CN110793978A

CN110793978A - Speed correction method, article detection method and apparatus

Info

Publication number: CN110793978A
Application number: CN201810869568.5A
Authority: CN
Inventors: 张兆宇; 底欣; 徐怡; 田军
Original assignee: Fujitsu Ltd
Current assignee: Fujitsu Ltd
Priority date: 2018-08-02
Filing date: 2018-08-02
Publication date: 2020-02-14
Anticipated expiration: 2038-08-02
Also published as: CN110793978B

Abstract

The embodiment of the invention provides a speed correction method, an article detection method and a device, wherein when the angular speed of the swing changes and/or when the maximum normal distance caused by the relative position change of an article and a predetermined shaft changes, the speed correction device corrects a first speed vector according to the value of the angular speed before and after the change or the value of the maximum normal distance before and after the change, and obtains a second speed vector. Thus, the radial velocity of the article caused by the article swinging about the predetermined axis can be corrected, and the radial velocity corresponding to the article placed at any angle or the article at any swinging angular velocity can be obtained.

Description

Speed correction method, article detection method and apparatus

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a speed correction method, an article detection method, and an article detection apparatus.

Background

In recent years, the safety problem in public places is more and more emphasized, and how to detect dangerous goods such as control instruments, flammable and explosive goods and the like becomes an important problem. At present, a detection device for dangerous goods is widely applied to various dense occasions such as airports, railway stations, subway stations, stadiums and the like.

Hazardous material detection devices can be divided into two types: contact and contactless. Contact detection devices require that a suspicious object (e.g., a bottle containing a liquid) be placed on the detection device for detection, while non-contact detection devices are capable of initiating detection and distinguishing whether the suspicious object is a dangerous object when the suspicious object moves within a certain range of the detection device.

It should be noted that the above background description is only for the sake of clarity and complete description of the technical solutions of the present invention and for the understanding of those skilled in the art. Such solutions are not considered to be known to the person skilled in the art merely because they have been set forth in the background section of the invention.

Disclosure of Invention

At present, aiming at a non-contact detection device, one of common detection methods is an X-ray detection method, but the method is generally high in cost, and long-term use of the method can affect the physical health of workers; in addition, since different articles are made of different materials, their reflection characteristics are also different. Such a difference can be used for detecting an article, that is, a sensor is arranged to transmit a signal to the article to be detected, and the intensity of the reflected signal of the article to be detected is compared with training data in a training set to realize object detection. Since dangerous goods are often concealed on a person, the detection of the goods can be performed by utilizing the influence of the swing of the goods concealed on the person on the reflected signal of the goods along with the walking of the person.

However, the inventor has found that when the object to be detected is hidden on a person, the hiding angle cannot be a specific one, or the swing generated along with the walking of the person cannot be a fixed one. Different hiding angles or different swing angular velocities can cause different radial velocities of the object to be detected due to the swing, and the training set can only contain one or a few preset angles or training data which are corresponding to one or a few swing angular velocities and are related to the radial velocities, and can not exhaust all the training data which are corresponding to all the angles or swing angular velocities and are related to the radial velocities, so that when the actual test data are compared with the training data of the training set, the missing detection and the false detection of the object to be detected can be caused due to the inconsistency of the hiding angles or the swing angular velocities.

The embodiment of the invention provides a speed correction method and a speed correction device, which can correct the radial speed generated by the article swinging relative to a preset shaft, so that the radial speed corresponding to the article placed at any angle or the article at any swinging angular speed can be obtained.

The embodiment of the invention also provides an article detection method and device, which utilize the influence of the swing of the article hidden on the human body along with the walking of the human on the reflected signal of the article to detect the article, and utilize the speed correction method to expand the training set or correct the test data, thereby realizing the accurate detection of the article hidden on the human body and having lower detection cost.

According to a first aspect of embodiments of the present invention, there is provided a speed correction apparatus in which an article oscillates about a predetermined axis to produce a first speed vector in a radial direction; the device includes:

and a correcting unit for correcting the first velocity vector based on a value of a maximum normal distance before and after a first change, which is a change in angular velocity of the oscillating movement, or a value of a maximum normal distance before and after a second change, which is a change in the maximum normal distance from the article to the predetermined axis due to a change in relative positions of the article and the predetermined axis, to obtain a second velocity vector.

According to a second aspect of embodiments of the present invention, there is provided an article detection apparatus, wherein the apparatus comprises:

the acquisition unit is used for acquiring a training set comprising at least two second feature sets in advance;

the receiving and transmitting unit is used for transmitting a transmitting signal to an article to be detected and receiving a reflected signal reflected by the article to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction;

a processing unit for processing the transmission signal and the reflection signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit;

the determining unit is used for comparing a first feature set extracted from the feature signal with a second feature set in the training set and determining the articles contained in the articles to be detected according to the comparison result;

the transmitting and receiving unit is also used for transmitting the transmitting signal to a known article and receiving a training reflected signal reflected by the known article; wherein the known article oscillates about a predetermined axis, the known article producing a first velocity vector in a radial direction;

the processing unit is further used for processing the transmitting signal and the training reflected signal to obtain a training characteristic signal; the training characteristic signal comprises a signal intensity value corresponding to a first preset number group of speed values corresponding to a preset distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit; a second feature set extracted from the training feature signal;

the acquisition unit includes: the velocity correction device and the generation unit according to the first aspect;

the speed correction device performs speed correction on the first speed vector corresponding to the second feature set according to the value of the speed before and after the first change or the value of the maximum normal distance before and after the second change to obtain at least one second speed vector corresponding to at least one group of values before and after the change; the generating unit generates at least one second feature set corresponding to the at least one second velocity vector, combines the extracted second feature set with the generated at least one second feature set, and generates the training set.

According to a third aspect of the embodiments of the present invention, there is provided an article detection apparatus, wherein the apparatus includes a transceiver unit, a processing unit, an extraction unit, the speed correction apparatus of the first aspect, and a determination unit;

the transmitting and receiving unit transmits a transmitting signal to an article to be detected and receives a reflected signal reflected by the article to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction;

the processing unit is used for processing the transmitting signal and the reflected signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit;

the extraction unit is used for extracting a first feature set from the feature signal;

the speed correction device performs speed correction on a third speed vector corresponding to the first feature set according to the value of the speed before and after the first change or the value of the maximum normal distance before and after the second change to obtain at least one second speed vector corresponding to at least one group of values before and after the change;

the determining unit is configured to generate at least one first feature set corresponding to the at least one second velocity vector; and comparing the first feature set extracted by the extraction unit with the generated at least one feature set respectively with the feature sets in the training set to obtain at least two detection results, and processing the at least two detection results to determine the articles contained in the articles to be detected.

The embodiment of the invention has the advantages that the radial speed generated by the articles swinging relative to the preset shaft can be corrected, so that the radial speed corresponding to the articles placed at any angle or the articles at any swinging angular speed can be obtained. And/or, utilize the swing produced along with the walking of people to the reflected signal influence of the article hidden on the human body, come to carry on the detection of the article, utilize the above-mentioned speed correction method to expand the training set or correct the test data, and then realize the detection of the article hidden on the human body accurately, the detection cost is lower.

Specific embodiments of the present invention are disclosed in detail with reference to the following description and drawings, indicating the manner in which the principles of the invention may be employed. It should be understood that the embodiments of the invention are not so limited in scope. The embodiments of the invention include many variations, modifications and equivalents within the spirit and scope of the appended claims.

Features that are described and/or illustrated with respect to one embodiment may be used in the same way or in a similar way in one or more other embodiments, in combination with or instead of the features of the other embodiments.

It should be emphasized that the term "comprises/comprising" when used herein, is taken to specify the presence of stated features, integers, steps or components but does not preclude the presence or addition of one or more other features, integers, steps or components.

Drawings

Many aspects of the invention can be better understood with reference to the following drawings. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. For convenience in illustrating and describing some parts of the present invention, corresponding parts may be enlarged or reduced in the drawings. Elements and features depicted in one drawing or one embodiment of the invention may be combined with elements and features shown in one or more other drawings or embodiments. Moreover, in the drawings, like reference numerals designate corresponding parts throughout the several views, and may be used to designate corresponding parts for use in more than one embodiment.

In the drawings:

FIG. 1 is a schematic view of a speed correction apparatus in the present embodiment 1;

FIG. 2 is a schematic view illustrating the swing of the object to be inspected in the embodiment 1;

FIG. 3 is a schematic diagram of an embodiment of a calibration unit in the embodiment 1;

FIG. 4 is a schematic view of the present embodiment 1 illustrating the shaft deflection;

FIG. 5 is a schematic diagram of an embodiment of a calibration unit in the embodiment 1;

FIG. 6 is a schematic view of the present embodiment 1 illustrating the shaft deflection;

FIG. 7 is a schematic diagram of an embodiment of a calibration unit in the embodiment 1;

FIG. 8 is a schematic view of the article detection apparatus in this embodiment 2;

FIG. 9 is a schematic diagram of the microwave sensor transmitting signal;

FIG. 10 is a schematic view showing the constitution of a processing unit in the present embodiment 1;

FIG. 11 is a schematic view of the article detection apparatus in this embodiment 3;

FIG. 12 is a schematic diagram showing the hardware configuration of the speed correction apparatus in this embodiment 4;

FIG. 13 is a schematic diagram showing the hardware configuration of the article detection apparatus in this embodiment 5;

FIG. 14 is a flowchart of a speed correction method in the present embodiment 6;

FIG. 15 is a flowchart of an article detection method according to this embodiment 7;

fig. 16 is a flowchart of the article detection method in embodiment 8.

Detailed Description

The foregoing and other features of embodiments of the present invention will become apparent from the following description, taken in conjunction with the accompanying drawings. These embodiments are merely exemplary and are not intended to limit the present invention. In order to enable those skilled in the art to easily understand the principle and the implementation manner of the present invention, the embodiment of the present invention is described by taking the example of transmitting the microwave signal, but it is understood that the embodiment of the present invention is not limited to transmitting the microwave signal.

The following describes a specific embodiment of the present invention with reference to the drawings.

Example 1

This embodiment 1 provides a speed correction apparatus in which an article is oscillated about a predetermined axis to generate a first speed vector in a radial direction; fig. 1 is a schematic view of the configuration of the apparatus, and as shown in fig. 1, the apparatus 100 includes:

a correcting unit 101 for correcting the first velocity vector based on a value of the angular velocity before and after a first change which is a change in the angular velocity of the oscillating movement or a value of the maximum normal distance before and after a second change which is a change in the maximum normal distance from the article to the predetermined axis due to a change in the relative position of the article and the predetermined axis, to obtain a second velocity vector.

In this embodiment, the article is oscillated about a predetermined axis to produce a first velocity vector in a radial direction; radial refers to the direction of oscillation.

Fig. 2 is a schematic view of the article being oscillated, as shown in fig. 2, about a predetermined axis. During the swinging, the article swings along the A axis and the B axis opposite to the A axis on two sides of the preset axis. The swing may be a plurality of reciprocating motions, or may be a single swing along the a axis and the B axis radially opposite to the a axis, and the embodiment is not limited thereto. The radial direction (the direction of the first velocity vector) is the direction of the a-axis or the B-axis; the driving force generated by the driving unit may control the article to swing, or the article may be placed on a human body and swing as the human walks, which is not limited in this embodiment. In addition, fig. 2 illustrates the predetermined axis in the middle of the object, but the present embodiment is not limited thereto, and the predetermined axis may be located at one side of the object, and the article swings along the a-axis or the B-axis at one side of the predetermined axis to generate a first velocity vector along the a-axis or the B-axis (radial direction).

In this embodiment, since the article may swing along with the walking of the person, the article may also move in a radial direction along with the walking of the person in addition to the swing about the predetermined axis, that is, the article may also move in a radial direction in the walking direction as a whole in addition to the first velocity vector, and the radial direction may be the same as the walking speed of the person. For convenience of description, the magnitude of the first velocity vector is a moving velocity of the human reference object, the first velocity vectors at the respective positions on the article are the same or different, and the objects to be corrected are the velocity vectors at the respective positions on the article.

The velocity correction caused by the first variation is explained below with reference to fig. 3.

When the article swings faster or slower (e.g., as a person walks, the change in the person's step frequency causes the article to swing), the angular velocity of the swing changes, which in turn causes the first velocity vector to change.

Fig. 3 is a schematic diagram of an embodiment of the calibration unit 101, and as shown in fig. 3, the calibration unit 101 includes:

a first calculation unit 301 for calculating a first ratio of the angular velocity after the first change and the angular velocity before the first change; the magnitude of the second velocity vector is equal to the product of the first ratio and the magnitude of the first velocity vector.

In this embodiment, the first velocity vector may be regarded as a linear velocity corresponding to the angular velocity, the linear velocity is changed at the same rate as the angular velocity, the second velocity vector is in the same direction as the first velocity vector, and the ratio of the magnitude of the second velocity vector to the magnitude of the first velocity vector is equal to the first ratio. The first ratio is used as a correction factor that can be corrected for the first velocity vector at each location on the article.

The velocity correction resulting from the second variation is explained below with reference to fig. 4-7.

In this embodiment, a deflection of the article in the plane of the article relative to the predetermined axis or a translation of the article in the plane of the article relative to the predetermined axis (e.g. a deflection or translation caused by a change in the concealed position and angle of the article as a person walks) both cause a change in the relative position of the article and the predetermined axis, the maximum normal distance representing the distance from the point on the article furthest from the predetermined axis to the predetermined axis.

Fig. 4 is a schematic view of the deflection of the article relative to the predetermined axis in the plane of the article resulting in a change in the relative position of the article and the predetermined axis, as shown in fig. 4, the article being deflected relative to the predetermined axis O by an angle α, the distance from point C on the article to the predetermined axis O being the maximum normal distance, the maximum normal distance changing from x to L, the point C resulting in a change in the magnitude of the first velocity vector (linear velocity) at point C due to the change in the maximum normal distance.

Fig. 5 is a schematic diagram of an embodiment of the calibration unit, and as shown in fig. 5, the calibration unit 101 includes:

a second calculating unit 501, configured to calculate a magnitude of the second velocity vector according to the angle of the deflection and a magnitude of the first velocity vector when the second change is a maximum normal distance change caused by the deflection of the article with respect to the axis on the plane on which the article is located.

In this embodiment, the second calculation unit 501 may include:

a first calculation module 5011 for calculating the maximum normal distance after the second change according to the angle of the deflection;

the second calculating module 5012 is configured to calculate a second ratio of the maximum normal distance after the second change to the maximum normal distance before the second change, and use a product of the second ratio and the size of the first velocity vector as the second velocity vector.

In this embodiment, the first calculation module 5011 can calculate the maximum normal distance after the second change L according to the following formula:

α is equal to the angle of deflection, x is equal to the maximum normal distance before the second change, and y is equal to the distance from the location on the article corresponding to the maximum normal distance before the second change to a horizontal axis perpendicular to the predetermined axis and passing through the center of mass or center of gravity of the article.

The second calculation module 5012 calculates the magnitude of the second velocity vector to be equal to:

wherein v is₁Equal to the magnitude of the first velocity vector, L/x is referred to as the second ratio.

After the offset, the direction of the second velocity vector of the location is opposite to the direction of the first velocity vector of the location as the location C of the maximum normal distance on the article changes from one side of the predetermined axis to the other.

Figure 6 is a schematic view of the article translating in the plane of the article with respect to the predetermined axis resulting in a change in the relative position of the article and the predetermined axis, as shown in figure 6, the article translating in the plane of the article with respect to the predetermined axis O by a distance d. Due to the translation, the distance from the point C on the article to the predetermined axis O is the maximum normal distance, the maximum normal distance is changed from x to L, and the size (linear velocity) of the first velocity vector of the point C is also changed due to the change of the maximum normal distance of the point C.

Fig. 7 is a schematic diagram of an embodiment of the calibration unit, and as shown in fig. 7, the calibration unit 101 includes:

a third calculating unit 701 for calculating a maximum normal distance before the second change from the first velocity vector and an angular velocity corresponding to the first velocity vector;

a magnitude calculation unit 702, configured to calculate a magnitude of the second velocity vector according to a distance that the article translates on a plane where the article is located with respect to the predetermined axis, the maximum normal distance before the second change, and a magnitude of the first velocity vector;

a direction determining unit 703 for determining the direction of the second velocity vector based on the distance of the translation and the direction of the first velocity vector.

In this embodiment, the third calculation unit 701 calculates the first velocity vector v according to point C in fig. 6₁And angular velocity ω of point C₁Calculating the maximum normal distance x before the second change to be equal to v₁/ω₁(third ratio), the magnitude calculation unit 702 calculates the magnitude v of the second velocity vector of the point C₂Equal to:

v₂＝ω₂x (x + d) formula (3);

after the axis translation, when the position C of the maximum normal distance on the article changes from one side to the other side of the predetermined axis, the direction determination unit 703 determines that the direction of the second velocity vector of the position is opposite to the direction of the first velocity vector of the position.

For convenience of explanation, the followingThe position C on the article in fig. 4 and 6 is illustrated, but the embodiment is not limited thereto, and the speed correction method for other positions on the article is the same as that for the position C, and the second ratio or the third ratio is used as the correction factor for v on other positions₁The correction is made (using equations 2 or 3) and is not exemplified here.

With the above-described embodiments, the radial velocity generated by the article swinging about the predetermined axis can be corrected, so that the radial velocity corresponding to an article placed at an arbitrary angle or an article at an arbitrary swinging angular velocity can be obtained, and the test data or training data at the time of article detection can be extended according to the velocity correction, so that the article detection accuracy can be improved, which will be described below with reference to embodiments 2 and 3.

Example 2

Embodiment 2 provides an article detection apparatus; fig. 8 is a schematic view of the configuration of the article detection apparatus, and as shown in fig. 8, the apparatus 800 includes: an acquisition unit 801, a transceiver unit 802, a processing unit 803, and a determination unit 804;

the obtaining unit 801 is configured to obtain a training set including at least two second feature sets in advance;

the transceiving unit 802 sends a transmitting signal to an article to be detected and receives a reflected signal reflected by the article to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction;

the processing unit 803 is configured to process the transmission signal and the reflection signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit;

the determining unit 804 is configured to compare the first feature set extracted from the feature signal with the second feature set in the training set, and determine an article included in the article to be detected according to a comparison result;

the transceiver unit 802 is further configured to send the transmission signal to a known article, and receive a training reflection signal reflected by the known article; wherein the known article oscillates about a predetermined axis, the known article producing a first velocity vector in a radial direction;

the processing unit 803 is further configured to process the transmission signal and the training reflection signal to obtain a training characteristic signal; the training characteristic signal comprises a signal intensity value corresponding to a first preset number group of speed values corresponding to a preset distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit; a second feature set extracted from the training feature signal;

the acquisition unit 801 includes: the speed correction device 100 and the generation unit 8011 in embodiment 1;

wherein, the speed calibration device 100 performs speed calibration on the first speed vector corresponding to the second feature set according to the value of the speed before and after the first change or the value of the maximum normal distance before and after the second change, so as to obtain at least one second speed vector corresponding to at least one set of values before and after the change; the generating unit 8011 generates at least one second feature set corresponding to the at least one second velocity vector, combines the extracted second feature set with the generated at least one second feature set, and generates the training set.

Therefore, the object hidden on the human body is detected by utilizing the influence of the swing of the object generated along with the walking of the human body on the reflected signal of the object, the training set is expanded by utilizing the speed correction, the object hidden on the human body is accurately detected, and the detection cost is low.

In this embodiment, the transceiver unit 802 has functions of transmitting and receiving signals, and may be implemented by a microwave sensor. For example, the transceiver 802 is a microwave sensor operating at 24.05 GHz-24.25 GHz, and transmits a microwave (e.g., millimeter Wave) signal, such as Frequency-modulated Continuous Wave (FMCW), to the object to be detected, but the embodiment is not limited thereto.

The transceiver unit 802 may also be a sensor other than a microwave sensor or use a microwave sensor other than doppler radar technology. For example, the transceiver 802 may also be a microwave device operating in the Ka band 27GHz to 40GHz, or the transceiver 802 may also be a terahertz device, which is not listed here.

For example, the transceiver module 802 is a microwave sensor operating in an FMCW mode, and fig. 9 is a schematic diagram of a transmitting signal operating in the FWCW mode, where the FMCW signal is sawtooth-shaped. As shown in fig. 9, B represents the amount of change (bandwidth of modulation) in the frequency of the transmission signal in one cycle, the frequency varying linearly in one cycle, and the frequency being f at the minimum₀Maximum frequency of f_TAnd T represents the length of the period. For convenience of description, one period is referred to as one chirp, and the second predetermined number (m) of chirps is referred to as one frame (frame).

In the present embodiment, the distance resolution d of the transceiving unit 802_resCan be determined from the bandwidth B of the modulation of the transmitted signal and the speed of light c, i.e. d_resc/2B, corresponding to a velocity resolution of v_res＝λ/2T_f，T_fThe time length of one frame is equal to mT, which is only an example, and the embodiment is not limited thereto.

In this embodiment, the article to be detected is oscillated about the predetermined axis to generate a third velocity vector in the radial direction in a manner as described with reference to fig. 2 in embodiment 1; the radial direction refers to the direction of the object to be detected toward the transceiver unit 802.

In this embodiment, since the article to be detected can swing along with the walking of the person, the article to be detected can also move radially toward the transceiver unit along with the walking of the person in addition to the swing relative to the predetermined axis, that is, the article to be detected can also generate a radial moving speed toward the transceiver unit as a whole in addition to generating moving speeds opposite in the radial direction on both sides of the axis, and the radial moving speed is the same as the walking speed of the person. For convenience of description, the third velocity vector of the object to be detected is the motion velocity of the human reference object.

In this embodiment, the transceiver 802 receives a reflected signal reflected by an object to be detected, the reflected signal and the transmitted signal have a frequency difference, the frequency difference is proportional to a distance between the transceiver and the object to be detected, and the transmitted signal and the reflected signal are processed to obtain a baseband signal. When the article to be detected has a radial velocity relative to the transceiver 802, the frequency of the baseband signal changes, and the changed frequency includes the velocity and the distance information between the article to be detected and the transceiver, and the velocity and the distance information can be obtained by performing two-dimensional fourier transform (2D-FFT), that is, the characteristic signal is obtained. Wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of speed values comprises a second predetermined number of speed values equal to an integral multiple of the radial speed resolution of the transceiver module 802, the predetermined distance being determined according to the distance between the item to be detected and the transceiver module 802.

How the processing unit 803 processes the transmission signal and the reflection signal to obtain a characteristic signal is described below.

Fig. 10 is a schematic diagram of a configuration of the processing unit 803, and as shown in fig. 10, the processing unit 803 includes:

a preprocessing module 1001, configured to perform frequency mixing sampling on the transmit signal and the reflected signal to obtain a first predetermined number of baseband signal matrices, where the transmit signal is a periodic signal; wherein the number of rows of a baseband signal matrix is equal to the second predetermined number, each row representing a baseband signal value of a sampling point of one period; the column number of the baseband signal matrix is equal to a third preset number, and the third preset number is the number of sampling points of the baseband signal in one period;

a transform module 1002, configured to perform fourier transform on each baseband signal matrix row by row to obtain a first signal matrix including the second predetermined number of signal strength values respectively corresponding to the third predetermined number of distances; performing Fourier transform on each first signal matrix column by column to obtain a second signal matrix containing signal intensity values corresponding to the second predetermined number of velocity values respectively corresponding to the third predetermined number of distances;

a selecting module 1004 for selecting, for each second signal matrix, a signal strength value corresponding to a set of speed values corresponding to the predetermined distance from the third predetermined number of distances to obtain the characteristic signal.

In this embodiment, as can be known from the foregoing, the transmission signal and the reflection signal have a frequency difference, and the preprocessing module 1001 performs frequency mixing processing on the transmission signal and the reflection signal with different frequencies, and then converts the signals into a first predetermined number of baseband signal matrixes through the analog-to-digital converter.

The first predetermined number n may be one or at least two, each frame corresponds to a baseband signal matrix, the number of rows of each baseband signal matrix is a second predetermined number, and is equal to the number m of chirp included in one frame, and the number of columns of the baseband signal matrix is a third predetermined number, and is equal to the number k of sampling points of the baseband signal in one period; each row represents the baseband signal value of a sampling point in a chirp, where the values of m, n and k can be determined according to the requirement, which is not limited in this embodiment.

The one baseband signal matrix is as follows:

in this embodiment, the transform module 1002 performs two-dimensional fourier transform on each baseband signal matrix, that is, performs fourier transform on each baseband signal matrix row by row, and then obtains first signal matrices, where each first signal matrix is specifically as follows:

the first signal matrix includes the second predetermined number of signal strength values respectively corresponding to the third predetermined number of distances, that is, includes the third predetermined number multiplied by the second predetermined number of signal strength values, wherein the number of columns of the first signal matrix is the same as the number k of columns of the baseband signal matrix, each column of signal strength values corresponds to one distance, the third predetermined number of distances is equal to an integral multiple of the distance resolution, and the multiple corresponding to each column is equal to the column serial number where the column is located. For example, the distance resolution corresponding to the 1 st column of signal strength values is 1 time, the distance resolution corresponding to the k th column of signal strength values is k times, and the number of rows of the first signal matrix is the same as the number of rows m of the baseband signal matrix.

In this embodiment, since the transformation module 1002 performs the above-mentioned processing on each baseband signal matrix of the first predetermined number of baseband signal matrices, the first predetermined number of first signal matrices are obtained.

In this embodiment, after completing fourier transform row by row, the transform module 1002 performs fourier transform on each first signal matrix column by column to obtain second signal matrices, where each second signal matrix is specifically as follows:

a second signal matrix including signal intensity values corresponding to the second predetermined number of velocity values respectively corresponding to the third predetermined number of distances; the method comprises multiplying a third predetermined number by a second predetermined number of signal intensity values, wherein the number of columns of the second signal matrix is the same as the number of columns k of the baseband signal matrix, each column of signal intensity values corresponds to a distance, the third predetermined number of distances is equal to an integral multiple of the distance resolution, and the multiple corresponding to each column is equal to the column serial number where the column is located. For example, the distance resolution corresponding to the signal strength value of the 1 st column is 1 time, the distance resolution corresponding to the signal strength value of the k th column is k times, the number of rows of the first signal matrix is the same as the number of rows m of the baseband signal matrix, each row of signal strength values corresponds to one speed value, the second predetermined number of speed values is equal to an integral multiple of the speed resolution, the multiple corresponding to each row is equal to the row number where the row is located, for example, the speed resolution corresponding to the signal strength value of the 1 st row is 1 time, and the speed resolution corresponding to the signal strength value of the m th row is m times (third speed vector corresponding to each position of the object to be detected).

In this embodiment, since the transforming module 1002 performs the above-mentioned processing on each of the first signal matrixes of the first predetermined number, the second signal matrixes of the first predetermined number are obtained.

In this embodiment, the specific transformation formula of the two-dimensional fourier transform (e.g., two-dimensional fast fourier transform) may refer to the prior art, and is not described herein again.

In this embodiment, the selecting module 1004 selects, for each second signal matrix, a signal strength value corresponding to a set of speed values corresponding to the predetermined distance from the third predetermined number of distances to obtain the characteristic signal.

How to determine the predetermined distance and how to select a signal strength value corresponding to a set of velocity values corresponding to the predetermined distance are described in detail below.

In this embodiment, the processing unit 803 may further include:

a first determining module 1003 for determining the predetermined distance;

in one embodiment, the first determining module 1003 determines the maximum signal strength value among the signal strength values of each row of the first signal matrix; determining a distance corresponding to the maximum signal strength value from the third predetermined number of distances; the first distance, which occurs the most frequently, is taken as the predetermined distance from among the distances determined from all the rows of the first signal matrix.

In this embodiment, the distance corresponding to the maximum signal strength value is the distance between the article to be detected and the transceiver unit, and since the article to be detected can move and swing along with the walking of the person, the distances corresponding to the maximum signal strength values in each row (corresponding to each chirp) of the first signal matrix are not necessarily the same, and the first determining module 1003 may use the first distance with the largest occurrence frequency as the predetermined distance among the distances determined by all rows of the first signal matrix.

Taking the first signal matrix as an example, X represents the maximum signal strength value, and the first row has the maximum signal strengthDistance corresponding to the value of 2 xd_resThe distance corresponding to the maximum signal strength value of the second row is 2 × d_resThe distance corresponding to the maximum signal strength value in the third row is 2 × d_resThe distance corresponding to the maximum signal intensity value in the fourth row is 3 xd_resTherefore, the first distance with the largest number of occurrences is 2 × d_resThe predetermined distance is equal to 2 xd_resThe selection module 403 selects the second signal matrix corresponding to the first signal matrix with a distance of 2 × d_resAnd obtaining the characteristic signal according to the signal intensity value corresponding to the corresponding set of speed values.

In one embodiment, the first determining module 1003 determines the maximum signal strength value among the signal strength values of the first signal matrix, and uses the second distance corresponding to the maximum signal strength value as the predetermined distance.

In this embodiment, the first determining module 1003 determines the maximum signal strength value not in each row unit, but in all row units of the first signal matrix, and selects the second distance corresponding to the maximum signal strength value as the predetermined distance.

Taking the first signal matrix as an example, X represents the maximum signal intensity value in the first signal matrix, and the distance corresponding to the maximum signal intensity value is 2 × d_resThe predetermined distance is equal to 2 xd_resThe selecting module 1003 selects the second signal matrix corresponding to the first signal matrix with a distance of 2 × d_resAnd obtaining the characteristic signal according to the signal intensity value corresponding to the corresponding set of speed values.

In this embodiment, optionally, after determining the first distance or the second distance, the first determining module 1003 may further select a fourth predetermined number of distances adjacent to the first distance or the second distance from the third predetermined number of distances; and in the second signal matrix, comparing the first distance or the second distance with the sum of the signal intensity values corresponding to each group of speed values corresponding to the fourth predetermined number of distances, and taking the third distance corresponding to the maximum sum of the signal intensity values as the predetermined distance.

For example, the first distance or the second distance determined by the first determining module 1003 is i × d_resThe first determining module 1003 is further configured to determine the value from d_res,2×d_res,....,i×d_res,...,k×d_resIs selected with i × d_resThe adjacent j distances i-j, i-j +1, …, i-1, i, i +1, …, i + j-1, i + j are respectively calculated as (i-j) x d_res，(i-j+1)×d_res…，(i-1)×d_res，i×d_res，(i+1)×d_res，…，(i+j-1)×d_res，(i+j)×d_resThe sum of the signal strength values corresponding to each set of corresponding velocity values, i.e. in the second signal matrix, and (i-j) x d are calculated separately_res，(i-j+1)×d_res…，(i-1)×d_res，i×d_res，(i+1)×d_res，…，(i+j-1)×d_res，(i+j)×d_resThe sum of the corresponding column data, and a third distance corresponding to the largest column, e.g., (i +1) × d_resThe selection module 1004 selects the sum distance as (i +1) × d for the second signal matrix corresponding to the first signal matrix as the predetermined distance_resAnd obtaining the characteristic signal by using the signal intensity value corresponding to the corresponding set of speed values, namely taking the column data of the maximum column as the characteristic signal.

In this embodiment, to avoid the interference and reduce the computational complexity, the first determining module 1003 may further determine the predetermined distance from the screened distances by setting a first threshold to screen the distances, that is, the processing unit 803 may further include: a second determining module (not shown, optional) for determining that less than or equal to the third predetermined number of distancesA distance of a first threshold; e.g. for d_res,2×d_res,...,i×d_res,...,k×d_resSetting the first threshold value to p × d_resThe first determination module 1004 determines from d_res,2×d_res,...,p×d_resThe first determining module determines the predetermined distance from the following first signal matrix and second signal matrix, and the specific method is as described above and is not described herein again.

In this embodiment, when the first predetermined number is 1, a row of signal strength values selected from the second signal matrix is used as the characteristic signal, the row of signal strength values includes a second predetermined number m of velocity values, the m velocity values are respectively equal to integral multiples of the radial velocity resolution, the multiples corresponding to the m velocity values are different and are respectively 1,2, …, m, and the distance corresponding to the row of signal strength values is equal to the predetermined distance.

In this embodiment, when the first predetermined number n is greater than 1, the selecting module 1004 selects a row of signal intensity values from n second signal matrices (each second signal matrix corresponds to a frame) (the selecting manner is the same, as described above, which is not repeated here, and the predetermined distance corresponding to each second signal matrix is the same or different), combines the selected n rows of signal intensity values to form a feature matrix as the feature signal, where the number of rows of the feature matrix is equal to n, the number of rows of the feature matrix is equal to the second predetermined number m, and each row of the feature matrix corresponds to a signal intensity value corresponding to the second predetermined number of speed values selected from one second signal matrix, and the feature matrix is shown as follows:

it should be noted that, for convenience of description, each module in fig. 10 converts a signal into a matrix form for transformation and processing, but this embodiment is not limited to this, and the baseband signal may not need to be sampled and converted into the matrix form, and a two-dimensional fourier transform is directly performed according to a formula (refer to the prior art), and a signal intensity value at a predetermined distance is selected to form a feature signal, which is not described herein again.

In this embodiment, the characteristic signal obtained by performing two-dimensional fourier transform on the baseband signal includes information of a distance, a velocity value, and a signal intensity value corresponding to the distance velocity value, and the processing unit 803 may directly extract a signal whose velocity value and signal intensity value satisfy a predetermined condition from the characteristic signal as the first feature set; in one embodiment, for more intuitive data extraction, the processing unit 803 may convert the feature signal into a micro-doppler map from which the first set of features is extracted.

For example, the horizontal axis of the microporpram represents the frame number, the frame numbers frame 1, frame 2, …, frame n corresponding to the columns of the feature matrix, respectively, and the vertical axis corresponds to the velocity value (corresponding to the third velocity vector) in one frame, wherein the velocity value is based on the v of each row of the feature matrix_res，…，m×v_resOne to one mapping. In addition, the gray scale of each coordinate point determined by the abscissa and ordinate represents the signal intensity value of the signal, i.e., the signal intensity value in the feature matrix. For example, when m is 5, the third row velocity resolution of the feature matrix may be 3 × v_resMapping to a velocity value with ordinate 0, corresponding to a second line velocity resolution of 2 x v_resMapping to a velocity value with ordinate 1 (or-1), the first line velocity resolution v_resMapped to a speed value with ordinate 2 (or-2), and the fourth line speed resolution 4 x v_resMapped to a speed value with ordinate-1 (or 1), with a fifth line speed resolution of 5 x v_resMapping into a speed value with a vertical coordinate of-2 (or 2); for example, for the signal strength value S of the ith row and the jth column of the feature matrix, the value S is added_ijWhen mapping to the micro-doppler plot, the abscissa p corresponds to the column number j (frame number) of the feature matrix, and the ordinate q corresponds to the velocity value corresponding to the i-th row of the feature matrix (for example, i × v is obtained by the above method_resConverted speed value), the gray scale of the coordinate point is the signal intensity valueS_ijThe above is merely an example, and the present embodiment is not limited thereto, and the velocity value in the ordinate is only one relative reference value (corresponding to the third velocity vector).

For example, converting the signal intensity value of the sampling point of which the signal intensity value is greater than or equal to the second threshold value in the micro doppler spectrogram into a first value, and converting the signal intensity value of the sampling point of which the signal intensity value is less than the second threshold value in the micro doppler spectrogram into a second value to obtain a binarized micro doppler spectrogram; removing sampling points with speed values larger than or equal to a third threshold value and smaller than or equal to a fourth threshold value from the binarized micro Doppler spectrogram; taking the signal intensity value or the value obtained by multiplying the speed value by the intensity value of the sampling point as a first feature set, comparing the first feature set with at least two second feature sets in a training set by the determining unit 804, and determining the articles contained in the articles to be detected according to the comparison result; the specific comparison method may also use the existing machine learning (e.g. support vector machine SVM, convolutional neural network CNN) method, which is not described herein again.

The following describes in detail how to obtain at least two second feature sets in the training set.

In this embodiment, the transceiver 802 is used in advance to transmit the same signal for different known articles at the same position as the article to be detected, and receive the reflected signal reflected by the known article, and the processing unit 803 processes the transmitted signal and the reflected signal in the same manner as described above to extract the second feature set.

Since the speed values corresponding to the sampling points in the first feature set extracted actually may be different from the speed values corresponding to the sampling points in the second feature set extracted during training (for example, the speed changes due to the first change and the second change in embodiment 1), if the first feature set is directly compared with the second feature set, the detection result may be inaccurate.

In this embodiment, speed correction apparatus 100 in obtaining unit 801 obtains at least one (n) second speed vector corresponding to at least one set of values before and after the change (n) by performing speed correction on the first speed vector corresponding to the second feature set (which may be obtained from the ordinate in the micro doppler spectrogram, where the first speed vector corresponds to the speed vector of the known item obtained during training, and the third speed quality corresponds to the speed vector of the unknown item obtained during detection) according to the value of the speed after the first change or the value of the maximum normal distance before and after the second change.

It should be noted that the first change and the second change in the present embodiment are not actually changes, but speed correction is performed on the first speed vector corresponding to the sampling point in the second feature set on the assumption that the change has occurred, for example, when the angle of the article relative to the axis is α 1, α 2 … or α n, n second speed vectors corrected according to the angle α 1, α 2 … or α n are obtained according to the above formula (1) (x and y of the article are known), the generation unit 8011 generates n second feature sets respectively according to the n second speed vectors corrected according to the sampling point, and combines the generated n second feature sets with one extracted second feature set as a training set, or, for example, the angular speed after the first change is ω₁Or ω₂… or ω_nWhen, respectively calculate ω₁Or ω₂… or ω_nA first ratio to the angular velocity before the first change; the corrected second velocity vectors are obtained according to the n assumed angular velocities, the generating unit 8011 generates n second feature sets according to the n corrected second velocity vectors of the sampling points, and combines the generated n second feature sets with a previously extracted second feature set to serve as a training set.

By the device of the embodiment, the object hidden on the human body is detected by utilizing the influence of the swing of the object along with the walking of the human body on the reflected signal of the object, the training set is expanded by utilizing the speed correction, the object hidden on the human body is accurately detected, and the detection cost is low.

Example 3

Embodiment 3 provides an article detection apparatus; fig. 11 is a schematic configuration diagram of the article detection apparatus, and as shown in fig. 11, the apparatus 1100 includes: a speed correction device 1101, a transmission/reception unit 1102, a processing unit 1103, an extraction unit 1104, a determination unit 1105 in embodiment 1;

the transceiving unit 1102 sends a transmitting signal to an article to be detected and receives a reflected signal reflected by the article to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction;

the processing unit 1103 is configured to process the transmission signal and the reflection signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit;

the extracting unit 1104 is configured to extract a first feature set from the feature signal;

the speed correction device 1101 performs speed correction on a third speed vector corresponding to the first feature set according to the value of the angular speed before and after the first change or the value of the maximum normal distance before and after the second change, so as to obtain at least one second speed vector corresponding to at least one group of values before and after the change;

the determining unit 1105 is configured to generate at least one first feature set corresponding to the at least one second velocity vector; the first feature set extracted by the extracting unit 1104 and the generated at least one feature set are respectively compared with the feature sets in the training set to obtain at least two detection results, and the at least two detection results are processed to determine the articles contained in the articles to be detected.

In this embodiment, the speed correction device 100 in embodiment 1 may be referred to as an implementation of the speed correction device 1101, the transceiver unit 1102, the processing unit 1103, and the extraction unit 1104 in embodiment 2 may be referred to as a specific implementation of the transceiver unit 802 and the processing unit 803 in embodiment 2, and an obtaining manner of the feature set in the training set is similar to the second feature set extracted in embodiment 2, and is not described again here.

Since the speed value corresponding to the sampling point in the first feature set extracted actually may be different from the speed value corresponding to the sampling point in the feature set extracted during training (for example, the speed changes due to the first change and the second change in embodiment 1), if the first feature set is directly compared with the feature set in the training set, the detection result may be inaccurate. In this embodiment, the first feature set is expanded to obtain at least two first feature sets and at least two detection results, and the at least two detection results are processed to determine the articles contained in the articles to be detected.

In this embodiment, the velocity correction apparatus 100 performs velocity correction on a third velocity vector (which may be obtained from the ordinate of the micro doppler spectrogram and corresponds to the first velocity vector before correction in embodiment 1) corresponding to the first feature set according to the value of the angular velocity before and after the first change or the value of the maximum normal distance before and after the second change, so as to obtain at least one (n) second velocity vector corresponding to at least one set of values before and after the (n) change.

It should be noted that the first change and the second change in the present embodiment are not actually changes, but the third velocity vector corresponding to the sampling point in the first feature set is subjected to velocity correction on the assumption that the change has occurred, and for example, when the article is translated by a distance d1, d2 …, or dn with respect to the axis, the angular velocity ω is calculated according to the above equation (3) (angular velocity ω) in accordance with the above equation (3)₂Known as actually measured, for example, measured according to the human step frequency, etc.), n second velocity vectors corrected according to the distance d1, d2 …, or dn are obtained, the determining unit 1105 respectively generates n first feature sets according to the n second velocity vectors corrected at the sampling point, and compares the generated n first feature sets with a first feature set extracted by the previous extracting unit 1104 respectively to obtain the detection result of each first feature set and the training set, or, for example, obtains the detection result of each first feature set and the training setAssume that the angular velocity after the first change is ω₁Or ω₂… or ω_nWhen, respectively calculate ω₁Or ω₂… or ω_nA first ratio to the angular velocity before the first change; obtaining the corrected second velocity vector according to the n assumed angular velocities, generating n first feature sets by the determining unit 1105 according to the n second velocity vectors (corresponding to the n assumed angular velocity corrections) corrected at the sampling point, respectively, and comparing the generated n first feature sets with a first feature set extracted by the extracting unit 1104 to obtain a detection result of each first feature set and a training set, respectively, where a specific implementation of the velocity correction may refer to fig. 3 or fig. 6 to 7 in embodiment 1, and details of the implementation are not repeated here.

In this embodiment, the determining unit 1105 compares the n first feature sets with the training set respectively to obtain n detection results, counts the occurrence ratio of each detection result, and takes the detection result with the highest occurrence ratio as the final detection result; or when the article detection device is applied to the security inspection field, n detection results are obtained, if at least one detection result in the n detection results is a dangerous article, the detected dangerous article is taken as a final detection result, and the safety is improved.

By the device of the embodiment, the object hidden on the human body is detected by utilizing the influence of the swing of the object along with the walking of the human body on the reflected signal of the object, the test data is expanded by utilizing the speed correction, the object hidden on the human body is accurately detected, and the detection cost is low.

Example 4

Embodiment 4 further provides a speed correction device, fig. 12 is a schematic diagram of a hardware configuration of the speed correction device according to the embodiment of the present invention, and as shown in fig. 12, the speed correction device 1200 may include: an interface (not shown), a Central Processing Unit (CPU)1220, a memory 1210, and a transceiver unit 1240; the memory 1210 is coupled to the central processor 1220. Wherein the memory 1210 may store various data; further, a program for article detection is stored, and the program is executed under the control of the central processor 1220, and various preset values, predetermined conditions, and the like are stored.

In one embodiment, the functions of the speed correction device 1200 may be integrated into the central processor 1220. Wherein the article oscillates about a predetermined axis to produce a first velocity vector in a radial direction; the central processor 1220 may be configured to: correcting the first velocity vector based on the value of the angular velocity before and after the first change, which is a change in angular velocity of the oscillating movement, or the value of the maximum normal distance before and after the second change, which is a change in the maximum normal distance of the article to the predetermined axis due to a change in the relative positions of the article and the predetermined axis, to obtain a second velocity vector.

In one embodiment, the central processor 1220 may be further configured to: calculating a first ratio of the angular velocity after the first change and the angular velocity before the first change; the magnitude of the second velocity vector is equal to the product of the first ratio and the magnitude of the first velocity vector.

In one embodiment, the central processor 1220 may be further configured to: and when the second change is the maximum normal distance change caused by the deflection of the article relative to the axis on the plane of the article, calculating the magnitude of the second velocity vector according to the angle of the deflection and the magnitude of the first velocity vector.

In one embodiment, the central processor 1220 may be further configured to: calculating the maximum normal distance after the second change according to the deflection angle; and calculating a second ratio of the maximum normal distance after the second change to the maximum normal distance before the second change, and taking the product of the second ratio and the size of the first speed vector as the second speed vector.

In one embodiment, the central processor 1220 may be further configured to: calculating the maximum normal distance before the second change according to the first velocity vector and the angular velocity corresponding to the first velocity vector; calculating the magnitude of the second velocity vector according to the distance of the article in translation relative to the axis on the plane where the article is located, the maximum normal distance before the second change and the magnitude of the first velocity vector; the direction of the second velocity vector is determined from the distance of the translation and the direction of the first velocity vector.

In one embodiment, the central processor 1220 may be further configured to: the direction of the second velocity vector determining the position is opposite to the direction of the first velocity vector of the position when the position of the maximum normal distance on the article changes from one side of the axis to the other.

In one embodiment, the central processor 1220 may be further configured to: the maximum normal distance L after the second change is calculated according to the following formula:

The embodiment of the central processing unit 1220 can refer to embodiment 1, and is not repeated here.

In another embodiment, the speed correction device 1200 may be disposed on a chip (not shown) connected to the central processing unit 1220, and the function of the speed correction device may be realized by the control of the central processing unit 1220.

It is noted that the speed correction device 1200 does not necessarily include all of the components shown in fig. 12; further, the speed correction apparatus 1200 may further include components not shown in fig. 12, which can be referred to in the related art.

Through the embodiment, the radial speed generated by the article swinging relative to the preset axis can be corrected, so that the radial speed corresponding to the article placed at any angle or the article at any swinging angular speed can be obtained, the test data or the training data during the article detection can be expanded according to the speed correction, and the article detection precision is improved.

Example 5

Embodiment 5 further provides an article detection apparatus, fig. 13 is a schematic hardware configuration diagram of the article detection apparatus according to the embodiment of the present invention, and as shown in fig. 13, an article detection apparatus 1300 may include: an interface (not shown), a Central Processing Unit (CPU)1320, a memory 1310, and a transceiver unit 1340; memory 1310 is coupled to central processor 1320. Wherein memory 1310 may store various data; further, a program for article detection is stored, and the program is executed under the control of the central processor 1320, and various preset values, predetermined conditions, and the like are stored.

In one embodiment, the functionality of the item detection apparatus 1300 may be integrated into the central processor 1320. Wherein the central processor 1320 may be configured to: obtaining a training set comprising at least two second feature sets in advance; controlling a transmitting and receiving unit to transmit a transmitting signal to an article to be detected and receive a reflected signal reflected by the article to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction; processing the transmitted signal and the reflected signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit; comparing a first feature set extracted from the feature signal with a second feature set in the training set, and determining the articles contained in the articles to be detected according to the comparison result, wherein the first speed vector corresponding to the second feature set is subjected to speed correction according to the value of the angular speed before and after the first change or the value of the maximum normal distance before and after the second change, so as to obtain at least one second speed vector corresponding to at least one group of values before and after the change; generating at least one second feature set corresponding to the at least one second velocity vector, and generating the training set by combining the extracted second feature set with the generated at least one second feature set.

In one embodiment, the functionality of the item detection apparatus 1300 may be integrated into the central processor 1320. Wherein the central processor 1320 may be configured to: controlling the transmitting and receiving unit to transmit a transmitting signal to the object to be detected and receive a reflected signal reflected by the object to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction; processing the transmitted signal and the reflected signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit; extracting a first feature set from the feature signal; performing speed correction on a third speed vector corresponding to the first feature set according to the value of the speed before and after the first change or the value of the maximum normal distance before and after the second change to obtain at least one second speed vector corresponding to at least one group of values before and after the change; generating at least one first feature set corresponding to the at least one second velocity vector; and comparing the first feature set extracted by the extraction unit with the generated at least one feature set respectively with the feature sets in the training set to obtain at least two detection results, and processing the at least two detection results to determine the articles contained in the articles to be detected.

The embodiment of the cpu 1320 can refer to embodiment 2 or 3, and will not be repeated here.

In another embodiment, the article detection apparatus 1300 may be disposed on a chip (not shown) connected to the central processing unit 1320, and the function of the article detection apparatus may be realized by the control of the central processing unit 1320.

It is noted that the item detection apparatus 1300 does not necessarily include all of the components shown in fig. 13; in addition, the article detection apparatus 1300 may further include components not shown in fig. 13, which can be referred to in the prior art.

By the device of the embodiment, the object hidden on the human body is detected by utilizing the influence of the swing of the object on the reflected signal of the object along with the walking of the human body, the training data or the test data are expanded by utilizing the speed correction, the object hidden on the human body is accurately detected, and the detection cost is low.

Example 6

Embodiment 6 of the present invention provides a speed correction method, and since the principle of solving the problem of this method is similar to that of the apparatus of embodiment 1, the specific implementation thereof can refer to the implementation of the apparatus of embodiment 1, and the description thereof is not repeated where the contents are the same.

FIG. 14 is a flow chart of one embodiment of the speed correction method of the present embodiment wherein the articles are oscillated about a predetermined axis to produce a first speed vector in a radial direction; referring to fig. 14, the method includes:

step 1401, the first velocity vector is corrected according to the value of the angular velocity before and after the first change, which is the change of the angular velocity of the swing, or the value of the maximum normal distance before and after the second change, which is the change of the maximum normal distance from the article to the predetermined axis due to the change of the relative position of the article and the predetermined axis, to obtain the second velocity vector.

In one embodiment, in step 1401, a first ratio of the angular velocity after the first change and the angular velocity before the first change is calculated; the magnitude of the second velocity vector is equal to the product of the first ratio and the magnitude of the first velocity vector.

In one embodiment, in step 1401, when the second change is a change in a maximum normal distance due to the deflection of the item relative to the axis in a plane in which the item is located, a magnitude of the second velocity vector is calculated based on the angle of the deflection and a magnitude of the first velocity vector.

For example, the maximum normal distance after the second change (formula (1)) is calculated from the angle of deflection; and calculating a second ratio of the maximum normal distance after the second change to the maximum normal distance before the second change, and taking the product of the second ratio and the size of the first speed vector as the second speed vector.

In one embodiment, in step 1401, the maximum normal distance before the second change is calculated from the first velocity vector and the angular velocity corresponding to the first velocity vector; calculating the magnitude of the second velocity vector according to the distance of the article in translation relative to the axis on the plane where the article is located, the maximum normal distance before the second change and the magnitude of the first velocity vector; determining a direction of the second velocity vector based on the distance of the translation and the direction of the first velocity vector, wherein after the axis translation, the direction of the second velocity vector determining the position is opposite to the direction of the first velocity vector determining the position when the position of the maximum normal distance on the article changes from one side of the axis to the other.

In this embodiment, reference may be made to embodiment 1 for a specific implementation of step 1401, which is not described herein again.

Example 7

Embodiment 7 of the present invention provides an article detection method, and since the principle of solving the problem of this method is similar to that of the apparatus of embodiment 2, the specific implementation thereof can refer to the implementation of the apparatus of embodiment 2, and the description thereof is not repeated where the contents are the same.

Fig. 15 is a flowchart of an embodiment of an article detection method according to the present embodiment, and referring to fig. 15, the method includes:

step 1501, the transceiver unit is further configured to send the transmission signal to a known article, and receive a training reflection signal reflected by the known article; wherein the known article oscillates about a predetermined axis, the known article producing a first velocity vector in a radial direction;

step 1502, performing speed correction on the first speed vector corresponding to the second feature set according to the value of the angular velocity before and after the first change or the value of the maximum normal distance before and after the second change, to obtain at least one second speed vector corresponding to at least one set of values before and after the change;

step 1503, generating at least one second feature set corresponding to the at least one second velocity vector, and generating the training set by combining the extracted second feature set with the generated at least one second feature set;

step 1504, the transceiver unit sends a transmitting signal to the article to be detected and receives a reflected signal reflected by the article to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction;

step 1505, processing the transmitted signal and the reflected signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit;

step 1506, comparing the first feature set extracted from the feature signal with at least two second feature sets in the training set, and determining the articles contained in the articles to be detected according to the comparison result.

In this embodiment, the specific implementation of step 1501-1506 may refer to embodiment 2, which is not described herein again.

By the device of the embodiment, the object hidden on the human body is detected by utilizing the influence of the swing of the object along with the walking of the human body on the reflected signal of the object, the training data is expanded by utilizing the speed correction, the object hidden on the human body is accurately detected, and the detection cost is low.

Example 8

Embodiment 8 of the present invention provides an article detection method, and since the principle of solving the problem of this method is similar to that of the apparatus in embodiment 3, the specific implementation thereof can refer to the implementation of the apparatus in embodiment 3, and the description thereof is not repeated where the contents are the same.

Fig. 16 is a flowchart of an embodiment of an article detection method according to the present embodiment, and referring to fig. 16, the method includes:

step 1601, the transceiver unit sends a transmitting signal to an article to be detected and receives a reflected signal reflected by the article to be detected; wherein the article to be detected swings relative to the predetermined axis, and the article to be detected generates a third velocity vector along the radial direction;

step 1602, processing the transmission signal and the reflection signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal intensity value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of velocity values comprises a second predetermined number of velocity values equal to an integer multiple of the radial velocity resolution of the transceiver unit;

step 1603, extracting a first feature set from the feature signal;

step 1604, performing a speed correction on a third speed vector corresponding to the first feature set according to the value of the angular velocity before and after the first change or the value of the maximum normal distance before and after the second change, to obtain at least one second speed vector corresponding to at least one set of values before and after the change;

step 1605, generating at least one first feature set corresponding to the at least one second velocity vector; and comparing the extracted first feature set with the generated at least one feature set respectively with the feature sets in the training set to obtain at least two detection results, and processing the at least two detection results to determine the articles contained in the articles to be detected.

In this embodiment, reference may be made to embodiment 3 for specific implementation of step 1601-1605, which is not described herein again.

An embodiment of the present invention also provides a computer-readable program, where when the program is executed in an article detection apparatus, the program causes a computer to execute the article detection method in the article detection apparatus as in embodiment 7 or 8 above.

An embodiment of the present invention further provides a storage medium storing a computer-readable program, where the computer-readable program enables a computer to execute the article detection method in the above embodiment 7 or 8 in an article detection apparatus.

An embodiment of the present invention also provides a computer-readable program, wherein when the program is executed in a speed correction apparatus, the program causes a computer to execute the speed correction method as in embodiment 6 above in the speed correction apparatus.

An embodiment of the present invention also provides a storage medium storing a computer-readable program, where the computer-readable program enables a computer to execute the speed correction method in the above embodiment 6 in a speed correction device.

The method of article detection/speed correction in an article detection/speed correction device described in connection with the embodiments of the invention may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. For example, one or more of the functional block diagrams and/or one or more combinations of the functional block diagrams illustrated in FIGS. 1, 3, 5, 7-8, 10-13 may correspond to individual software modules of a computer program flow or may correspond to individual hardware modules. These software modules may correspond to the various steps shown in fig. 14-16, respectively. These hardware modules may be implemented, for example, by solidifying these software modules using a Field Programmable Gate Array (FPGA).

A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. A storage medium may be coupled to the processor such that the processor can read information from, and write information to, the storage medium; or the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The software module may be stored in the memory of the article detection/speed correction device or in a memory card that is insertable into the article detection/speed correction device.

One or more of the functional block diagrams and/or one or more combinations of the functional block diagrams described with respect to fig. 1, 3, 5, 7-8, 10-13 may be implemented as a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any suitable combination thereof, for performing the functions described herein. One or more of the functional block diagrams and/or one or more combinations of the functional block diagrams described with respect to fig. 1, 3, 5, 7-8, 10-13 may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in communication with a DSP, or any other such configuration.

While the invention has been described with reference to specific embodiments, it will be apparent to those skilled in the art that these descriptions are illustrative and not intended to limit the scope of the invention. Various modifications and alterations of this invention will become apparent to those skilled in the art based upon the spirit and principles of this invention, and such modifications and alterations are also within the scope of this invention.

With regard to the embodiments including the above embodiments, the following remarks are also disclosed.

Supplementary note 1, a speed correction method, wherein the article is oscillated with respect to a predetermined axis to generate a first speed vector in a radial direction; the method comprises the following steps:

and correcting the first velocity vector according to the value of the angular velocity before and after the first change or the value of the maximum normal distance before and after the second change to obtain a second velocity vector, wherein the first change is the change of the angular velocity of the swing, and the second change is the change of the maximum normal distance from the article to the predetermined shaft caused by the change of the relative positions of the article and the predetermined shaft.

Supplementary note 2, the method according to supplementary note 1, wherein the correcting the first velocity vector according to the value of the first before-change back angular velocity to obtain the second velocity vector comprises:

calculating a first ratio of the angular velocity after the first change and the angular velocity before the first change; the magnitude of the second velocity vector is equal to the product of the first ratio and the magnitude of the first velocity vector.

Supplementary note 3, the method according to supplementary note 1, wherein the change in the relative position of the article and the predetermined axis comprises a deflection of the article relative to the axis on a plane on which the article is located, and the correcting of the first velocity vector according to the value of the maximum normal distance before and after the second change to obtain the second velocity vector comprises:

and when the second change is the maximum normal distance change caused by the deflection of the article relative to the shaft on the plane of the article, calculating the magnitude of the second velocity vector according to the deflection angle and the magnitude of the first velocity vector.

Note 4, the method according to note 3, wherein calculating the magnitude of the second velocity vector from the angle of deflection and the magnitude of the first velocity vector comprises:

calculating the maximum normal distance after the second change according to the deflection angle;

and calculating a second ratio of the maximum normal distance after the second change to the maximum normal distance before the second change, and taking the product of the second ratio and the size of the first speed vector as the second speed vector.

Supplementary note 5, the method according to supplementary note 1, wherein the changing of the relative position of the article and the predetermined axis comprises a translation of the article on a plane on which the article is located relative to the axis, and the correcting of the first velocity vector according to the value of the maximum normal distance before and after the second change to obtain the second velocity vector comprises:

calculating the maximum normal distance before the second change according to the first speed vector and the angular speed corresponding to the first speed vector;

calculating the size of the second speed vector according to the translation distance of the article on the plane where the article is located relative to the shaft, the maximum normal distance before the second change and the size of the first speed vector;

determining a direction of the second velocity vector based on the distance of the translation and the direction of the first velocity vector.

Supplementary note 6 the method of supplementary note 5, wherein after the axis translation, when the position of the maximum normal distance on the article changes from one side of the axis to the other side, the direction of the second velocity vector determining the position is opposite to the direction of the first velocity vector of the position.

Note 7, the method according to note 4, wherein the maximum normal distance L after the second change is calculated according to the following formula:

wherein α is equal to the angle of deflection, x is equal to the maximum normal distance before the second change, and y is equal to the distance from the location on the article corresponding to the maximum normal distance before the second change to a horizontal axis that is perpendicular to the predetermined axis and passes through the center of mass or center of gravity of the article.

Reference 8 discloses an article detection method, wherein the method comprises:

obtaining a training set comprising at least two second feature sets in advance;

the transmitting and receiving unit transmits a transmitting signal to an article to be detected and receives a reflected signal reflected by the article to be detected; the article to be detected swings relative to the preset shaft, and the article to be detected generates a third speed vector along the radial direction;

processing the emission signal and the reflection signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal strength value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of speed values comprises a second predetermined number of speed values equal to an integer multiple of the radial speed resolution of the transceiver unit;

comparing the first feature set extracted from the feature signal with the second feature set in the training set, and determining the articles contained in the articles to be detected according to the comparison result;

the pre-obtaining of the training set comprising at least two second feature sets comprises:

the transceiver unit is further used for sending the transmitting signal to a known article and receiving a training reflected signal reflected by the known article; wherein a known article oscillates about a predetermined axis, the known article producing a first velocity vector in a radial direction;

processing the emission signal and the training reflection signal to obtain a training characteristic signal; the training characteristic signal comprises a signal intensity value corresponding to a first preset number group of speed values corresponding to a preset distance; each set of speed values comprises a second predetermined number of speed values equal to an integer multiple of the radial speed resolution of the transceiver unit; a second feature set extracted from the training feature signal;

performing speed correction on the first speed vector corresponding to the second feature set according to the value of the first before-after-change speed or the value of the maximum normal distance before and after the second change to obtain at least one second speed vector corresponding to at least one group of values before and after the change;

generating at least one second feature set corresponding to the at least one second velocity vector, and generating the training set by combining the extracted second feature set with the generated at least one second feature set.

Reference 9, a method for detecting an article, wherein the method comprises:

extracting a first feature set from the feature signal;

performing speed correction on a third speed vector corresponding to the first feature set according to the value of the speed before and after the first change or the value of the maximum normal distance before and after the second change to obtain at least one second speed vector corresponding to at least one group of values before and after the change;

generating at least one first feature set corresponding to the at least one second velocity vector; and comparing the extracted first feature set with the generated at least one feature set respectively with the feature sets in the training set to obtain at least two detection results, and processing the at least two detection results to determine the articles contained in the articles to be detected.

Supplementary note 10, the method according to supplementary note 9, wherein the detection result with the high occurrence ratio among the at least two detection results is counted, the detection result with the high occurrence ratio is taken as a final detection result, and the article contained in the article to be detected is determined according to the final detection result.

Claims

1. A speed correction device, wherein an article is oscillated about a predetermined axis to produce a first speed vector in a radial direction; the device comprises:

a correction unit for correcting the first velocity vector according to a value of a first before-change back angular velocity or a value of a second before-change maximum normal distance to obtain a second velocity vector; wherein the first variation is a variation in angular velocity of the oscillation and the second variation is a variation in maximum normal distance of the article to the predetermined axis due to a variation in relative position of the article and the predetermined axis.

2. The apparatus of claim 1, wherein the correction unit comprises:

a first calculation unit for calculating a first ratio of the angular velocity after the first change and the angular velocity before the first change; the magnitude of the second velocity vector is equal to the product of the first ratio and the magnitude of the first velocity vector.

3. The apparatus according to claim 1, wherein the article and predetermined axis relative position variation comprises deflection of the article relative to the axis in a plane in which the article lies, the correction unit comprising:

a second calculating unit for calculating a magnitude of the second velocity vector from an angle of the deflection and a magnitude of the first velocity vector when the second change is a maximum normal distance change due to the deflection of the article relative to the axis on a plane on which the article is located.

4. The apparatus of claim 3, wherein the second computing unit comprises:

a first calculation module for calculating the second changed maximum normal distance from the angle of deflection;

a second calculating module, configured to calculate a second ratio between the maximum normal distance after the second change and the maximum normal distance before the second change, and use a product of the second ratio and the size of the first velocity vector as the second velocity vector.

5. The apparatus according to claim 1, wherein the article and predetermined axis relative position variation comprises article translation relative to the axis in a plane in which the article lies, the correction unit comprising:

a third calculation unit configured to calculate a maximum normal distance before the second change from the first velocity vector and an angular velocity corresponding to the first velocity vector;

a magnitude calculation unit, configured to calculate a magnitude of the second velocity vector according to a distance that the article translates relative to the axis on a plane where the article is located, a maximum normal distance before the second change, and a magnitude of the first velocity vector; and

a direction determination unit for determining a direction of the second velocity vector from the distance of the translation and the direction of the first velocity vector.

6. The apparatus according to claim 5, wherein the direction determination unit determines that the direction of the second velocity vector of the location is opposite to the direction of the first velocity vector of the location when the location of the maximum normal distance on the article changes from one side of the axis to the other side after the article is translated on the plane on which the article is located with respect to the axis.

7. The apparatus of claim 4, wherein the first calculating module calculates the second varied maximum normal distance L according to the formula:

8. An article detection apparatus, wherein the apparatus comprises:

the receiving and transmitting unit is used for sending a transmitting signal to an article to be detected and receiving a reflected signal reflected by the article to be detected; the article to be detected swings relative to the preset shaft, and the article to be detected generates a third speed vector along the radial direction;

a processing unit for processing the transmission signal and the reflection signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal strength value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of speed values comprises a second predetermined number of speed values equal to an integer multiple of the radial speed resolution of the transceiver unit; and

the determining unit is used for comparing a first feature set extracted from the feature signals with a second feature set in the training set, and determining the articles contained in the articles to be detected according to the comparison result;

the transceiver unit is further configured to send the transmission signal to a known article and receive a training reflection signal reflected by the known article; wherein the known article oscillates about a predetermined axis, the known article producing a first velocity vector in a radial direction;

the processing unit is further configured to process the transmission signal and the training reflection signal to obtain a training characteristic signal; the training characteristic signal comprises a signal intensity value corresponding to a first preset number group of speed values corresponding to a preset distance; each set of speed values comprises a second predetermined number of speed values equal to an integer multiple of the radial speed resolution of the transceiver unit; a second feature set extracted from the training feature signal;

the acquisition unit includes: the speed correction device and the generation unit according to any one of claims 1 to 7;

the speed correction device performs speed correction on the first speed vector corresponding to the second feature set according to the value of the speed after the first change or the value of the maximum normal distance before and after the second change to obtain at least one second speed vector corresponding to at least one group of values before and after the change; the generating unit generates at least one second feature set corresponding to the at least one second velocity vector, combines the extracted second feature set with the generated at least one second feature set, and generates the training set.

9. An article detection device, wherein the device comprises a transceiver unit, a processing unit, an extraction unit, a speed correction device according to any one of claims 1 to 7, a determination unit;

the processing unit is used for processing the transmitting signal and the reflected signal to obtain a characteristic signal; wherein the characteristic signal comprises a signal strength value corresponding to a first predetermined number of sets of velocity values corresponding to a predetermined distance; each set of speed values comprises a second predetermined number of speed values equal to an integer multiple of the radial speed resolution of the transceiver unit;

10. The apparatus according to claim 9, wherein the determining unit counts a detection result with a high occurrence ratio among the at least two detection results, takes the detection result with the high occurrence ratio as a final detection result, and determines the article included in the article to be detected according to the final detection result.