JP2021004874A - Article detection method and apparatus - Google Patents

Abstract

To provide an article detection method and an apparatus.SOLUTION: The apparatus comprises: an acquisition unit which acquires test data of an article to be detected, wherein the test data is a micro-Doppler image acquired on the basis of an echo signal by a radar and the echo signal is the signal after reflection by the article to be detected of the signal of radar transmission; a determination unit which determines an effective reflection region in the micro-Doppler image on the basis of the weighted average speed of each frame in the micro-Doppler image; a processing unit which performs processing on the effective reflection region; and a detection unit which performs article detection on the basis of the feature in the micro-Doppler image after the effective reflection region is processed.SELECTED DRAWING: Figure 3

Description

The present invention relates to the field of article detection technology.

In recent years, the issue of safety in public places has been emphasized. For example, how to detect dangerous goods such as regulated products, combustibles, and explosives has become an important issue. So far, hazardous material detectors have been widely applied in densely populated areas such as airports, train stations, subway stations, and gymnasiums. Dangerous goods detection devices can be divided into two types: contact type and non-contact type. The contact type detection device needs to place a suspicious object (for example, a bottle containing a liquid) on the detection device before performing detection, and the non-contact type detection device causes the suspicious object to move within a predetermined range to the detection device. It is possible to start detection and identify whether the suspicious object belongs to dangerous goods.

So far, when it comes to non-contact detectors, one of the most common detection methods is the X-ray detection method. However, the method is usually expensive and can adversely affect the health of workers when used for long periods of time. Also, since the materials of different articles are different, there are differences in their reflection characteristics, and such differences can be used for detecting articles, that is, a radar is provided to send a signal to articles waiting to be detected. The detection can be performed by performing an analysis using an echo signal from an article waiting to be detected and obtaining a micro Doppler image including information such as intensity and radial velocity.

The inventor of the present invention has discovered the following. That is, since dangerous objects are often hidden by pedestrians, the accuracy of detection is affected by the moving speed of pedestrians when detecting articles based on micro-Doppler images obtained by radar. For example, when the moving speed of the pedestrian at the time of actual detection is different from the moving speed at the time of training, the effective reflection area in the micro Doppler image for indicating the classification of the article has a different position and width, so that the detection result May cause inaccuracies.

FIG. 1 is a diagram showing a micro Doppler image acquired at the time of training, and FIG. 2 is a diagram showing a micro Doppler image acquired at the time of actual detection. For the same detected article, as shown in FIG. 1, during training, the pedestrian's movement speed is relatively slow, so that more reflex features can be obtained, and the effective reflex region is relatively low in position and wide. However, as shown in FIG. 2, the moving speed of the pedestrian is relatively fast at the time of actual detection, so that the effective reflection region containing a large amount of effective reflection features has a relatively high position and a comparative width. Narrow. Therefore, if the article is detected by directly comparing the reflection features in FIGS. 1 and 2, an erroneous detection result may be obtained.

An object of the present invention is to accurately find an effective reflection region indicating the classification of an object in a micro-Doppler image and improve the detection accuracy, so that the detection result is affected by the moving speed of a pedestrian carrying an article (limitation). ) Is provided to provide an article detection method and device capable of preventing the reception.

According to the first aspect of the embodiment of the present invention, an article detection device is provided, wherein the device is
An acquisition unit, which is used to obtain test data for an article awaiting detection, the test data is a microdoppler image acquired by a radar based on an echo signal, and the echo signal is a detection of a signal transmitted by the radar. What is the signal after reflection by the waiting item;
A determination unit that determines the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image;
A processing unit that performs processing on the effective reflection region; and a detection unit that detects an article based on a feature in a micro-Doppler image after the effective reflection region has been processed. ..

According to the second aspect of the embodiment of the present invention, an article detection method is provided, wherein the method is described.
The test data of the article waiting to be detected is acquired, and the test data is a microdoppler image acquired based on the echo signal by the radar, and the echo signal is a signal of the signal transmitted by the radar after being reflected by the article waiting to be detected. ;
The effective reflection region in the micro-Doppler image is determined based on the weighted average velocity of each frame in the micro-Doppler image;
Processing the effective reflection region; and including performing article detection based on features in the microDoppler image after the effective reflection region has been processed.

The beneficial effects of the present invention are as follows: that is, based on the weighted average velocity of each frame in the micro-Doppler image, the effective reflection region in the micro-Doppler image is determined and processing is performed on the effective reflection region. By performing the article detection based on the features in the micro Doppler image after the effective reflection region has been processed, the effective reflection region indicating the classification of the object in the micro Doppler image can be accurately found, and the detection accuracy can be obtained. Can be improved so that the detection result is not affected by the moving speed of the pedestrian holding the article.

It is a figure which shows the micro-Doppler image by the training data. It is a figure which shows the micro Doppler image by a test data. It is a figure which shows the article detection apparatus in Example 1 of this invention. It is a figure which shows the transmission signal of a microwave sensor. It is a figure which shows the swing of the article waiting for detection in Example 1 of this invention. It is a figure which shows the calculation of the weighted average velocity in Example 1 of this invention. It is a figure which shows the structure of the determination unit in Example 1 of this invention. It is a figure which shows the structure of the 1st determination unit in Example 1 of this invention. It is a figure which shows the structure of the 2nd determination unit in Example 1 of this invention. It is a figure which shows the structure of the processing unit in Example 1 of this invention. It is a figure which shows the insertion of the frame in Example 1 of this invention. It is a figure which shows the other structure of the processing unit in Example 1 of this invention. It is a figure which shows the hardware composition of the article detection apparatus in Example 2 of this invention. It is a flowchart of the article detection method in Example 3 of this invention.

Hereinafter, suitable examples for carrying out the present invention will be described in detail with reference to the attached drawings.

In order for those skilled in the art to easily understand the principles and embodiments of the present invention, examples of the present invention will be described using a microwave radar as an example, but examples of the present invention will be described in the microwave radar. Not limited.

Example 1 of the present invention provides an article detection device, FIG. 3 is a configuration diagram of the article detection device, and as shown in FIG. 3, the article detection device 300 includes the following.

Acquisition unit 301: Acquires test data of an article awaiting detection, the test data is a microdoppler image acquired by the radar based on an echo signal, and the echo signal is a signal transmitted by the radar reflected by the article awaiting detection. It is a signal after
Confirmation unit 302: Determines the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image;
Processing unit 303: Processing is performed on the effective reflection region;
Detection unit 304: Article detection is performed based on the features in the micro Doppler image after the effective reflection region has been processed.

With the above-mentioned apparatus in the embodiment of the present invention, the effective reflection region in the micro Doppler image is determined based on the weighted average velocity of each frame in the micro Doppler image, the effective reflection region is processed, and the effective reflection region is processed. By performing article detection based on the features in the micro-Doppler image after the effective reflection region has been processed, the effective reflection region that indicates the classification of the object in the micro-Doppler image can be accurately found and the detection accuracy can be improved. Therefore, it is possible to prevent the detection result from being affected by the moving speed of the pedestrian carrying the article.

In an embodiment of the invention, the radar transmits an electromagnetic wave through a transmitting antenna and receives an echo signal after the electromagnetic wave is reflected by a different object. For example, the radar may be a microwave radar operating from 24.05GHz to 24.25GHz, which transmits a microwave (eg, millimeter wave) signal to the object awaiting detection, eg, frequency-modulated continuous wave (Frequency-). Modified Continuous Wave (FMCW) may be transmitted, but the embodiments of the present invention are not limited to this, for example, the radar is a microwave device operating in the Ka band of 27 GHz to 40 GHz. It may be a microwave (for example, millimeter wave) radar working at 77 GHz to 81 GHz, but a comprehensive list is omitted here.

For example, when the radar is a microwave sensor working in the Frequency-modulated Continuous Wave (FMCW) method, FIG. 4 is a diagram showing a transmission signal when working in the FMCW method. The FMCW signal is serrated and, as shown in FIG. 4, B indicates the amount of change (frequency modulation bandwidth) in the frequency of the transmitted signal within one period, and the frequency is linear within one period. The smallest frequency is f ₀ , the largest frequency is f _T , and T indicates the length of the period. Hereinafter, for convenience of explanation, one cycle is referred to as one chirp, and the second predetermined number of chirps is referred to as one frame.

In an embodiment of the invention, the radar range resolution d _res may be determined based on the frequency modulation bandwidth B and the speed of light c of the transmitted signal, i.e. d _res = c / 2B and the corresponding velocity resolution. Is v _res = λ / 2T _f , where T _f indicates the time length of one frame and is equal to mT. It should be noted that this is merely an example, and the examples of the present invention are not limited thereto.

In an embodiment of the present invention, an article awaiting detection is carried by a moving object (eg, a pedestrian), which swings about a predetermined axis (the human walking process also moves left and right along the central axis of the human body). (Swinging back and forth), the moving body produces opposing radial movement speeds on both sides of the axis, which radial direction refers to the direction in which the moving body faces the radar. FIG. 5 is a diagram showing the swing of the moving body. As shown in FIG. 5, when the moving body swings and swings with respect to a predetermined axis, the moving body is along the C axis and the D axis whose radial directions are opposite to the C axis on both sides of the axis, respectively. The swing may be a plurality of reciprocating movements, but the embodiment of the present invention is not limited to this.

In an embodiment of the invention, in addition to the moving body, eg, a human being, being able to generate diametrically opposed moving velocities v1 on both sides of the axis, it also has a radial moving speed v2 facing the radar as a whole. It can also be generated, and its radial movement speed is the same as the walking speed of humans. Hereinafter, for convenience of explanation, the moving speed in the radial direction in the micro Doppler image is the sum (overlap-add method) of v1 and v2.

Hereinafter, how the acquisition unit 301 obtains the test data, that is, a micro Doppler image will be described.

The micro-Doppler effect is a physical phenomenon caused by an object and the resulting tremors, and a micro-Doppler image that reflects the target tremor features can be extracted from the radar echo signal. Specifically, the radar receives the echo signal due to the reflection of the moving object, and there is a frequency difference between the echo signal and the transmitting signal, and the frequency difference is the distance between the radar and the moving object. The baseband signal is directly proportional to the transmitted signal and the reflected signal, and the baseband signal is acquired, when the moving body has opposite moving speeds along the radial direction with respect to the radar. The changing frequency includes the radial movement speed v1 and the distance information between the moving object and the radar, which enables two-dimensional Fourier transform (2D-FFT). , The above-mentioned microdoppler image can be obtained.

In the embodiment of the present invention, frequency mixed sampling is performed on the transmission signal and the echo signal to obtain a first predetermined number (n) of baseband signal matrices, of which the transmission signal is a periodic signal. Among them, the number of rows of one baseband signal matrix is equal to the second predetermined number m, each row shows the baseband signal value of the sampling point of one period, and the number of columns of the baseband signal matrix is It is equal to the third predetermined number k, and the third predetermined number k is the number of sampling points in one period of the baseband signal. A two-dimensional Fourier transform is performed on each baseband signal matrix to obtain a second signal matrix, and the value of each element in the matrix indicates the signal strength. For each second signal matrix (each second signal matrix corresponds to one frame frame), one column may be selected from k columns (for example, the column where the element having the highest signal strength value is located). However, the embodiment of the present invention is not limited to this), and the n columns selected from the n second signal matrices are merged to obtain the feature matrix, and the value of each element in the matrix is a signal. Indicates strength. The feature matrix can be converted into a micro Doppler image.

One baseband signal matrix is as follows.
(Outside 1)

One second signal matrix is as follows.
(Outside 2)

The feature matrix is as follows.
(Outside 3)

1 and 2 are diagrams showing micro Doppler images obtained by conversion. As shown in FIGS. 1 and 2, the abscissa indicates the frame number, which corresponds to the frame numbers frame 1, frame 2, ..., Frame n of each column of the feature matrix, respectively, and the ordinate coordinate is one. Corresponds to radial velocities in the frame, of which the velocities are transformed in a one-to-one mapping based on ν _res , ..., _M × ν _res in each row of the feature matrix, and also. The gray scale of the coordinate points determined by the abscissa and ordinate coordinates represents the signal intensity value of the signal, that is, the signal intensity value in the feature matrix, and the one-to-one mapping method refers to the prior art. Therefore, the detailed description thereof will be omitted here.

In the embodiments of the present invention, the device may further include the following (optional; not shown):

Calculation unit: Calculates the weighted average velocity of each frame in the micro-Doppler image.

In the embodiment of the present invention, the calculation unit obtains the weighted sum of the speed after sampling and the corresponding intensity value of each frame in the micro-Doppler image, and then divides the result by the sum of the corresponding intensity values. The weighted average speed of the frame. Among them, the corresponding intensity value is larger than the first threshold value, and the speed after the sampling is an integral multiple of the radar speed resolution.

For example, for the jth frame of the micro Doppler image, the velocity after sampling is i × ν _res , of which i = 0, 1, ..., K-1, of which K is the micro Doppler image. Select an intensity value that is equal to the number of chirps contained in all frames in, and the intensity value corresponding to the speed after sampling is expressed as E _j (i), and the corresponding intensity value is larger than the first threshold TH1. Calculate the weighted average velocity, for example, calculate c (j) based on the following formula (1), and when c (j) is not an integer, perform truncation and perform the weighted average velocity of the frame j. Can be obtained as c (j) × ν _rex , of which E _j (i)> TH1.

In the embodiment of the present invention, the weighted average velocity of each frame is calculated in order by adopting the above formula (1). FIG. 6 is a diagram showing the calculation of the weighted average velocity. As shown in FIG. 6, for the second frame of the microdoppler image, the speed after sampling is i × ν _res , of which i = 0, 1, ..., 6, of which ν. _{If res} = 1 and the first threshold value TH1 = 0, then c (j) = 3, and the weighted average velocity is 3ν _res .

In the embodiment of the present invention, the determination unit 302 can determine the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame calculated by the calculation unit. The details will be described below.

FIG. 7 is a configuration diagram of the confirmation unit 302, and as shown in FIG. 7, the confirmation unit 302 includes a first confirmation unit 701 and / or a second confirmation unit 702.

The first determination unit 701 determines the start frame and the end frame of the effective reflection region in the micro Doppler image based on the weighted average velocity of each frame in the micro Doppler image.

The second determination unit 702 determines the minimum and maximum velocities of the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image.

FIG. 8 is a configuration diagram of an embodiment of the first determination unit 701, and as shown in FIG. 8, the first determination unit 701 includes the following.

First calculation module 801: Calculates the first average value of the weighted average velocity value above the second threshold value in each frame;
First determination module 802: The frame with the smallest frame number and the frame with the largest frame number whose weighted average velocity value is within the first predetermined range of the first average value are determined, and the frame with the smallest frame number is defined as the start frame. Confirm, and confirm the frame confirmation with the maximum frame number as the end frame.

を計算する。本発明の実施例では、該第一所定範囲は、ニーズに応じて確定されても良く、例えば、該第一所定範囲は、
（外５）

の±所定値THの範囲であっても良く、THの値は、同じであっても良く、異なっても良いが、以下、同じであるケースを例として説明する。該第一平均値の第一所定範囲が
（外６）

であるとし、各フレームの加重平均速度を確定し、加重平均速度が該第一所定範囲内にあるフレーム番号最小のフレーム及びフレーム番号最大のフレームを確定し、そして、フレーム番号最小のフレームを開始フレームと確定し、フレーム番号最大のフレームを終了フレームと確定し、これは、該有効反射領域の左右境界を確定しているに相当する。なお、これは、例示に過ぎず、本発明の実施例は、これに限定されず、例えば、さらに、該マイクロドップラー画像の第一フレームを開始フレームとし、最後の1つのフレームを終了フレームとしても良いが、ここでは、網羅的な列挙を省略する。 In the embodiment of the present invention, the first calculation module 801 calculates the weighted average velocity of all frames in the microdoppler image, then selects the weighted average velocity of the second threshold TH2 or more, and the first of them. Average value (outside 4)

To calculate. In the embodiment of the present invention, the first predetermined range may be determined according to the needs, for example, the first predetermined range may be determined.
(Outside 5)

The value of TH may be in the range of ± predetermined value TH, and the TH value may be the same or different, but the same cases will be described below as an example. The first predetermined range of the first average value is (outside 6)

Then, the weighted average speed of each frame is determined, the frame with the minimum frame number and the frame with the maximum frame number whose weighted average speed is within the first predetermined range are determined, and the frame with the minimum frame number is started. The frame is determined, and the frame having the largest frame number is determined as the end frame, which corresponds to determining the left-right boundary of the effective reflection region. It should be noted that this is merely an example, and the embodiment of the present invention is not limited to this. For example, the first frame of the micro Doppler image may be used as a start frame, and the last frame may be used as an end frame. Good, but I will omit the exhaustive enumeration here.

In the embodiments of the present invention, all reflection features between the start frame and the end frame may be the effective reflection region, optionally further defining the upper and lower boundaries of the effective reflection region, and the start frame and The end frame may also be used to determine the effective reflection region. Hereinafter, how to determine the upper and lower boundaries of the effective reflection region will be described.

FIG. 9 is a configuration diagram of an embodiment of the second deterministic unit 702, and as shown in FIG. 9, the second determinant unit 702 includes the following.

Second determination module 901: Determines the weighted average velocity corresponding to the frame with the smallest weighted average velocity among the frames between the start frame and the end frame;
Third determination module 902: The weighted average speed plus the speed of the fifth threshold value is defined as the maximum speed, and the weighted average speed minus the speed of the sixth threshold value is defined as the minimum speed.

In the embodiment of the present invention, the corresponding minimum weighted average velocity c (j) _min × ν _rez is found from each frame between the start frame and the end frame, and c (j) _min × ν _rez + TH5 is set as the maximum velocity, that is, effective. The lower boundary of the reflection region is set, and c (j) _min × ν _rez −TH6 is set as the minimum velocity, that is, the upper boundary of the effective reflection region, which corresponds to determining the upper and lower boundaries of the effective reflection region. It should be noted that this is merely an example, and the embodiment of the present invention is not limited to this. For example, the minimum speed of the micro-Doppler image may be the upper boundary and the maximum speed may be the lower boundary. Then, the exhaustive enumeration is omitted.

In the embodiment of the present invention, after the effective reflection region is determined, the processing unit 303 processes the effective reflection region so that the detection result is not affected by the moving speed of the pedestrian holding the article.

FIG. 10 is a configuration diagram of an embodiment of the processing unit 303, and as shown in FIG. 10, the processing unit 303 includes the following.

First Judgment Module 1001: Judges whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the third threshold value;
First processing module 1002: When the judgment result of the first judgment module 1001 is no, a frame of the first quantity is inserted between the start frame and the end frame, and the first quantity is the third threshold minus the frame. The difference in numbers.

In the embodiment of the present invention, when the difference in the number of frames between the start frame and the end frame is smaller than the third threshold value TH3, it means that the effective reflection region in the current microdoppler image is relatively narrow, and the detection In order to make the result unaffected by the moving speed of the pedestrian holding the article, the first processing module 1002 inserts a first quantity of frames to increase the width of the effective reflection area. The first quantity can be the difference between the third threshold TH3 minus the number of frames.

In the embodiment of the present invention, the first processing module 1002 can insert the first quantity of frames by the following method. That is, for example, when trying to insert the frame F, for the frame F, the weighted average velocity c (F) × ν _res of the frame F is calculated by using the above formula (1), and the start frame and the start frame and the frame F are calculated. The weighted average velocity c (j) _min × ν _res corresponding to the frame C having the smallest weighted average velocity in each frame between the end frames is determined, and the weighted average velocity of frame F is set to c (j) _{min of} frame C. After aligning with × ν _res , the frame F is inserted into an adjacent position before or after the frame C, and the data of the frame F is cut based on the upper and lower boundaries of the effective reflection region, for example. , Frame F data is cut within the range of (c (j) _min × ν _rez −TH6, c (j) _min × ν _rez + TH5). FIG. 11 is a diagram showing the insertion of a frame. As shown in FIG. 11, the effective reflection region is determined based on the start frame, end frame, maximum velocity, and minimum velocity, and the weighted average velocity of frame F is the weighted average velocity of frame C (minimum weight in the effective reflection region). After aligning with the average velocity), it can be inserted behind the frame C and cut based on the maximum and minimum velocities of the effective reflection region to obtain the processed effective reflection region.

In the embodiment of the present invention, the first quantity of frames may be preset or may be selectable. As shown in FIG. 10, the processing unit 303 may further include the following.

を計算し、計算したフレームjの信号強度値の平均値が

を満たすときに、該フレームjを第一数量のフレームのうちの1つのフレームとして選択する。 Insertion frame selection module 1003: The first quantity of frames is selected from the training collection, and the insertion frame selection module 1003 calculates the average value of the signal strength values equal to or higher than the seventh threshold value in each frame data in the training collection. The calculation is performed, and the frame of the first quantity is selected from the frames where the average value of the eighth threshold value or more and the ninth threshold value or less is located. For example, in the jth frame of the training collection (micro-Doppler image obtained during training), the signal strength values are E _j (i), i = 0, 1, ..., K-1, and E. _{For those with j} (i) ≥ TH7, the average value of the signal strength values E (j) above the seventh threshold TH7 in the frame j (outside 7)

Is calculated, and the average value of the calculated signal strength values of frame j is

When the condition is satisfied, the frame j is selected as one of the frames of the first quantity.

FIG. 12 is a configuration diagram of another embodiment of the processing unit 303, and as shown in FIG. 12, the processing unit 303 includes the following.

Second judgment module 1201: Judges whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the fourth threshold value;
Second processing module 1202: When the judgment result of the second judgment module is yes, the second quantity of frames between the start frame and the end frame is removed, and the second quantity is the difference in the number of frames minus. This is the fourth threshold.

In the embodiment of the present invention, when the difference in the number of frames between the start frame and the end frame is the fourth threshold value TH4 or more, it means that the effective reflection region in the current microdoppler image is relatively wide, and the detection In order to make the result unaffected by the moving speed of the pedestrian holding the article, the second processing module 1202 removes the second quantity of frames to reduce the width of the effective reflection area. The second quantity is the difference in the number of frames minus the fourth threshold.

In the examples of the present invention, the second quantity of frames may be randomly selected, or may be frames in which the weighted average velocities in the effective reflection region overlap. Not limited to this.

In the embodiment of the present invention, the detection unit 304 inputs the microdoppler image after the effective reflection region has been processed into the trained convolutional neural network and outputs the result as a dangerous or safe or article type. Acquires the detection result of the article waiting for detection. The embodiment of the present invention is not limited to this, and detection may be performed using another method, for example, a support vector machine, but a comprehensive enumeration is omitted here.

Further, in the embodiment of the present invention, at least one (p; p ≧ 1) radars may be provided, and at the time of acquiring training data or test data, all the radars have their corresponding micros in the manner described above. The effective reflection area of the Doppler image is acquired, and the above processing is performed on each effective reflection area, and the micro Doppler images corresponding to each radar after the effective reflection area is processed are merged, and then after the merger. The detection can be performed using the Micro Doppler image article of the above, and the merger may be a top-bottom merger or a left-right merger, but the examples of the present invention are not limited thereto.

In an embodiment of the invention, the device further comprises a training unit (not shown; option), which has previously been tested on different articles using at least one radar and of different types of articles. A training collection (eg, a microdoppler image) is extracted, used as the input of the convolutional neural network, the corresponding label (article type, and safety / danger) is used as the output of the convolutional neural network, and a predetermined algorithm. By training the convolutional neural network using the above, the convolutional neural network after training can be obtained. The prior art can be referred to for the specific training process, and detailed description thereof will be omitted here.

Further, in the embodiment of the present invention, the numerical range of each of the above-mentioned threshold values is not limited, and may be set based on experience.

With the above-mentioned apparatus in the embodiment of the present invention, the effective reflection region in the micro Doppler image is determined based on the weighted average velocity of each frame in the micro Doppler image, the effective reflection region is processed, and the effective reflection region is processed. By detecting the feature article in the micro Doppler image after the effective reflection region processing, it is possible to accurately find the effective reflection region that indicates the classification of the object in the micro Doppler image and improve the detection accuracy. The result can be made unaffected by the moving speed of the pedestrian holding the article.

Example 2 of the present invention further provides an article detection device. FIG. 13 is a hardware configuration diagram of the article detection device 1300 according to the embodiment of the present invention. As shown in FIG. 13, the article detector 1300 may include one interface (not shown), a central processing unit (CPU) 1320, and a storage device 1310, which is connected to the central processing unit 1320. Will be done. Among them, the storage device 1310 can store various data, can further store the article detection program, and executes the program under the control of the central processing unit 1320 to execute various predetermined programs. Values, predetermined conditions, etc. can be stored.

In one embodiment, the functionality of the article detector may be integrated into the central processing unit 1320. Among them, the central processing device 1320 may be configured as follows, that is, it is a microdoppler image obtained by acquiring test data of an article waiting to be detected and the test data is acquired by a radar based on an echo signal. The echo signal is a signal of the radar transmission signal after reflection by the object waiting to be detected; the effective reflection region in the microdoppler image is determined based on the weighted average velocity of each frame in the microdoppler image; the effective reflection. The region is processed; and the article is detected based on the features in the microdoppler image after the effective reflection region has been processed.

In one embodiment, the central processing unit 1320 may be further configured as follows, i.e., calculate the weighted average speed of each frame in the microDoppler image.

In one embodiment, the central processing unit 1320 may be further configured as follows, i.e., after determining the weighted sum of the post-sampling speed and corresponding intensity value of each frame in the microDoppler image. , The result of dividing by the sum of the corresponding intensity values is taken as the weighted average velocity of the frame. Among them, the corresponding intensity value is larger than the first threshold value, and the speed after the sampling is an integral multiple of the radar speed resolution.

In one embodiment, the central processing apparatus 1320 may be further configured as follows, i.e., based on the weighted average speed of each frame in the micro-Doppler image, of the effective reflection region in the micro-Doppler image. Determine the start and end frames; and / or determine the minimum and maximum velocities of the effective reflection region in the microDoppler image based on the weighted average velocity of each frame in the microDoppler image.

In one embodiment, the central processing unit 1320 may further be configured as follows, i.e., calculate the first mean of the weighted average velocity values above the second threshold in each frame; the weighted average velocity. The frame with the smallest frame number and the frame with the largest frame number whose values are within the first predetermined range of the first average value are determined, the frame with the smallest frame number is determined as the start frame, and the frame with the largest frame number is the end frame. Is confirmed.

In one embodiment, the central processing unit 1320 may be further configured as follows, i.e., to determine if the difference in the number of frames between the start frame and the end frame is greater than or equal to the third threshold. When the judgment result is no, the first quantity of frames is inserted between the start frame and the end frame, and the first quantity is the difference between the third threshold value and the number of frames; or with the start frame. It is determined whether the difference in the number of frames from the end frame is equal to or greater than the fourth threshold value; when the determination result is yes, the second quantity frame between the start frame and the end frame is removed, and the second quantity is removed. Is the difference in the number of frames minus the fourth threshold value.

In one embodiment, the central processing apparatus 1320 may be further configured as follows, i.e., the weight corresponding to the frame with the lowest weighted average speed of the frames between the start frame and the end frame. The average speed is determined; the weighted average speed plus the speed of the fifth threshold is defined as the maximum speed, and the weighted average speed minus the speed of the sixth threshold is defined as the minimum speed.

In one embodiment, the central processing unit 1320 may further be configured as follows, i.e., select the first quantity of frames from the training collection, of which the training collection is provided by the insertion frame selection module. The average value of the signal strength values of the seventh threshold value or more in each frame data in the above is calculated, and the frame of the first quantity is selected from the frames in which the average value of the eighth threshold value or more and the ninth threshold value or less is located. ..

For a specific example of the central processing unit 1320, the first embodiment can be referred to, and detailed description thereof will be omitted here.

In another embodiment, the above-mentioned article detection device may be provided on a chip (not shown) connected to the central processing unit 1320, and the function of the article detection device may be realized by controlling the central processing unit 1320.

Note that the apparatus 1300 does not have to include all the parts shown in FIG. 13, and may further include, for example, a radar 1340, and the first embodiment of the radar can be referred to. In addition, the device 1300 may further include components not shown in FIG. 13, for which prior art can be referred to.

With the above-mentioned apparatus in the embodiment of the present invention, the effective reflection region in the micro Doppler image is determined based on the weighted average velocity of each frame in the micro Doppler image, the effective reflection region is processed, and the effective reflection region is processed. By performing article detection based on the characteristics of the micro Doppler image after the effective reflection region processing, it is possible to accurately find the effective reflection region that indicates the classification of the object in the Icro Doppler image and improve the detection accuracy. Therefore, the detection result can be made unaffected by the moving speed of the pedestrian holding the article.

Example 3 of the present invention provides a method for detecting an article, and the principle of the method for solving a problem is the same as that of the device of Example 1. Therefore, for specific implementation thereof, the device of Example 1 is implemented. Can be referred to, and duplicate explanations having the same contents are omitted.

FIG. 14 is a flowchart of an article detection method according to an embodiment of the present invention, and as shown in FIG. 14, the method includes the following operations.

Operation 1401: Acquire test data of an article awaiting detection, the test data is a microdoppler image acquired by a radar based on an echo signal, and the echo signal is due to an article awaiting detection of a signal transmitted by the radar. It is a signal after reflection;
Operation 1402: Determine the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image;
Operation 1403: Process the effective reflection area;
Operation 1404: Article detection is performed based on the features in the micro-Doppler image after the effective reflection region has been processed.

In the embodiment of the present invention, the acquisition unit 301, the confirmation unit 302, the processing unit 303, and the detection unit 304 in the first embodiment can be referred to for the specific embodiment of the operation 1401-1404. It will be merged here and duplicate explanations will be omitted.

In an embodiment of the invention, the method may further include the following operation (optional; not shown), i.e., the operation of calculating the weighted average velocity of each frame in the microDoppler image. Of these, the weighted average velocity of the frame is the result of dividing the speed after sampling of each frame and the weighted sum of the corresponding intensity values by the sum of the corresponding intensity values in the micro-Doppler image. Among them, the corresponding intensity value is larger than the first threshold value, and the speed after the sampling is an integral multiple of the radar speed resolution.

In an embodiment of the invention, in operation 1402, the start and end frames of the effective reflection region in the micro-Doppler image are determined based on the weighted average velocity of each frame in the micro-Doppler image; and / or the micro-Doppler. The minimum and maximum velocities of the effective reflection region in the microDoppler image may be determined based on the weighted average velocity of each frame in the image.

In the embodiment of the present invention, the first average value of the weighted average velocity value equal to or higher than the second threshold value in each frame is calculated; the frame number minimum in which the weighted average velocity value is within the first predetermined range of the first average value. The frame and the frame with the maximum frame number are determined, the frame with the minimum frame number is set as the start frame, and the frame with the maximum frame number is set as the end frame.

In the embodiment of the present invention, the weighted average speed corresponding to the frame having the minimum weighted average speed among the frames between the start frame and the end frame is determined; the weighted average speed plus the speed of the fifth threshold is the maximum. The speed is defined as the weighted average speed minus the speed of the sixth threshold value as the minimum speed.

In the embodiment of the present invention, in operation 1403, it is determined whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the third threshold value; when the determination result is no, the start frame and the end frame are determined. Insert a first quantity of frames in between, the first quantity being the third threshold minus the difference in the number of frames; or when the difference in the number of frames between the start frame and the end frame is greater than or equal to the fourth threshold. If the determination result is yes, the second quantity of frames is removed from between the start frame and the end frame, and the second quantity is the difference in the number of frames minus the fourth threshold value.

In the embodiment of the present invention, the first quantity of frames may be selected from the training collection. For example, the average value of the signal strength values equal to or higher than the seventh threshold value in each frame data in the training collection is calculated, and the first value is calculated. The first quantity of frames can be selected from the frames in which the average value of the eighth threshold value or more and the ninth threshold value or less is located.

By the method described above in the examples of the present invention, the effective reflection region in the micro Doppler image is determined based on the weighted average velocity of each frame in the micro Doppler image, the effective reflection region is processed, and the effective reflection region is processed. By performing article detection based on the features in the micro Doppler image after the effective reflection region processing, it is possible to accurately find the effective reflection region that indicates the classification of the object in the micro Doppler image and improve the detection accuracy. , The detection result can be prevented from being restricted by the moving speed of a pedestrian holding an article.

An embodiment of the present invention further provides a computer-readable program, of which, when the program is executed in the article detector, the program tells the computer the article detection in Example 3 in the article detector. Let the method run.

An embodiment of the present invention further provides a storage medium that stores a computer-readable program, of which the computer-readable program causes a computer to perform the article detection method of Example 3 in an article detector.

The methods, devices, and the like described in the examples of the present invention can be realized by hardware, software modules executed by a processor, or a combination of both. For example, one or more functions in the functional block diagram shown in FIGS. 1 to 13 and / or a combination of one or more functions in the functional block diagram may correspond to each software module in the computer program. It may correspond to a hardware module. In addition, each of these software modules can correspond to each step shown in FIG. These hardware modules can be realized by solidifying these software modules using, for example, FPGA (field-programmable gate array).

Further, the apparatus, method, etc. according to the embodiment of the present invention may be realized by software, may be realized by hardware, or may be realized by a combination of hardware and software. The present invention also relates to such a computer-readable program, i.e., when the program is executed by a logic component, the logic component can realize the above-mentioned device or component, or the above-mentioned logic. The component can implement the method described above or a step thereof. Furthermore, the present invention also relates to a storage medium that stores the above-mentioned program, such as a hard disk, a magnetic disk, an optical disk, a DVD, or a fresh memory.

In addition, the following additional notes will be further disclosed with respect to the above-mentioned examples and the like.

(Appendix 1)
It is an article detection device
It is used to obtain test data of an article awaiting detection, and the test data is a microdoppler image acquired by a radar based on an echo signal, and the echo signal is a signal transmitted by the radar after being reflected by the article awaiting detection. Acquisition unit that is a signal;
A determination unit that determines the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image;
An apparatus including a processing unit that performs processing on the effective reflection region; and a detection unit that detects an article based on a feature in a micro-Doppler image after the effective reflection region is processed.

(Appendix 2)
The device described in Appendix 1, and further
A device comprising a computing unit that calculates the weighted average velocity of each frame in the micro-Doppler image.

(Appendix 3)
The device described in Appendix 2
The calculation unit obtains the weighted sum of the speed after sampling and the corresponding intensity value of each frame in the micro-Doppler image, and then divides by the sum of the corresponding intensity values to obtain the weighted average speed of the frame. ,apparatus.

(Appendix 4)
The device described in Appendix 3
The device, wherein the corresponding intensity value is greater than the first threshold and the speed after sampling is an integral multiple of the radar speed resolution.

(Appendix 5)
The device described in Appendix 1
The confirmation unit is
A first determination unit that determines the start and end frames of the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image; and / or the weighted average of each frame in the micro-Doppler image. A device comprising a second determinant unit that determines the minimum and maximum velocities of an effective reflection region in the microDoppler image based on velocity.

(Appendix 6)
The device described in Appendix 5
The first confirmation unit is
The first calculation module that calculates the first average value of the weighted average velocity value above the second threshold value in each frame; and the frame and frame with the smallest frame number whose weighted average velocity value is in the first predetermined range of the first average value. A device that includes a first confirmation module that determines the frame with the largest number, determines the frame with the smallest frame number as the start frame, and determines the frame with the largest frame number as the end frame.

(Appendix 7)
The device described in Appendix 6
The processing unit
The first determination module that determines whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the third threshold value; and when the determination result of the first determination module is no, the start frame and the end frame. A frame of the first quantity is inserted between them, and the first processing module in which the first quantity is the difference between the third threshold value and the number of frames is included.
Alternatively, the processing unit may
A second determination module that determines whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the fourth threshold value; and when the determination result of the second determination module is Yes, the start frame and the end frame. An apparatus comprising a second processing module in which a second quantity of frames is removed from between and the second quantity is the difference in the number of frames minus the fourth threshold.

(Appendix 8)
The device described in Appendix 6
The second confirmation unit is
A second determination module that determines the weighted average speed corresponding to the frame with the smallest weighted average speed among the frames between the start frame and the end frame; and the maximum weighted average speed plus the speed of the fifth threshold. A device comprising a third determination module, wherein the speed is the weighted average speed minus the speed of the sixth threshold, and the minimum speed is the speed.

(Appendix 9)
The device described in Appendix 7
The processing unit further
Includes an insertion frame selection module that selects the first quantity of frames from the training collection.
The insertion frame selection module calculates the average value of the signal strength values above the seventh threshold value in each frame data in the training collection, and starts from the frame where the average value above the eighth threshold value and below the ninth threshold value is located. A device that selects a quantity of frames.

(Appendix 10)
It is an article detection method
The test data of the article waiting to be detected is acquired, and the test data is a microdoppler image acquired based on the echo signal by the radar, and the echo signal is a signal of the signal transmitted by the radar after being reflected by the article waiting to be detected. ;
The effective reflection region in the micro-Doppler image is determined based on the weighted average velocity of each frame in the micro-Doppler image;
A method comprising performing a process on the effective reflection area; and performing article detection based on features in a microDoppler image after the effective reflection area has been processed.

(Appendix 11)
The method described in Appendix 10, and further
A method comprising calculating the weighted average velocity of each frame in the micro-Doppler image.

(Appendix 12)
The method described in Appendix 11
Calculating the weighted average velocity of each frame in the micro-Doppler image
A method comprising obtaining the weighted sum of the speed after sampling and the corresponding intensity value of each frame in the micro-Doppler image and then dividing by the sum of the corresponding intensity values to obtain the weighted average speed of the frame. ..

(Appendix 13)
The method described in Appendix 12
The method, wherein the corresponding intensity value is greater than the first threshold and the speed after sampling is an integral multiple of the radar speed resolution.

(Appendix 14)
The method described in Appendix 10
Determining the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image
Determine the start and end frames of the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image; and / or
A method comprising determining the minimum and maximum velocities of an effective reflection region in the microDoppler image based on the weighted average velocity of each frame in the microDoppler image.

(Appendix 15)
The method described in Appendix 14,
Determining the start and end frames of the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image
Calculate the first average value of the weighted average velocity value above the second threshold value in each frame; and the frame with the smallest frame number and the frame with the largest frame number whose weighted average velocity value is in the first predetermined range of the first average value. A method that includes determining the frame with the lowest frame number as the start frame, and the frame with the highest frame number as the end frame.

(Appendix 16)
The method described in Appendix 15
Performing processing on the effective reflection region
It is determined whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the third threshold value; and when the determination result is no, the first quantity of frames is inserted between the start frame and the end frame. , Including that the first quantity is the difference between the third threshold value and the number of frames.
Or,
It is determined whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the fourth threshold value; and when the determination result is yes, the second quantity of frames is removed from between the start frame and the end frame. , The method comprising: the second quantity being the difference in the number of frames minus the fourth threshold.

(Appendix 17)
The method described in Appendix 15
Determining the minimum and maximum velocities of the effective reflection region in the microDoppler image based on the weighted average velocity of each frame in the microDoppler image
The weighted average speed corresponding to the frame having the smallest weighted average speed among the frames between the start frame and the end frame is determined; and the speed of the weighted average speed plus the fifth threshold is defined as the maximum speed. A method comprising setting the weighted average speed minus the speed of the sixth threshold to the minimum speed.

(Appendix 18)
The method described in Appendix 16
Further processing is performed on the effective reflection region.
Including selecting the first quantity of frames from the training collection
Selecting the first quantity of frames from the training collection
The average value of the signal strength values above the seventh threshold value in each frame data in the training collection is calculated, and the first quantity of frames is selected from the frames where the average value above the eighth threshold value and below the ninth threshold value is located. The method, including that.

Although the preferred embodiment of the present invention has been described above, the present invention is not limited to this embodiment, and any modification to the present invention belongs to the technical scope of the present invention unless the gist of the present invention is deviated.

Claims (10)

It is an article detection device
The test data of the article waiting to be detected is acquired, and the test data is a microdoppler image acquired based on the echo signal by the radar, and the echo signal is a signal of the signal transmitted by the radar after being reflected by the article waiting to be detected. Acquisition unit;
A determination unit that determines the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image;
A device comprising a processing unit that performs processing on the effective reflection region; and a detection unit that detects an article based on features in a micro-Doppler image after the effective reflection region has been processed. The device according to claim 1, further
A device comprising a computing unit that calculates the weighted average velocity of each frame in the micro-Doppler image. The device according to claim 2.
The calculation unit obtains the weighted sum of the speed after sampling and the corresponding intensity value of each frame in the micro-Doppler image, and then divides by the sum of the corresponding intensity values to obtain the weighted average speed of the frame. ,apparatus. The device according to claim 3.
The device, wherein the corresponding intensity value is greater than the first threshold and the speed after sampling is an integral multiple of the radar speed resolution. The device according to claim 1.
The confirmation unit is
A first determination unit that determines the start and end frames of the effective reflection region in the micro-Doppler image based on the weighted average velocity of each frame in the micro-Doppler image; and / or the weighted average of each frame in the micro-Doppler image. A device comprising a second determinant unit that determines the minimum and maximum velocities of an effective reflection region in the microDoppler image based on velocity. The device according to claim 5.
The first confirmation unit is
The first calculation module that calculates the first average value of the weighted average velocity value above the second threshold value in each frame; and the frame and frame with the smallest frame number whose weighted average velocity value is in the first predetermined range of the first average value. A device that includes a first confirmation module that determines the frame with the largest number, determines the frame with the smallest frame number as the start frame, and determines the frame with the largest frame number as the end frame. The device according to claim 6.
The processing unit
The first determination module that determines whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the third threshold value; and when the determination result of the first determination module is no, the start frame and the end frame. A frame of the first quantity is inserted between them, and the first processing module in which the first quantity is the difference between the third threshold value and the number of frames is included.
Alternatively, the processing unit may
A second determination module that determines whether the difference in the number of frames between the start frame and the end frame is equal to or greater than the fourth threshold value; and when the determination result of the second determination module is Yes, the start frame and the end frame. An apparatus comprising a second processing module in which a second quantity of frames is removed from between and the second quantity is the difference in the number of frames minus the fourth threshold. The device according to claim 6.
The second confirmation unit is
A second determination module that determines the weighted average speed corresponding to the frame with the smallest weighted average speed among the frames between the start frame and the end frame; and the maximum weighted average speed plus the speed of the fifth threshold. A device comprising a third determination module, wherein the speed is the weighted average speed minus the speed of the sixth threshold, and the minimum speed is the speed. The device according to claim 7.
The processing unit
It also includes an insertion frame selection module that selects the first quantity of frames from the training collection.
The insertion frame selection module calculates the average value of the signal strength values above the seventh threshold value in each frame data in the training collection, and starts from the frame where the average value above the eighth threshold value and below the ninth threshold value is located. A device that selects a quantity of frames. It is an article detection method
The test data of the article waiting to be detected is acquired, and the test data is a microdoppler image acquired based on the echo signal by the radar, and the echo signal is a signal of the signal transmitted by the radar after being reflected by the article waiting to be detected. ;
The effective reflection region in the micro-Doppler image is determined based on the weighted average velocity of each frame in the micro-Doppler image;
A method comprising performing a process on the effective reflection area; and performing an article detection based on features in a microDoppler image after the effective reflection area has been processed.

Images