A millimeter wave radar background environment automatic learning method and device

Abstract

The application discloses a kind of millimeter wave radar background environment automatic learning method and device, through the extraction of intermediate frequency signal amplitude-frequency characteristic, according to its characteristic variation in a certain time to judge, deduce whether the current background environment changes, and then more accurately extract the real information of target, the accurate background environment of millimeter wave radar can be obtained when working, the interference target in background environment is effectively eliminated by background environment, the detection probability of real target is improved, the false alarm probability of radar is reduced, and the reliability of radar is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial control, in particular to a millimeter wave radar background environment automatic learning method and device. BACKGROUND

[0002] The application of millimeter wave radar in the field of intelligent industrial control has become increasingly mature. With the rapid development of industry, millimeter wave radar is widely used in industrial control due to its high precision, small size, light weight, non-contact and other characteristics, and further improves the intelligent process in the industrial field. As the role of millimeter wave radar in the field of industrial control becomes indispensable, the accuracy of its detection results becomes increasingly important. Deviation in detection results may cause property loss or even affect people's health.

[0003] Millimeter wave radar is easily affected by environmental factors such as temperature, humidity, and changes in the environment itself during use. Changes in these background environments will affect the accuracy and reliability of millimeter wave radar. Moreover, millimeter wave radar is applied in various industrial backgrounds. In some complex backgrounds, radar is more likely to be disturbed, and the probability of false positives will also increase. Therefore, there is an urgent need for an algorithm that can learn the environment background to improve the adaptability of the radar to the environment. SUMMARY

[0004] The purpose of the present application is to provide a millimeter wave radar background environment automatic learning method and device for eliminating the influence of environmental factors on millimeter wave radar measurement results and improving the accuracy of millimeter wave radar detection results.

[0005] To achieve the above purpose, the present application provides the following technical scheme:

[0006] The present application discloses a millimeter wave radar background environment automatic learning method, which specifically comprises the following steps:

[0007] S1, the millimeter wave radar samples the background environment to obtain an intermediate frequency sampling signal; the intermediate frequency sampling signal is frequency-decomposed to obtain an amplitude-frequency response curve;

[0008] S2, N amplitude-frequency response curves in a unit sampling duration are recorded to obtain N transient background matrices; the mean of the N transient background matrices is calculated to obtain a transient background mean matrix;

[0009] S3, the difference between the transient background mean matrix and each transient background matrix is calculated to obtain N transient background difference matrices; the maximum value of each transient background difference matrix is calculated to obtain N transient background difference matrix maximum values; the number of transient background difference matrix maximum values greater than a threshold value one is determined;

[0010] S31, if the number of the values greater than the threshold value one is not greater than the interference threshold value, changing the background environment, repeating steps S1-S3 until the number is greater than the interference threshold value, and entering step S4;

[0011] S32, if the number of the values greater than the threshold value one is greater than the interference threshold value, entering step S4;

[0012] S4, changing the background environment, repeating steps S1-S3 until the number is 0; recording the background environment of the current unit sampling duration to obtain a background matrix;

[0013] S5, repeating steps S1-S4 until the number of the cycles reaches a cycle threshold value K, obtaining K background matrices; calculating the mean value of the K background matrices to obtain a background mean value matrix;

[0014] S6, calculating the difference between the background mean value matrix and each background matrix to obtain K background difference value matrices; obtaining the maximum value of each background difference value matrix to obtain K background difference value matrix maximum values; judging the number of the values greater than a threshold value two in the K background difference value matrix maximum values;

[0015] S61, if the number of the values greater than the threshold value two is 0, obtaining a real environment background matrix;

[0016] S62, if the number of the values greater than the threshold value two is greater than 0, repeating steps S1-S6.

[0017] Preferably, the time interval between the N transient background matrices in step S2 is the sampling period of the millimeter wave radar.

[0018] Preferably, the threshold value one and the threshold value two are determined according to the duration of the moving target in the actual application scene.

[0019] Preferably, the value of K is 5-10.

[0020] Preferably, the value of the interference threshold value is N / 5-N / 2.

[0021] The application further discloses a millimeter wave radar background environment automatic learning device, which comprises a memory and one or more processors, the memory stores executable codes, and the one or more processors execute the executable codes to implement the millimeter wave radar background environment automatic learning method.

[0022] The application further discloses a computer readable storage medium, which stores a program, and the program is executed by a processor to implement the millimeter wave radar background environment automatic learning method.

[0023] The application has the following beneficial effects:

[0024] The application discloses a millimeter wave radar background environment automatic learning method and device.

[0025] The features and advantages of the present application will be described in detail with reference to the embodiments combined with the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a flowchart of the algorithm of the present application;

[0027] Figure 2 is a spectrum diagram without background correction;

[0028] Figure 3 is a spectrum diagram with learning background correction;

[0029] Figure 4 is an ambient noise spectrum diagram without background correction;

[0030] Figure 5 is an ambient noise spectrum diagram without background correction at-10 DEG C;

[0031] Figure 6 is an ambient noise spectrum diagram with learning background correction;

[0032] Figure 7 is a structure diagram of the millimeter wave radar background environment automatic learning device. DETAILED DESCRIPTION

[0033] To make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below with reference to the drawings and embodiments. However, it should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the scope of the present application. In addition, in the following description, the description of the known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present application.

[0034] Referring to Figure 1 The present application provides a millimeter wave radar background environment automatic learning method, which is suitable for the scene that the radar and the background environment are relatively static, such as indoor personnel monitoring, parking space monitoring, etc.

[0035] Specifically comprising the following steps:

[0036] S1, the millimeter wave radar samples the background environment to obtain an intermediate frequency sampling signal; the intermediate frequency sampling signal is frequency-decomposed to obtain an amplitude-frequency response curve;

[0037] S2, record N amplitude-frequency response curves in a unit sampling duration to obtain N transient background matrices; calculate the mean of the N transient background matrices to obtain a transient background mean matrix;

[0038] S3, calculate the difference between the transient background mean matrix and each transient background matrix to obtain N transient background difference matrices; find the maximum value of each transient background difference matrix to obtain N transient background difference matrix maximum values; determine the number of transient background difference matrix maximum values greater than a threshold value one;

[0039] S31, if the number of values greater than the threshold value one is not greater than an interference threshold value, change the background environment and repeat steps S1-S3 until the number is greater than the interference threshold value, and enter step S4;

[0040] S32, if the number of values greater than the threshold value one is greater than the interference threshold value, enter step S4;

[0041] S4, change the background environment and repeat steps S1-S3 until the number is 0; record the background environment of the current unit sampling duration to obtain a background matrix;

[0042] S5, repeat steps S1-S4 until the number of cycles reaches a cycle threshold value K to obtain K background matrices; calculate the mean of the K background matrices to obtain a background mean matrix;

[0043] S6, calculate the difference between the background mean matrix and each background matrix to obtain K background difference matrices; find the maximum value of each background difference matrix to obtain K background difference matrix maximum values; determine the number of background difference matrix maximum values greater than a threshold value two;

[0044] S61, if the number of values greater than the threshold value two is 0, obtain a real environment background matrix;

[0045] S62, if the number of values greater than the threshold value two is greater than 0, repeat steps S1-S6.

[0046] In a feasible embodiment, the time interval between the N transient background matrices in step S2 is the sampling period of the millimeter wave radar.

[0047] In a feasible embodiment, the threshold value one and the threshold value two are determined according to the duration of the moving target in the actual application scenario.

[0048] In one feasible embodiment, the value of K is 5 to 10.

[0049] In one feasible embodiment, the value of the interference threshold is determined according to the actual application scenario, and ranges from N / 5 to N / 2.

[0050] Example:

[0051] This example uses an actual millimeter-wave radar product for testing and verification. This radar is specifically designed for indoor personnel monitoring. First, the radar is placed in a cluttered room. The radar begins sampling the background environment of the room. The sampling period t of this radar is 100ms. In this example, the unit sampling duration T is set to 10s, and the number of periods N is 100. This yields 100 sets of amplitude-frequency response curves F1, F2, F3, ..., F 100 .

[0052] Next, the mean value of the amplitude-frequency response curve over 10 seconds is calculated.

[0053] Then calculate the maximum value of 100 sets of background difference matrices.

[0054] (δF-Fi) Max =M δFi (i = 1, 2, ..., 100)

[0055] Get M δFi After (i = 1, 2, ..., 100), determine the number of values ​​greater than the threshold one, using P. num1 This indicates that with a threshold of 50 selected, this instance is the first to obtain P. num1 The value is 5. Based on the application scenario in this example, the interference threshold M is selected as 20, and condition P is not met. num1 >M,(M <N),

[0056] The person then entered the room, walked around within the radar's detection range, and came out, receiving P for the second time. num1 The value is 31, satisfying condition P. num1 >M;

[0057] Next, keep the objects in the room still and do not let anyone enter the room. After 10 seconds, obtain P. num1 Set the value to 0 and record the background matrix δF at this point;

[0058] After repeating the above steps 5 times, we obtain δF1, δF2, ..., δF5. Then, we calculate the mean of the environmental background matrix for each of the 5 iterations. Calculate the maximum value of the background difference matrix after 5 more calculations.

[0059] (σF-δF i ) Max =M σFi(i = 1, 2, …, 5)

[0060] M is obtained σFi (i = 1, 2, …, k), and the number of which is greater than the threshold value two is represented by P num2 , the threshold value two is 100, and the P num2 obtained in this example is 0, and the real environment background σF obtained is recorded;

[0061] The σF is used as a basis to correct the transient background obtained in subsequent sampling, as follows,

[0062] As Figure 2 shown, in the current sampled transient background, there are three higher peaks 417.5, 246.6, and 247.2 at frequency points (17, 33), (17, 38), and (17, 42). At this time, the detection threshold of the radar is limited to 100, and the frequency points (17, 33), (17, 38), and (17, 42) all meet the detection threshold, and if not handled, the radar will output these three targets as effective targets for processing, and misreport that there is a person. After obtaining the environmental background σF using the above algorithm, the current sampled transient background is corrected, and a new set of amplitude-frequency response curves is obtained as shown in Figure 3 , the peaks at frequency points (17, 33), (17, 38), and (17, 42) are filtered out, reducing the risk of false positives.

[0063] As Figure 4 shown, in the current sampled transient background, there are only some small peaks, which are all background noise in the environment, and although they do not currently meet the detection threshold, they will not cause false positives, but as the environmental temperature changes, the intensity of these noises also changes, as shown in Figure 5 , the amplitude-frequency response curve obtained by the radar in the same environmental background when the temperature changes to -10°C, from the figure, it can be seen that as the temperature decreases, multiple frequency points in the background noise exceed the detection threshold, such as the peak value of frequency point (17, 14) is 142.9, at this time, the radar will also consider these noises as effective targets when processing, causing false alarms. After using the algorithm of the present application, the radar will automatically learn the environmental background in real time during the running process from normal temperature to low temperature, and calibrate and correct the existing amplitude-frequency response curve, as shown in Figure 6 , the corrected curve has very low noise, far from reaching the detection threshold, and thus eliminates the false positives caused by the background environment.

[0064] The embodiment of the millimeter wave radar background environment automatic learning device can be applied to any device with data processing capability, which can be a device such as a computer or the like. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a logical device, it is formed by reading the corresponding computer program instructions in the non-volatile memory into the memory for running by the processor of the device with data processing capability. From the hardware level, as shown in Figure 7 The processor, memory, network interface, and non-volatile memory shown in Figure 6 In addition to the processor, memory, network interface, and non-volatile memory shown in

[0065] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts refer to the part of the method embodiment. The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or distributed on multiple network units. According to actual needs, part or all of the modules can be selected to achieve the purpose of the present application. Those skilled in the art can understand and implement without creative labor.

[0066] The embodiment of the present application also provides a computer readable storage medium, which stores a program, and the program is executed by a processor to realize the millimeter wave radar background environment automatic learning device in the above embodiment.

[0067] The computer readable storage medium can be an internal storage unit of any of the aforementioned devices with data processing capability, such as a hard disk or a memory. The computer readable storage medium can also be an external storage device of any of the aforementioned devices with data processing capability, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. Further, the computer readable storage medium can include both an internal storage unit and an external storage device of any of the aforementioned devices with data processing capability. The computer readable storage medium is used to store the computer program and other programs and data required by the aforementioned devices with data processing capability, and can also be used to temporarily store data that has been output or is to be output.

[0068] The above merely provides the preferred embodiments of the application, and is not intended to limit the application. Any modification, equivalent replacement or improvement made within the spirit and principle of the application shall fall within the protection scope of the application.

Claims

1. An automatic learning method for millimeter-wave radar background environment, characterized in that, Specifically, the steps include the following: S1. The millimeter-wave radar samples the background environment to obtain an intermediate frequency (IF) sampling signal; the IF sampling signal is then decomposed to obtain the amplitude-frequency response curve. S2. Record N amplitude-frequency response curves within a unit sampling time to obtain N transient background matrices; Calculate the mean of N transient background matrices to obtain the transient background mean matrix; S3. Calculate the difference between the transient background mean matrix and each transient background matrix to obtain N transient background difference matrices; find the maximum value of each transient background difference matrix to obtain the maximum value of the N transient background difference matrices; determine the number of the N transient background difference matrices whose maximum values ​​are greater than a threshold. S31. If the number of elements greater than threshold 1 is not greater than the interference threshold, change the background environment and repeat steps S1-S3 until the number of elements is greater than the interference threshold, then proceed to step S4. S32. If the number of values ​​greater than threshold 1 is greater than the interference threshold, proceed to step S4. S4. Change the background environment and repeat steps S1-S3 until the count is 0; record the background environment for the current unit sampling time to obtain the background matrix; S5. Repeat steps S1-S4 until the number of iterations reaches the iteration threshold K, resulting in K background matrices; calculate the mean of the K background matrices to obtain the background mean matrix. S6. Calculate the difference between the background mean matrix and each background matrix to obtain K background difference matrices; find the maximum value of each background difference matrix to obtain the maximum value of the K background difference matrices; determine the number of the maximum values ​​of the K background difference matrices that are greater than the threshold value 2. S61. If the number of values ​​greater than the threshold is 0, then the real environment background matrix is ​​obtained. S62. If the number of values ​​greater than threshold 2 is greater than 0, then repeat steps S1-S6.

2. The automatic learning method for millimeter-wave radar background environment as described in claim 1, characterized in that: In step S2, the time interval between the N transient background matrices is the sampling period of the millimeter-wave radar.

3. The automatic learning method for millimeter-wave radar background environment as described in claim 1, characterized in that: The threshold one and threshold two are determined based on the duration of the moving target in the actual application scenario.

4. The automatic learning method for millimeter-wave radar background environment as described in claim 1, characterized in that: The value of K is between 5 and 10.

5. The automatic learning method for millimeter-wave radar background environment as described in claim 1, characterized in that, The interference threshold ranges from N / 5 to N / 2.

6. An automatic learning device for millimeter-wave radar background environment, characterized in that, The device includes a memory and one or more processors, wherein the memory stores executable code, and the one or more processors execute the executable code to implement the automatic learning method for millimeter-wave radar background environment as described in any one of claims 1-5.

7. A computer-readable storage medium, characterized in that: It stores a program that, when executed by a processor, implements the automatic learning method for millimeter-wave radar background environment as described in any one of claims 1-5.

Images