraband passes through, so as to achieve accurate target recognition and classification.

[0077] Input the quadrature demodulation data into the target detection model. The target detection model extracts features from the input quadrature demodulation data, determines the type of the detected item such as a controlled knife according to the extracted features, generates a security inspection result according to the type of the detected item, and outputs the security inspection result.

[0078] Among them, the target detection model can specifically be a convolutional neural network model. Using a convolutional neural network (CNN for short) model for target detection has the following advantages:

[0079] 1) The most important operation in a convolutional neural network is convolution, which can effectively extract the features of quadrature demodulation data. The convolution operation is a form of local connection and weight sharing operation, which can efficiently extract the spatial structure information of the input data. Through multiple layers of convolution operations, the convolutional neural network can gradually transform the input raw data into a more representative feature map, thereby achieving effective feature extraction and dimensionality reduction of the data.

[0080] 2) The pooling operation in a convolutional neural network is another commonly used feature extraction method, which can further reduce the dimension of the feature map and shorten the training time by downsampling the feature map. At the same time, the pooling operation can also enhance the robustness of the model.

[0081] 3) A convolutional neural network usually adopts a multi-layer network structure. By stacking multiple layers of convolution, pooling and other operations, rich feature information can be gradually extracted, and effective feature combination and extraction can be carried out. The multi-layer network structure can represent the input raw data at different abstraction levels, so as to better understand the information contained in the data.

[0082] Therefore, using a convolutional neural network model for target detection, the detection result is more accurate.

[0083] S506, when the security inspection result is that there are prohibited items, output a warning message.

[0084] In the embodiment of the present application, when the security inspection result is that there are prohibited items, a warning message is output so that security personnel can take actions in time. The warning message can specifically include but is not limited to various forms such as sound alarm information, light indication information, vibration prompt information, remote alarm information, etc.

[0085] In summary, the security inspection method of the embodiment of the present application does not require additional installation of infrared pair emitters and corresponding control systems, reducing the installation and maintenance costs. Moreover, it avoids the false alarm problems caused by the infrared pair emitters being vulnerable to environmental factors and having high requirements for installation and calibration, as well as the waste of the time and resources of security personnel and the incorrect increase in the statistical number of passing people due to false alarms. Additionally, since the detection signals collected by the coil sensors on the security inspection door are obtained in real time and the quadrature demodulation signals are processed using a sliding window to generate detection results, as long as a person passes through, the collected signals will definitely change compared to when no one passes through, and this part of the change can definitely be collected through the sliding window, thereby obtaining the corresponding detection results. Therefore, even if a person enters at a special angle, they can still be accurately detected. And through the sliding window method, the amount of data processed each time is small, so the detection of items carried by passing people can be achieved in real time, quickly, and accurately, avoiding the missed alarm problems caused by the limited detection range of infrared pair emitters and their high requirements for installation and calibration, and reducing potential safety hazards. In addition, using a target detection model to generate security inspection results, a target detection model based on deep learning, such as a convolutional neural network model, can make the generated security inspection results more accurate. The sliding step size can be adjusted as needed, taking into account both detection sensitivity and computational cost. In the case where the security inspection result indicates the presence of prohibited items, a warning message is output, which can facilitate security personnel to take timely actions when there are potential safety hazards. Before performing quadrature demodulation processing on the detection signals, band-pass filtering processing is performed on the detection signals and reference signals, which can reduce the noise interference in the detection signals and reference signals, filter out the noise components that do not match the frequency of the target signal, and thus improve the quality of the detection signals and reference signals. Before performing quadrature demodulation processing on the detection signals, downsampling processing is performed on the detection signals and reference signals, which can reduce the amount of data, improve the data processing efficiency, and save storage resources.

[0086] Figure 6 FIG. is a schematic flow chart of a method for determining the window length of a sliding window provided by an embodiment of the present application. As Figure 6 shown, the method for determining the window length of the sliding window in the embodiment of the present application may specifically include the following steps:

[0087] S601, obtain the sampling frequency of the coil sensors provided on the security inspection door.

[0088] In the embodiment of the present application, if the sliding window is too small, that is, the window length of the sliding window is too small, it may not be possible to capture the important features of the target; if the sliding window is too large, that is, the window length of the sliding window is too large, it may contain too much redundant information, reducing the detection sensitivity and causing a computational burden. The sampling frequency of the coil sensors determines the accuracy and response speed of the detection signals collected by the coil sensors. Therefore, the window length of the sliding window should take into account the sampling frequency of the coil sensors.

[0089] S602. Determine the window length of the sliding window according to the sampling frequency of the coil sensor.

[0090] In the embodiments of the present application, since the higher the sampling frequency of the coil sensor, the better the accuracy and response speed of the detection signal collected by the coil sensor. At this time, even if the window length of the sliding window is set relatively small, the important features of the target can still be captured. Therefore, the higher the sampling frequency of the coil sensor, the longer the determined window length of the sliding window.

[0091] Among them, the sliding window is used for: the security inspection device performs a sliding window process of a set length on the quadrature demodulation signal based on the sliding window to extract the quadrature demodulation data corresponding to the quadrature demodulation signal. The quadrature demodulation signal is obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor obtained in real time, and the set length is the same as the window length; the quadrature demodulation data is used for: the security inspection device generates a security inspection result according to the quadrature demodulation data. For the specific process of implementing security inspection based on the sliding window with the determined window length, refer to the relevant descriptions in the embodiments of the above security inspection method, which will not be elaborated here.

[0092] In summary, for the method for determining the window length of the sliding window in the embodiments of the present application, the window length of the sliding window is determined according to the sampling frequency of the coil sensor provided on the security door, so that the window length of the sliding window is reasonably set, avoiding the problem that if the window length is too small, the important features of the target may not be captured, and if the window length is too large, there may be too much redundant information, reducing the detection sensitivity and causing a computational burden. The solution for implementing security inspection based on the sliding window with the determined window length does not require additional installation of infrared transceivers and corresponding control systems, reducing the installation and maintenance costs, and avoiding the false alarm problem caused by the fact that infrared transceivers are vulnerable to environmental factors and have high requirements for installation and calibration, as well as the waste of time and resources of security personnel caused by false alarms and the problem of incorrect increase in the statistics of the number of people passing through. In addition, since the detection signal collected by the coil sensor on the security door is obtained in real time, and the quadrature demodulation signal is processed by means of a sliding window to generate a detection result, as long as a person passes through, the collected signal will definitely change compared with when no person passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And by means of the sliding window, the amount of data processed each time is small, so the detection of items carried by passing personnel can be realized in real time, quickly and accurately, avoiding the missed alarm problem caused by the limited detection range of infrared transceivers and their high requirements for installation and calibration, and reducing the security risk.

[0093] Figure 7A method for determining the window length of a sliding window provided in another embodiment of this application. As Figure 7 shown, based on the embodiment shown in Figure 6 , the method for determining the window length of the sliding window in the embodiment of this application may specifically include the following steps:

[0094] S701, obtain the sampling frequency of the coil sensor provided on the security gate.

[0095] In the embodiment of this application, step S701 is the same as step S601 in the above embodiment, and will not be elaborated here.

[0096] When the quadrature demodulation data is used for: the security inspection device inputs the quadrature demodulation data into the target detection model to obtain the security inspection result output by the target detection model, step S602 "determine the window length of the sliding window according to the sampling frequency of the coil sensor" in the above embodiment may specifically include the following steps S702.

[0097] S702, determine the window length according to the sampling frequency of the coil sensor and the data input size of the target detection model.

[0098] In the embodiment of this application, when using the target detection model to generate the security inspection result according to the quadrature demodulation data, since the target detection model will receive inputs of a fixed size during the training phase, the window length of the sliding window should match the input size expected by the target detection model.

[0099] Comprehensively consider the sampling frequency of the coil sensor and the input size of the target detection model, and set the appropriate window length of the sliding window.

[0100] As Figure 8 shown, the specific process of step S702 for determining the window length of the sliding window can be realized through the following steps:

[0101] S801, determine the range of the window length according to the sampling frequency of the coil sensor and the average time for a person to pass through the security gate.

[0102] In the embodiment of this application, as a first feasible implementation manner, the time for a person to pass through the security gate can be obtained through the monitoring image when the person passes through the security gate, and the average value of the time for a large number of people (which can be the same person or different people) to pass through the security gate is calculated to calculate the average time for a person to pass through the security gate.

[0103] As a second feasible implementation, the time when the current person passes through the security gate can be obtained by analyzing the characteristics of the detection signal. By averaging the times when a large number of people (which can be the same person or different people) pass through the security gate, the average time for a person to pass through the security gate can be calculated. For example, the time when a person passes through the security gate can be determined by obtaining the change time points of the detection signal when the person passes through the security gate. Based on the change time points, the time when the person passes through the security gate can be determined. According to the times when the person passes through the security gate determined multiple times, the average time for the person to pass through the security gate can be determined. Specifically, when no one passes through the security gate, the security gate is in an empty field state, and the detection signal is a straight line that tends to be stable. During the period from when a person enters the gate to when they leave the gate, the detection signal will change, that is, there will be at least the following time points: the time points when the straight line changes to a curve and when the curve returns to the straight line again. These time points affect the selection of the start and end positions of the sliding window, that is, the selection of the window length of the sliding window. In fact, these time points can reflect the time when a person passes through the security gate.

[0104] For example:

[0105] First, collect the data when a large number of people (which can be the same person or different people) pass through the security gate to ensure that these data cover the actual application scenarios. Then, analyze the characteristics of these data (that is, the characteristics of the detection signal), and find two time points (when the straight line changes to a curve and when the curve returns to the straight line) for each detection signal to determine the length of the detection signal. Assume that the sampling frequency of the coil sensor is 200KHz, that is, 200,000 points are sampled per second. After downsampling, 1,000 points are sampled per second. By analyzing the data when a person passes through the security gate, the signal lengths of most data are between 600 points and 800 points, that is, the time for most people to pass through the security gate is between 0.6 seconds and 0.8 seconds. Therefore, the range of the window length of the sliding window can be initially determined to be that the minimum can be 600 and the maximum can be 800, that is, [600, 800].

[0106] S802. Determine multiple candidate window lengths according to the input size of the target detection model and the range of the window length.

[0107] In the embodiments of the present application, the structures of the target detection models are different, and the requirements for the input size are also different. Assume that the input size required by the target detection model is a multiple of 8 (i.e., the length of the data input to the model is a multiple of 8). Based on the range of the window length of the sliding window in the above example being [600, 800], 26 candidate window lengths can be determined, with values of 600, 608, 616…, 800 respectively (for example, one candidate window length is 600 sampling points, one candidate window length is 608 sampling points, and so on). These 26 values are all multiples of 8. Since these 26 values all meet the requirements and are all suitable window lengths, it is necessary to rely on subsequent tests to determine which one is the best.

[0108] S803. Test the detection performance of the target detection model according to the candidate window lengths to obtain the test results.

[0109] In the embodiments of the present application, the detection performance of the target detection model may specifically include, but is not limited to, the response speed, calculation efficiency, and target detection accuracy of the target detection model, etc.

[0110] The response speed refers to the ability of the security gate to quickly respond to the detected items, and can also be understood as the calculation time consumption, that is, the time consumption from collecting data by the coil sensor to calculating the security inspection result. When the window length is small, the time consumption is low, and when the window length is large, the time consumption is high.

[0111] The calculation efficiency refers to the actual utilization rate of the security gate under the given hardware resources. If the window length is set large, the utilization rate of the calculation processing unit (processor) is high. Long-term excessive utilization rate will cause the processing unit to continuously work in a high-load state. High load means high power consumption. Long-term high power consumption is likely to increase the power supply pressure of the system, cause the device to overheat, and thus lead to failures or shorten the life of the device. In addition, high utilization rate indicates that the calculation processing unit has reached or is close to the limit of its processing capacity, which makes it difficult to expand or upgrade the system in the future because the existing computing resources are already unable to bear more workloads.

[0112] The target detection accuracy is the proportion of the security inspection results output by the target detection model that are consistent with the actual security inspection results.

[0113] Based on each candidate window length determined in step S802, test the detection performance of the target detection model to obtain the test results corresponding to each candidate window length, that is, the response speed, calculation efficiency, and target detection accuracy. Specifically, it includes the following steps:

[0114] When the window length is the candidate window length, obtain the time length between when the coil sensor collects the detection signal and when the target detection model outputs the security inspection result, as the test result corresponding to the response speed. When the window length is the candidate window length, obtain the utilization rate of the processor of the security door, as the test result corresponding to the computing efficiency. When the window length is the candidate window length, obtain the ratio of the security inspection result output by the target detection model to the actual security inspection result, as the test result corresponding to the target detection accuracy. Then, based on one or more test results, reasonably determine the window length among multiple candidate window lengths.

[0115] S804. Determine the window length among multiple candidate window lengths according to the test results.

[0116] In the embodiments of the present application, for example, at least one of the response speed, computing efficiency, and target detection accuracy corresponding to each candidate window length can be used as a screening basis to determine one as the final window length among multiple candidate window lengths. Exemplarily, the candidate window length corresponding to the case where the response speed is greater than the response speed threshold can be used as the final window length, or the candidate window length corresponding to the case where the computing efficiency is greater than the preset computing efficiency threshold can be used as the final window length, or the candidate window length corresponding to the case where the target detection accuracy is greater than the accuracy threshold can be used as the final window length, etc. Of course, the final window length can also be determined based on multiple test results.

[0117] Exemplarily, as Figure 9 shown, determining the final window length among multiple candidate window lengths can be specifically implemented through the following steps:

[0118] S901. Among multiple candidate window lengths, screen out the first candidate window length whose response speed is greater than the preset response speed threshold and whose computing efficiency is greater than the preset computing efficiency threshold.

[0119] In the embodiments of the present application, the response speed and computing efficiency are used as constraint conditions, and the target detection accuracy is used as a decisive factor. Specifically, among multiple candidate window lengths, exclude the values that do not meet the response speed and computing efficiency, that is, exclude the candidate window lengths whose response speed is equal to or less than the preset response speed threshold, or whose computing efficiency is equal to or less than the preset computing efficiency threshold, and retain the remaining candidate window lengths, that is, retain the candidate window lengths whose response speed is greater than the preset response speed threshold and whose computing efficiency is greater than the preset computing efficiency threshold, denoted as the first candidate window length.

[0120] S902. Among the first candidate window lengths, screen out the second candidate window length with the highest target detection accuracy.

[0121] In the embodiment of the present application, among the first candidate window lengths screened out, the first candidate window length with the highest target detection accuracy is screened out and denoted as the second candidate window length.

[0122] S903. Determine the second candidate window length as the window length.

[0123] In summary, for the method for determining the window length of the sliding window in the embodiment of the present application, according to the sampling frequency of the coil sensor provided on the security door, the window length of the sliding window is determined, so that the window length of the sliding window is reasonably set, avoiding the problem that if the window length is too small, important features of the target may not be captured, and if the window length is too large, there may be too much redundant information, reducing the detection sensitivity and causing a computational burden. The solution for implementing security inspection based on the sliding window with the determined window length does not require additional installation of infrared cross-beam sensors and corresponding control systems, reducing the installation and maintenance costs, and avoiding the false alarm problem caused by the fact that infrared cross-beam sensors are susceptible to environmental factors and have high requirements for installation and calibration, as well as the problems of wasting the time and resources of security personnel caused by false alarms and the incorrect increase in the statistics of the number of people passing through. In addition, since the detection signal collected by the coil sensor on the security door is obtained in real time, and the quadrature demodulation signal is processed by means of a sliding window to generate a detection result, as long as a person passes through, the collected signal will definitely change compared with when no person passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And by means of the sliding window, the amount of data processed each time is small, so the detection of items carried by passing personnel can be realized in real time, quickly and accurately, avoiding the missed alarm problem caused by the limited detection range of infrared cross-beam sensors and their high requirements for installation and calibration, and reducing the security risk. When setting the window length of the sliding window, the input size of the target detection model is considered, so that the data input into the target detection model meets the requirements of the model input size, and thus the output result of the model is more accurate. When setting the window length of the sliding window, the test results of the detection performance of the target detection model are combined, so that the detection performance of the target detection model is better.

[0124] Figure 10 This is a schematic structural diagram of a device for determining the window length of a sliding window provided by an embodiment of the present application. As Figure 10 shown, the device 1000 for determining the window length of the sliding window in the embodiment of the present application may specifically include: a first acquisition module 1001 and a determination module 1002.

[0125] The first acquisition module 1001 is configured to acquire the sampling frequency of the coil sensor provided on the security door.

[0126] A determination module 1002 is configured to determine the window length of a sliding window according to the sampling frequency of a coil sensor; the sliding window is used for: the security inspection device to perform a sliding window process of a set length on the quadrature demodulation signal based on the sliding window to extract quadrature demodulation data corresponding to the quadrature demodulation signal, the quadrature demodulation signal is obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor obtained in real time, and the set length is consistent with the window length; the quadrature demodulation data is used for: the security inspection device to generate a security inspection result according to the quadrature demodulation data.

[0127] In the embodiments of the present application, for the specific processes of each module and unit in the window length determination device of the sliding window in the embodiments of the present application to implement their functions, reference may be made to the relevant descriptions in the embodiments of the method for determining the window length of the sliding window above, and details are not described herein again.

[0128] In summary, the window length determination device of the sliding window in the embodiments of the present application determines the window length of the sliding window according to the sampling frequency of the coil sensor provided on the security inspection door, so that the window length of the sliding window is reasonably set, avoiding the problem that if the window length is too small, important features of the target may not be captured, and if the window length is too large, there may be too much redundant information, reducing the detection sensitivity and causing a computational burden. The solution for implementing security inspection based on the sliding window with the determined window length does not require additional installation of infrared transceivers and corresponding control systems, reducing the installation and maintenance costs, and avoiding the false alarm problems caused by the infrared transceivers being susceptible to environmental factors and having high requirements for installation and calibration, as well as the waste of the time and resources of security personnel and the incorrect increase in the statistics of the number of people passing through due to false alarms. In addition, since the detection signal collected by the coil sensor on the security inspection door is obtained in real time, and the quadrature demodulation signal is processed by means of a sliding window to generate a detection result, as long as a person passes through, the collected signal will definitely change compared with when no person passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And by means of the sliding window, the amount of data processed each time is small, so the detection of items carried by passing personnel can be realized in real time, quickly and accurately, avoiding the missed alarm problems caused by the limited detection range of the infrared transceivers and their high requirements for installation and calibration, and reducing the security risks.

[0129] Figure 11 It is a structural schematic diagram of a window length determination device of a sliding window provided in another embodiment of the present application. As Figure 11 shown, in the window length determination device 1000 of the sliding window in the embodiments of the present application, when the quadrature demodulation data is used for: the security inspection device inputs the quadrature demodulation data into a target detection model to obtain a security inspection result output by the target detection model, the determination module 1002 further includes: a determination unit 1101. Among them:

[0130] A determination unit 1101, configured to determine a window length according to the sampling frequency of a coil sensor and the data input size of a target detection model.

[0131] In a feasible implementation manner of an embodiment of this application, the determination unit 1101 is further configured to: determine a range of the window length according to the sampling frequency of the coil sensor and the average time for a person to pass through the security checkpoint; determine a plurality of candidate window lengths according to the data input size of the target detection model and the range of the window length; test the detection performance of the target detection model according to the candidate window lengths to obtain a test result; and determine the window length from the plurality of candidate window lengths according to the test result.

[0132] In a feasible implementation manner of an embodiment of this application, the determination unit 1101 is further configured to: obtain the change time points of the detection signal when a person passes through the security checkpoint; determine the time for the person to pass through the security checkpoint according to the change time points; and determine the average time for a person to pass through the security checkpoint according to the times for the person to pass through the security checkpoint determined multiple times.

[0133] In a feasible implementation manner of an embodiment of this application, the detection performance of the target detection model includes at least one of the following: the response speed of the target detection model, the calculation efficiency, and the target detection accuracy rate; the determination unit 1101 is further configured to perform at least one of the following: obtain the time length between when the coil sensor collects the detection signal and when the target detection model outputs the security inspection result when the window length is the candidate window length, as the test result corresponding to the response speed; obtain the utilization rate of the processor of the security checkpoint when the window length is the candidate window length, as the test result corresponding to the calculation efficiency; and obtain the ratio of the security inspection result output by the target detection model and the actual security inspection result being consistent when the window length is the candidate window length, as the test result corresponding to the target detection accuracy rate.

[0134] In a feasible implementation manner of an embodiment of this application, the detection performance of the target detection model includes the response speed of the target detection model, the calculation efficiency, and the target detection accuracy rate; the determination unit 1101 is further configured to: screen out a first candidate window length whose response speed is greater than a preset response speed threshold and whose calculation efficiency is greater than a preset calculation efficiency threshold from the plurality of candidate window lengths; screen out a second candidate window length with the highest target detection accuracy rate from the first candidate window lengths; and determine the second candidate window length as the window length.

[0135] In an embodiment of this application, for the specific processes of each module and unit in the window length determination device of the sliding window in the embodiment of this application to implement their functions, reference may be made to the relevant descriptions in the embodiment of the window length determination method of the sliding window above, which will not be elaborated here.

[0136] In summary, the window length determination device of the sliding window according to the embodiments of the present application determines the window length of the sliding window according to the sampling frequency of the coil sensor provided on the security gate, so that the window length of the sliding window is reasonably set, avoiding the problem that if the window length is too small, important features of the target may not be captured, and if the window length is too large, there may be too much redundant information, reducing the detection sensitivity and causing a computational burden. The solution for implementing security inspection based on the sliding window with the determined window length does not require additional installation of infrared transceivers and corresponding control systems, reducing the installation and maintenance costs, and avoiding the false alarm problem caused by the fact that infrared transceivers are vulnerable to environmental factors and have high requirements for installation and calibration, as well as the problem of wasting the time and resources of security personnel caused by false alarms and the problem of incorrect increase in the statistics of the number of people passing through. In addition, since the detection signal collected by the coil sensor on the security gate is obtained in real time and the quadrature demodulation signal is processed by using the sliding window method to generate a detection result, as long as a person passes through, the collected signal will definitely change compared with when no person passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And by using the sliding window method, the amount of data processed each time is small, so the detection of items carried by passing people can be realized in real time, quickly and accurately, avoiding the missed alarm problem caused by the limited detection range of infrared transceivers and their high requirements for installation and calibration, and reducing the security risk. When setting the window length of the sliding window, the input size of the target detection model is considered, so that the data input into the target detection model meets the requirements of the model input size, and further makes the output result of the model more accurate. When setting the window length of the sliding window, the test results of the detection performance of the target detection model are combined, so that the detection performance of the target detection model is better.

[0137] Figure 12 It is a schematic structural diagram of a security inspection device provided by an embodiment of the present application. As Figure 12 shown, the security inspection device 1200 according to the embodiment of the present application may specifically include: a second acquisition module 1201, a first processing module 1202, a second processing module 1203, and a generation module 1204. Among them:

[0138] The second acquisition module 1201 is configured to acquire in real time the detection signal collected by the coil sensor provided on the security gate.

[0139] The first processing module 1202 is configured to perform quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal.

[0140] The second processing module 1203 is configured to perform a sliding window process of a set length on the quadrature demodulation signal based on a set sliding window, so as to extract quadrature demodulation data corresponding to the quadrature demodulation signal. The set length is the same as the window length of the sliding window, and the window length is determined by a device for determining the window length of the sliding window according to any of the above embodiments.

[0141] The generation module 1204 is configured to generate a security inspection result according to the quadrature demodulation data.

[0142] In the embodiments of the present application, for the specific processes of each module and unit in the security inspection device of the embodiments of the present application to implement their functions, reference may be made to the relevant descriptions in the embodiments of the above security inspection method, which will not be elaborated here.

[0143] In summary, the security inspection device of the embodiments of the present application does not require additional installation of infrared pair emitters and corresponding control systems, reducing the installation and maintenance costs, and avoiding the false alarm problems caused by the fact that infrared pair emitters are vulnerable to environmental factors and have high requirements for installation and calibration, as well as the waste of time and resources of security personnel and the incorrect increase in the statistics of the number of people passing through caused by false alarms. In addition, since the detection signals collected by the coil sensors on the security inspection door are obtained in real time, and the quadrature demodulation signals are processed by means of a sliding window to generate detection results, as long as a person passes through, the collected signals will definitely change compared with when no person passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection results can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And by means of the sliding window, the amount of data processed each time is small, so the detection of items carried by passing personnel can be realized in real time, quickly and accurately, avoiding the missed alarm problems caused by the limited detection range of infrared pair emitters and the high requirements for installation and calibration, and reducing the security risks.

[0144] Figure 13 It is a schematic structural diagram of a security inspection device provided in another embodiment of the present application. As Figure 13 shown, on the basis of the embodiment shown in Figure 12 In the security inspection device 1200 of the embodiments of the present application, the generation module 1204 further includes: a generation unit 1301. Wherein:

[0145] The generation unit 1301 is configured to input the quadrature demodulation data into a target detection model, and obtain a security inspection result output by the target detection model. The target detection model is used to extract features from the quadrature demodulation data, determine the types of detected items according to the extracted features, and generate a security inspection result according to the types of detected items.

[0146] In the embodiments of the present application, for the specific processes of each module and unit in the security inspection device of the embodiments of the present application to implement their functions, reference may be made to the relevant descriptions in the embodiments of the above security inspection method, which will not be elaborated here.

[0147] In summary, the security inspection device of the embodiments of the present application does not require additional installation of infrared pair emitters and corresponding control systems, reducing the installation and maintenance costs. Moreover, it avoids the false alarm problems caused by the fact that infrared pair emitters are vulnerable to environmental factors and have high requirements for installation and calibration, as well as the waste of the time and resources of security personnel and the incorrect increase in the statistics of the number of people passing through due to false alarms. Additionally, since the detection signals collected by the coil sensors on the security inspection door are obtained in real time, and the quadrature demodulation signals are processed using a sliding window method to generate detection results, as long as a person passes through, the collected signals will definitely change compared to when no one passes through, and this part of the change can definitely be collected through the sliding window, thereby obtaining the corresponding detection results. Therefore, even if a person enters at a special angle, they can still be accurately detected. And through the sliding window method, the amount of data processed each time is small, so the detection of items carried by passing people can be realized in real time, quickly, and accurately, avoiding the missed alarm problems caused by the limited detection range of infrared pair emitters and their high requirements for installation and calibration, and reducing potential safety hazards. In addition, using a target detection model to generate security inspection results, a target detection model based on deep learning, such as a convolutional neural network model, can make the generated security inspection results more accurate. The sliding step size can be adjusted as needed, taking into account both detection sensitivity and computational cost. When the security inspection result indicates the presence of prohibited items, a warning message is output, which can facilitate security personnel to take timely actions when there are potential safety hazards. Before performing quadrature demodulation processing on the detection signals, band-pass filtering processing is performed on the detection signals and reference signals, which can reduce the noise interference in the detection signals and reference signals and filter out the noise components that do not match the frequency of the target signal, thereby improving the quality of the detection signals and reference signals. Before performing quadrature demodulation processing on the detection signals, downsampling processing is performed on the detection signals and reference signals, which can reduce the amount of data, improve the data processing efficiency, and save storage resources.

[0148] The embodiments of the present application further provide an electronic device. As Figure 14As shown, the electronic device 1400 can vary significantly due to different configurations or performances. It may include one or more processors 1401 and a memory 1402. One or more applications or data may be stored in the memory 1402. Among them, the memory 1402 can be short-term storage or persistent storage. The applications stored in the memory 1402 may include one or more modules (not shown in the figure). Each module may include a series of computer-executable instructions for the electronic device 1400. Further, the processor 1401 can be set to communicate with the memory 1402 and execute a series of computer-executable instructions in the memory 1402 on the electronic device 1400. The electronic device 1400 may also include one or more power supplies 1403, one or more wired or wireless network interfaces 1404, one or more input / output interfaces 1405, and one or more keyboards 1406.

[0149] Specifically, in this embodiment, the electronic device includes a memory and one or more programs. One or more of the programs are stored in the memory, and one or more of the programs may include one or more modules. Each module may include a series of computer-executable instructions for the electronic device and is configured to be executed by one or more processors. The one or more programs include the following computer-executable instructions for:

[0150] Obtain the sampling frequency of the coil sensor set on the security gate;

[0151] Determine the window length of the sliding window according to the sampling frequency of the coil sensor; the sliding window is used for: the security inspection device performs a sliding window process of a set length on the quadrature demodulation signal based on the sliding window to extract the quadrature demodulation data corresponding to the quadrature demodulation signal. The quadrature demodulation signal is obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor obtained in real time. The set length is the same as the window length; the quadrature demodulation data is used for: the security inspection device generates a security inspection result according to the quadrature demodulation data.

[0152] Or, it is used for the following computer-executable instructions:

[0153] Obtain in real time the detection signal collected by the coil sensor set on the security gate;

[0154] Perform quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal;

[0155] Based on the set sliding window, perform a sliding window process of a set length on the quadrature demodulation signal to extract the quadrature demodulation data corresponding to the quadrature demodulation signal. The set length is the same as the window length of the sliding window, and the window length is determined according to the method for determining the window length of the sliding window in any of the above embodiments;

[0156] Generate security inspection results based on quadrature demodulation data.

[0157] The electronic device according to the embodiment of the present application does not require additional installation of an infrared pair emitter and the corresponding control system, reducing the installation and maintenance costs, and avoiding the false alarm problems caused by the infrared pair emitter being vulnerable to environmental factors and having high requirements for installation and calibration, as well as the waste of the time and resources of security personnel and the incorrect increase in the number of people passing through due to false alarms. In addition, since the detection signal collected by the coil sensor on the security inspection door is obtained in real time, and the quadrature demodulation signal is processed by using a sliding window to generate a detection result, as long as a person passes through, the collected signal will definitely change compared with when no person passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And through the sliding window method, the amount of data processed each time is small, so the detection of items carried by passing people can be realized in real time, quickly, and accurately, avoiding the missed alarm problems caused by the limited detection range of the infrared pair emitter and its high requirements for installation and calibration, and reducing the security risks. Determine the window length of the sliding window according to the sampling frequency of the coil sensor set on the security inspection door, so that the window length of the sliding window is reasonably set, avoiding the problem that the window length is too small and may not be able to capture the important features of the target, and the window length is too large and may contain too much redundant information, reducing the detection sensitivity and causing a computational burden.

[0158] The embodiment of the present application also proposes a readable storage medium, on which one or more computer programs are stored. The one or more computer programs include instructions. When the program or instructions are executed by a processor in an electronic device including multiple application programs, the processor in the electronic device can execute each process of the above security inspection method embodiment, and specifically be used to execute:

[0159] Obtain the sampling frequency of the coil sensor set on the security inspection door;

[0160] Determine the window length of the sliding window according to the sampling frequency of the coil sensor; The sliding window is used for: The security inspection device performs a sliding window process with a set length on the quadrature demodulation signal based on the sliding window to extract the quadrature demodulation data corresponding to the quadrature demodulation signal. The quadrature demodulation signal is obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor obtained in real time, and the set length is the same as the window length; The quadrature demodulation data is used for: The security inspection device generates security inspection results according to the quadrature demodulation data.

[0161] Or, specifically be used to execute:

[0162] Obtain in real time the detection signal collected by the coil sensor set on the security door;

[0163] Perform quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal;

[0164] Based on a set sliding window, perform a sliding window process of a set length on the quadrature demodulation signal to extract the quadrature demodulation data corresponding to the quadrature demodulation signal. The set length is the same as the window length of the sliding window, and the window length is determined according to the window length determination method of the sliding window in any of the above embodiments;

[0165] Generate a security inspection result according to the quadrature demodulation data.

[0166] The readable storage medium of the embodiment of the present application does not require additional installation of an infrared pair emitter and the corresponding control system, reducing the installation and maintenance costs, and avoiding the false alarm problem caused by the infrared pair emitter being vulnerable to environmental factors and having high requirements for installation and calibration, as well as the waste of the time and resources of security personnel and the problem of incorrect increase in the statistics of the number of people passing through due to false alarms. In addition, since the detection signal collected by the coil sensor on the security door is obtained in real time, and the quadrature demodulation signal is processed and a detection result is generated by using a sliding window method, as long as a person passes through, the collected signal will definitely change compared with when no person passes through, and this part of the change can definitely be collected through the sliding window, and then the corresponding detection result can be obtained. Therefore, even if a person enters at a special angle, they can still be accurately detected. And through the sliding window method, the amount of data processed each time is small, so the detection of items carried by passing people can be realized in real time, quickly and accurately, avoiding the missed alarm problem caused by the limited detection range of the infrared pair emitter and its high requirements for installation and calibration, and reducing the security risk. Determine the window length of the sliding window according to the sampling frequency of the coil sensor set on the security door, so that the window length of the sliding window is reasonably set, avoiding the problem that the window length is too small and may not be able to capture the important features of the target, and the window length is too large and may contain too much redundant information, reducing the detection sensitivity and causing a computational burden.

[0167] The system, device, module or unit illustrated in the above embodiments can be specifically implemented by a computer processing unit or an entity, or by a product with a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0168] For the convenience of description, when describing the above device, it is divided into various units according to functions for separate description. Of course, when implementing the present application, the functions of each unit can be implemented in one or more software and / or hardware.

[0169] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0170] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0171] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0172] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0173] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0174] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0175] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0176] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0177] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0178] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and reference can be made to the relevant part of the method embodiment for the relevant content.

[0179] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for determining the window length of a sliding window, characterized in that Including: Obtaining the sampling frequency of the coil sensor disposed on the security gate; Determining the window length of the sliding window according to the sampling frequency of the coil sensor; the sliding window is used for: the security inspection device to perform a sliding window process of a set length on the quadrature demodulation signal based on the sliding window to extract the quadrature demodulation data corresponding to the quadrature demodulation signal, and the quadrature demodulation signal is obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor obtained in real time, and the set length is consistent with the window length; The quadrature demodulation data is used for: the security inspection device to generate a security inspection result according to the quadrature demodulation data.

2. The method according to claim 1, characterized in that, The quadrature demodulation data is used for: the security inspection device to input the quadrature demodulation data into the target detection model to obtain the security inspection result output by the target detection model; The determining the window length of the sliding window according to the sampling frequency of the coil sensor includes: Determining the window length according to the sampling frequency of the coil sensor and the data input size of the target detection model.

3. The method according to claim 2, wherein The determining the window length according to the sampling frequency of the coil sensor and the data input size of the target detection model includes: Determining the range of the window length according to the sampling frequency of the coil sensor and the average time for a person to pass through the security gate; Determining a plurality of candidate window lengths according to the data input size of the target detection model and the range of the window length; Testing the detection performance of the target detection model according to the candidate window lengths to obtain a test result; Determining the window length from the plurality of candidate window lengths according to the test result.

4. The method according to claim 3, wherein Also including: Obtaining the change time point of the detection signal when a person passes through the security gate; Determining the time for a person to pass through the security gate according to the change time point; Determining the average time for a person to pass through the security gate according to the time for a person to pass through the security gate determined multiple times.

5. The method according to claim 3, characterized in that, The detection performance of the target detection model includes at least one of the following: the response speed, calculation efficiency, and target detection accuracy of the target detection model; The testing the detection performance of the target detection model according to the candidate window lengths to obtain a test result includes at least one of the following: Obtaining the time length between when the coil sensor collects the detection signal and the target detection model outputs the security inspection result when the window length is the candidate window length as the test result corresponding to the response speed; Obtaining the utilization rate of the processor of the security gate when the window length is the candidate window length as the test result corresponding to the calculation efficiency; Obtaining the ratio of the security inspection result output by the target detection model and the actual security inspection result being consistent when the window length is the candidate window length as the test result corresponding to the target detection accuracy.

6. The method according to claim 3, wherein The detection performance of the target detection model includes the response speed, calculation efficiency, and target detection accuracy of the target detection model; The determining the window length from the plurality of candidate window lengths according to the test result includes: Among the multiple candidate window lengths, screen out the first candidate window length whose response speed is greater than a preset response speed threshold and whose calculation efficiency is greater than a preset calculation efficiency threshold; Among the first candidate window lengths, screen out the second candidate window length with the highest target detection accuracy; Determine the second candidate window length as the window length.

7. A security inspection method, characterized in that, Including: Obtain in real time the detection signal collected by the coil sensor arranged on the security gate; Perform quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal; Based on a set sliding window, perform a sliding window process of a set length on the quadrature demodulation signal to extract the quadrature demodulation data corresponding to the quadrature demodulation signal, where the set length is consistent with the window length of the sliding window, and the window length is determined according to the method for determining the window length of the sliding window described in any one of claims 1-6; Generate a security inspection result according to the quadrature demodulation data.

8. The method according to claim 7, wherein The generating a security inspection result according to the quadrature demodulation data includes: Input the quadrature demodulation data into a target detection model to obtain the security inspection result output by the target detection model, where the target detection model is used to extract features from the quadrature demodulation data, determine the type of the item to be detected according to the extracted features, and generate the security inspection result according to the type of the item to be detected.

9. A window length determination device for a sliding window, characterized in that Including: A first acquisition module for acquiring the sampling frequency of the coil sensor arranged on the security gate; A determination module for determining the window length of the sliding window according to the sampling frequency of the coil sensor; the sliding window is used for: the security inspection device performs a sliding window process of a set length on the quadrature demodulation signal based on the sliding window to extract the quadrature demodulation data corresponding to the quadrature demodulation signal, the quadrature demodulation signal is obtained by performing quadrature demodulation processing on the detection signal collected by the coil sensor acquired in real time, and the set length is consistent with the window length; The quadrature demodulation data is used for: the security inspection device generates a security inspection result according to the quadrature demodulation data.

10. A security inspection device, characterized in that, Including: A second acquisition module for acquiring in real time the detection signal collected by the coil sensor arranged on the security gate; A first processing module for performing quadrature demodulation processing on the detection signal to obtain a quadrature demodulation signal; A second processing module for performing a sliding window process of a set length on the quadrature demodulation signal based on a set sliding window to extract the quadrature demodulation data corresponding to the quadrature demodulation signal, where the set length is consistent with the window length of the sliding window, and the window length is determined according to the device for determining the window length of the sliding window described in claim 9; A generation module for generating a security inspection result according to the quadrature demodulation data.

11. An electronic device, characterized in that, Including a processor, a memory, and a program or instruction stored on the memory and executable on the processor, where when the program or instruction is executed by the processor, the steps of the method described in any one of claims 1-6 are implemented, or the steps of the method described in any one of claims 7-8 are implemented.

12. A readable storage medium, characterized in that, The program or instructions are stored on the readable storage medium, and when the program or instructions are executed by the processor, the steps of the method described in any one of claims 1-6 are implemented, or the steps of the method described in any one of claims 7-8 are implemented.

Images