A method for detecting the presence of people in a region

Abstract

The invention discloses a method for detecting the presence of personnel in a region, which comprises the following steps: collecting images and radar waves in a region; establishing a dark channel priority image processing strategy under a condition of weak illumination intensity, and establishing a regional illumination imaging model; defining a dark channel in an image by using any pixel point in an image; taking a minimum value of light intensity for an image; establishing a vibration model of a human body; establishing an exponential average adaptive background subtraction equation for radar waves; establishing a radar target enhancement equation; establishing a local energy ratio equation; and determining the coordinates of a target. The method combines images with radar applications, and uses a variety of algorithms to process the collected images and radar data, so as to accurately determine whether there are personnel in a region. Compared with traditional methods, the presence of personnel can be determined regardless of whether the illumination intensity in the region is weak or strong, and whether the personnel in the region are moving or stationary, and the method is simple to understand and highly operable.

Description

Technical Field

[0001] The present invention relates to the field of radar algorithm technology, and in particular to a method for detecting the presence of personnel in an area. Background Art

[0002] Meeting rooms, offices and other places need to turn on the air conditioner when people are present, but once people leave, they tend to forget to turn off the air conditioner, causing power consumption when no one is present. Or when the number of people in the area changes, increases or decreases, the air conditioner cannot adjust itself.

[0003] Patent publication number CN110286627A discloses a method, device, equipment and system for monitoring the presence status of indoor personnel, wherein the method includes: obtaining detection information sent by an intelligent detection device; comparing the detection information with corresponding preset standard information to obtain a comparison result; outputting the detection result of the presence status of indoor personnel according to the comparison result; the detection result includes whether there is someone in the room or no one in the room. When applying the technical solution provided by the embodiment of the present application, if you want to determine whether there is someone in the room, first obtain the detection information through the intelligent detection device, and then compare the detection information with the preset standard information, that is, use the preset standard information to determine what state the current detection information of the intelligent detection device belongs to, and finally determine whether there is someone in the room based on the state to which the current detection information belongs. With such a setting, it is possible to intelligently detect whether there is someone in the room through the intelligent detection device, so as to determine whether it needs to be powered on, which is convenient, fast and practical.

[0004] Authorization number CN107272078B authorizes a method for accurately sensing the presence status of people indoors. The method constructs and starts a perception system and a remote perception control management system. The perception system consists of an infrared sensor probe, a single-chip microcomputer and a pan-tilt motor device. The infrared sensor probe is driven by a pan-tilt motor and fixed in the middle of the upper part of one side in a confined space; wherein the single-chip microcomputer is used to process the receiving signal of the infrared sensor probe, control the motor and provide WebService services; the perception system has a built-in signal processing system, wherein the infrared sensor probe obtains infrared wave signals in real time, and the signal processing system is used to process infrared wave signals to form infrared heat source feedback values; the remote perception control management system is started to manage and control the operation of the perception system in the confined space; the remote perception monitoring system is started to monitor the infrared wave signal data reported by the perception system in real time.

[0005] However, the existing methods cannot effectively monitor people indoors. When the indoor light is dim or people move for a long time, such as when people are sitting quietly or speaking in a meeting, the level range of the sensor feedback changes little, so it cannot accurately sense the presence of the human body. Summary of the invention

[0006] In order to overcome the shortcomings and deficiencies of the prior art, the present invention provides a method for detecting the presence of personnel in an area.

[0007] The technical solution adopted by the present invention is that the method steps include:

[0008] Step S1: collect images and radar waves in the area to establish a data set of images and radar waves;

[0009] Step S2: establishing a dark channel priority processing strategy for images under conditions of weak illumination intensity, and establishing a regional illumination imaging model;

[0010] Step S3: Using any pixel point in the dataset image, define a dark channel in the image;

[0011] Step S4: taking the minimum value of the light intensity of the image, taking the minimum value of the points with low pixel values ​​in the image on the three channels, and completing the noise reduction process of the image;

[0012] Step S5: establishing a vibration model of the human body, wherein the vibration model is composed of a breathing signal, a heartbeat signal, and a noise signal;

[0013] Step S6: Establishing an exponential average adaptive background subtraction equation for radar wave data to remove background data;

[0014] Step S7: establishing a radar target enhancement equation, and performing a cross-correlation operation on the target enhancement equation in the slow time dimension to obtain a 3D matrix;

[0015] Step S8: compress the frequency dimension of the matrix of the radar target enhancement equation to obtain a two-dimensional matrix, and use the performance characteristics of the human body vibration model in the two-dimensional matrix to calculate the local energy ratio to obtain the distance information between the human body and the radar;

[0016] Step S9: using the three-position determination method to locate the coordinates of the person;

[0017] Step S10: Repeat steps S2, S3, S4, S5, S6, S7, S8 and S9 to determine the number of people in the area.

[0018] Furthermore, the dark channel priority processing strategy for establishing an image is expressed as:

[0019]

[0020] Among them, A(n) represents the image after reducing the noise signal, D b represents the background light intensity in the area, a(n) represents the transmittance of the light in the area, and a0 represents the transmittance threshold;

[0021] The regional illumination imaging model is established, and the expression is:

[0022] B b (n) = C b (n)*a(n)+(1-a(n))*D b

[0023] Among them, b represents the color attribute, b belongs to one or a combination of red light, green light, and blue light, B b (n) represents the light intensity at pixel n in the image after imaging, C b (n) represents the reflected light intensity of objects in the area.

[0024] Furthermore, for any pixel in the image, a dark channel in the image can be defined, and the expression is:

[0025]

[0026] Among them, E c (n) represents the dark channel with low light intensity in the image, X, Y, and Z represent red light, green light, and blue light respectively, m represents the pixel in the neighborhood Ω(n) centered on pixel n, and E b (m) indicates a clear image;

[0027] Take the minimum value of the light intensity of the image, the expression is:

[0028]

[0029] Among them, B b (m) represents the light intensity at pixel m in the image after imaging, m represents the pixel in the neighborhood Ω(n) centered at pixel n, E b (m) represents a clear image, a(m) represents the sum of the light transmittance of all objects in the area, and D b Indicates the background light intensity in the area.

[0030] Furthermore, the minimum value of the points with low pixel values ​​in the image is taken on the three channels, and the expression is:

[0031]

[0032] Since the dark channel satisfies:

[0033]

[0034] Without considering the background illumination offset, the brightest pixel in the dark channel can be expressed as:

[0035]

[0036] Transmittance, expressed as:

[0037]

[0038] Among them, B b (m) represents the light intensity at pixel m in the image after imaging, D b represents the background light intensity in the area, a(m) represents the sum of all transmittances of the light in the area, m represents the pixel in the neighborhood Ω(n) centered on pixel n, b represents the color attribute, X represents red, Y represents green, and Z represents blue.

[0039] Furthermore, the vibration model of the human body is expressed as:

[0040] F(i)=F1+f(i)

[0041] Among them, F(i) represents the vibration of the human body, F1 represents the unit distance, and f(i) represents the amplitude of small movement;

[0042] f(i)=f e (i)+f g (i)+h(i)

[0043] Among them, f e (i) represents the human breathing signal, f g (i) represents the human heartbeat signal, h(i) represents the comprehensive noise signal;

[0044]

[0045]

[0046] Among them, f1 represents the vibration amplitude of human breathing, f2 represents the vibration amplitude of human heartbeat, j1 represents the vibration frequency of human breathing, j2 represents the vibration frequency of human heartbeat, Indicates the initial phase of human breathing, Indicates the initial phase of the human heartbeat signal.

[0047] Furthermore, the exponential average adaptive background subtraction method is expressed as:

[0048] H x [y] = θ x [y]×H x-1 [y]+(1-θ x [y])×I x [y]

[0049] Among them, H x [y] represents the current background estimate, H x-1 [y] represents the real background at the previous moment, Ix [y] represents the received radar echo, θ x [y] represents the weight factor.

[0050] Furthermore, the target enhancement expression is:

[0051] J KL (y K ,y L , x′)=P{J K (y K , x n )J L (y L , x m )}

[0052]

[0053] Among them, J K (y K , x n ) represents the radar echo matrix No. 1, J L (y L , x m ) represents the radar echo matrix No. 2, P represents the cross-correlation operation, J KL (y K ,y L , x′) represents the value of the two radar echo matrices after cross-correlation operation, FFT ω represents the Fourier transform operation, Represents the value of the cross-correlation operation of two radar echo matrices in the frequency domain.

[0054] Furthermore, the local energy ratio is expressed as:

[0055] Q v =K r / (KK j )

[0056] Among them, Q v represents the local energy ratio, K r represents the inner window energy, K j represents the warning window energy, and K represents the outer window energy.

[0057] Furthermore, the coordinates of the personnel are expressed as:

[0058]

[0059] Among them, M R Indicates the distance between the human body and radar No. 1, M O Indicates the distance between the human body and radar No. 2, i R The horizontal axis represents the distance between the human body and radar No. 1, jR The vertical coordinate representing the distance between the human body and radar No. 1, i O The horizontal axis represents the distance between the human body and radar No. 2, j O The vertical coordinate represents the distance between the human body and radar No. 2, i represents the horizontal coordinate of the human body position, and j represents the vertical coordinate of the human body position.

[0060] Beneficial effects:

[0061] The invention provides a method for detecting the presence of people in an area. The method combines images with radar applications, and uses multiple algorithms to process the collected images and radar data to accurately determine whether there are people in the area. Compared with traditional methods, this method can determine the presence of people regardless of whether the light intensity in the area is weak or strong, and whether the people in the area are moving or stationary. This method combines the existing image processing algorithm with the radar wave processing algorithm, is easy to understand, and has strong operability. The method is applied to conference rooms, offices and other places to determine the presence of people, which helps to achieve adaptive adjustment of electronic products such as air conditioners or automatic shutdown of equipment, thereby achieving energy saving and consumption reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a general step flow chart of the present invention;

[0063] Figure 2 It is a vibration model diagram of the human body of the present invention. DETAILED DESCRIPTION

[0064] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other. The present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0065] like Figure 1 As shown, a method for detecting the presence of personnel in an area, the method steps include:

[0066] Step S1: collect images and radar waves in the area, and establish a data set of images and radar waves; use RGB-D camera for image collection. RGB-D camera is a widely used visual device. It can not only obtain image information of the environment, but also directly obtain depth information of corresponding points in the environment. Compared with the traditional binocular camera that relies on parallax to obtain depth information, RGB-D camera greatly reduces the complexity of calculation. The collection of radar wave data uses ultra-wideband radar, which has the characteristics of strong penetration, high distance resolution and strong anti-interference, and is widely used in ground exploration, through-wall imaging and post-disaster rescue. UWB radar can be divided into different systems, including carrier-free pulse radar, pseudo-random coding radar, frequency-modulated continuous wave radar and frequency-stepping radar. Carrier-free pulse radar has been widely developed due to its simple structure, low cost and low power loss. UWB radar detects and locates human targets by identifying the Doppler effect generated by micro-movements such as human breathing in the radar echo signal. By selecting the appropriate frequency band, UWB radar can penetrate obstacles and detect human bodies under non-visual conditions.

[0067] Step S2: Establish a dark channel priority processing strategy for images with weak light intensity and establish a regional light imaging model; affected by factors such as transmittance, dust, and light, the image quality in the region is uneven, resulting in low detection rate and accuracy. The dark channel priority image processing strategy uses the dark channel-related conclusions as a priori conditions, and has certain advantages in image enhancement and image restoration, providing a reliable basis for improving the positioning capability and accuracy of personnel detection.

[0068] Step S3: Use any pixel point in the dataset image to define the dark channel in the image; the background light is obtained from the original image through the dark channel, and the brightest top 0.1% pixels are taken according to the brightness in the dark channel image, and the value of the point with the highest brightness at the corresponding position in the original image is found, and the pigment value of this point is taken as the value of the background light.

[0069] Step S4: When there is a person at a certain pixel point in the image, the value of the dark channel will be very small. The minimum value of the light intensity is taken for the image. In order to solve the influence of light rate, dust, illumination, etc. in the area, it is necessary to find the transmittance that best suits the environment in the area. Assuming that the background light is homogeneous and uniform, the transmittance function of the image can be obtained. Then, the minimum value of the pixel value with low value in the image is taken on the three channels to complete the noise reduction processing of the image.

[0070] Step S5: Establish a vibration model of the human body, which is composed of a breathing signal, a heartbeat signal, and a noise signal; during the breathing process of the human body, the chest diaphragm relaxes and contracts to create a certain pressure difference between the inside and outside of the human body, so that oxygen can enter the lungs and accompanied by the ups and downs of the chest cavity, so that the human body produces the required displacement difference. Both breathing and heartbeat signals are low-frequency signals, usually lower than 3Hz. A normal healthy human body (including adults and children) has a heartbeat of about 50-160 times / min and a breathing of about 12-40 times / min in a static state. Since the ups and downs of the human heartbeat and breathing are similar to vibrations, the present invention represents the human life characteristic signal with the amplitude and frequency of sinusoidal vibrations. Assuming that the radar system is fixed to the positioning target and the distance between the antenna and the human body is affected by the ups and downs of the chest cavity, a vibration model of the human body can be established.

[0071] Step S6: Establish an exponential average adaptive background subtraction equation for radar wave data to remove background data; the present invention uses two radars, namely radar 1 and radar 2, to locate human body data. Human breathing signals are usually buried in background echoes such as walls with strong energy. Therefore, background information needs to be removed from the received data. When human body detection is performed based on chest displacement caused by breathing, human echo signals have a similar time delay and are easily mistaken for background information.

[0072] Step S7: Establish a radar target enhancement equation, and perform a cross-correlation operation on the target enhancement equation in the slow time dimension to obtain a three-dimensional matrix; the first dimension in the three-dimensional matrix represents the fast time of radar No. 1, that is, the distance between the target and radar No. 1; the second dimension represents the fast time of radar No. 2, that is, the distance between the target and radar No. 2; since the cross-correlation function between non-periodic noise signals tends to zero, and the amplitude of the homologous periodic breathing signals collected by the two radars will increase after cross-correlation processing, the enhancement of the breathing signal of interest can be achieved. In order to extract the breathing frequency, a fast Fourier transform is performed in the slow time dimension of the three-dimensional matrix. In addition, based on the prior knowledge of the human breathing signal, a rectangular window of 0.1Hz to 0.5Hz is added in the frequency domain to remove other high-frequency interference.

[0073] Step S8: compress the frequency dimension of the matrix of the radar target enhancement equation, and obtain a two-dimensional matrix. By using the performance characteristics of the human body's vibration model in the two-dimensional matrix and calculating the local energy ratio, the distance information between the human body and the radar is obtained; because the position of the person is a two-dimensional information, the three-dimensional matrix must be compressed. The frequency dimension is a description of human breathing and has no effect on the actual position information of the person, so it can be compressed.

[0074] Step S9: The coordinates of the person are located by using the three-position determination method, i.e., the position of radar 1, the position of radar 2, and the position of the person; after processing in step S8, the distance between the human body and radar 1 and the distance between the target and radar 2 can be obtained. Since the positions of the two radars are known, the two-dimensional positioning of the person can be achieved by the three-position determination method.

[0075] Step S10: Repeat steps S2, S3, S4, S5, S6, S7, S8 and S9 to determine the number of people in the area.

[0076] Dark channel priority image processing strategy, the expression is:

[0077]

[0078] Among them, A(n) represents the image after reducing the noise signal, D b represents the background light intensity in the area, a(n) represents the transmittance of the light in the area, and a0 represents the transmittance threshold;

[0079] The regional illumination imaging model is established, and the expression is:

[0080] B b (n) = C b (n)*a(n)+(1-a(n))*D b

[0081] Among them, b represents the color attribute, b belongs to one or a combination of red light, green light, and blue light, B b (n) represents the light intensity at pixel n in the image after imaging, C b (n) represents the reflected light intensity of objects in the area.

[0082] For any pixel in the image, the dark channel in the image can be defined as follows:

[0083]

[0084] Among them, E c (n) represents the dark channel with low light intensity in the image, X, Y, and Z represent red light, green light, and blue light respectively, m represents the pixel in the neighborhood Ω(n) centered on pixel n, and E b (m) indicates a clear image;

[0085] Take the minimum value of the light intensity of the image, the expression is:

[0086]

[0087] Among them, B b(m) represents the light intensity at pixel m in the image after imaging, m represents the pixel in the neighborhood Ω(n) centered at pixel n, E b (m) represents a clear image, a(m) represents the sum of the light transmittance of all objects in the area, and D b Indicates the background light intensity in the area.

[0088] For the points with low pixel values ​​in the image, take the minimum value on the three channels. The expression is:

[0089]

[0090] Since the dark channel satisfies:

[0091]

[0092] Without considering the background illumination offset, the brightest pixel in the dark channel can be expressed as:

[0093]

[0094] Transmittance, expressed as:

[0095]

[0096] Among them, B b (m) represents the light intensity at pixel m in the image after imaging, D b represents the background light intensity in the area, a(m) represents the sum of all transmittances of the light in the area, m represents the pixel in the neighborhood Ω(n) centered on pixel n, b represents the color attribute, X represents red, Y represents green, and Z represents blue.

[0097] The vibration model of the human body is expressed as:

[0098] F(i)=F1+f(i)

[0099] Among them, F(i) represents the vibration of the human body, F1 represents the unit distance, and f(i) represents the amplitude of small movement;

[0100] f(i)=f e (i)+f g (i)+h(i)

[0101] Among them, f e (i) represents the human breathing signal, f g (i) represents the human heartbeat signal, h(i) represents the comprehensive noise signal;

[0102]

[0103]

[0104] Among them, f1 represents the vibration amplitude of human breathing, f2 represents the vibration amplitude of human heartbeat, j1 represents the vibration frequency of human breathing, j2 represents the vibration frequency of human heartbeat, Indicates the initial phase of human breathing, Indicates the initial phase of the human heartbeat signal.

[0105] The adaptive background subtraction of exponential averaging is expressed as:

[0106] H x [y] = θ x [y]×H x-1 [y]+(1-θ x [y])×I x [y]

[0107] Among them, H x [y] represents the current background estimate, H x-1 [y] represents the real background at the previous moment, I x [y] represents the received radar echo, θ x [y] represents the weight factor.

[0108] The target enhancement expression is:

[0109] J KL (y K ,y L , x′)=P{J K (y K , x n )J L (y L , x m )}

[0110]

[0111] Among them, J K (y K , x n ) represents the radar echo matrix No. 1, J L (y L , x m ) represents the radar echo matrix No. 2, P represents the cross-correlation operation, J KL (y K ,y L , x′) represents the value of the two radar echo matrices after cross-correlation operation, FFT ω represents the Fourier transform operation, Represents the value of the cross-correlation operation of two radar echo matrices in the frequency domain.

[0112] The local energy ratio is expressed as:

[0113] Q v =K r / (KK j )

[0114] Among them, Q v represents the local energy ratio, K r represents the inner window energy, K j represents the warning window energy, and K represents the outer window energy.

[0115] The coordinates of the personnel are expressed as:

[0116]

[0117] Among them, M R Indicates the distance between the human body and radar No. 1, M O Indicates the distance between the human body and radar No. 2, i R The horizontal axis represents the distance between the human body and radar No. 1, j R The vertical coordinate representing the distance between the human body and radar No. 1, i O The horizontal axis represents the distance between the human body and radar No. 2, j O The vertical coordinate represents the distance between the human body and radar No. 2, i represents the horizontal coordinate of the human body position, and j represents the vertical coordinate of the human body position.

[0118] The invention provides a method for detecting the presence of people in an area. The method combines images with radar applications, uses multiple algorithms to process the collected images and radar data, and accurately determines whether there are people in the area. Compared with traditional methods, this method can determine the presence of people regardless of whether the light intensity in the area is weak or strong, and whether the people in the area are moving or stationary. This method combines the existing image processing algorithm with the radar wave processing algorithm, is easy to understand, and has strong operability. The method is applied to conference rooms, offices and other places to determine the presence of people, which helps to achieve adaptive adjustment of electronic products such as air conditioners.

[0119] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting the presence of people in an area, characterized in that , the method steps include: Step S1: collect images and radar waves in the area to establish a data set of images and radar waves; Step S2: establishing a dark channel priority processing strategy for images under conditions of weak illumination intensity, and establishing a regional illumination imaging model; Step S3: Using any pixel point in the dataset image, define a dark channel in the image; Step S4: taking the minimum value of the light intensity of the image, taking the minimum value of the points with low pixel values ​​in the image on the three channels, and completing the noise reduction process of the image; Step S5: establishing a vibration model of the human body, wherein the vibration model is composed of a breathing signal, a heartbeat signal, and a noise signal; Step S6: Establishing an exponential average adaptive background subtraction equation for radar wave data to remove background data; Step S7: establishing a radar target enhancement equation, and performing a cross-correlation operation on the target enhancement equation in the slow time dimension to obtain a 3D matrix; Step S8: compress the frequency dimension of the matrix of the radar target enhancement equation to obtain a two-dimensional matrix, and use the performance characteristics of the human body vibration model in the two-dimensional matrix to calculate the local energy ratio to obtain the distance information between the human body and the radar; Step S9: using the three-position determination method to locate the coordinates of the person; Step S10: Repeat steps S2, S3, S4, S5, S6, S7, S8 and S9 to determine the number of people in the area.

2. A method for detecting the presence of personnel in an area as claimed in claim 1, characterized in that: The dark channel priority processing strategy for establishing an image is expressed as: Among them, A(n) represents the image after reducing the noise signal, D b represents the background light intensity in the area, a(n) represents the transmittance of the light in the area, and a0 represents the transmittance threshold; The regional illumination imaging model is established, and the expression is: B b (n)=C b (n)*a(n)+(1-a(n))*D b Among them, b represents the color attribute, b belongs to one or a combination of red light, green light, and blue light, B b (n) represents the light intensity at pixel n in the image after imaging, C b (n) represents the reflected light intensity of objects in the area.

3. A method for detecting the presence of personnel in an area as claimed in claim 1, characterized in that: The dark channel in the image is defined by using any pixel point in the dataset image. The expression is: Among them, E c (n) represents the dark channel with low light intensity in the image, X, Y, and Z represent red light, green light, and blue light respectively, m represents the pixel in the neighborhood Ω(n) centered on pixel n, and E b (m) indicates a clear image; Take the minimum value of the light intensity of the image, the expression is: Among them, B b (m) represents the light intensity at pixel m in the image after imaging, m represents the pixel in the neighborhood Ω(n) centered at pixel n, E b (m) represents a clear image, a(m) represents the sum of the light transmittance of all objects in the area, and D b Indicates the background light intensity in the area.

4. A method for detecting the presence of personnel in an area as claimed in claim 1, characterized in that: The minimum value of the points with low pixel values ​​in the image is taken on the three channels, and the expression is: Since the dark channel satisfies: Without considering the background illumination offset, the brightest pixel in the dark channel can be expressed as: Transmittance, expressed as: Among them, B b (m) represents the light intensity at pixel m in the image after imaging, D b represents the background light intensity in the area, a(m) represents the sum of all transmittances of the light in the area, m represents the pixel in the neighborhood Ω(n) centered on pixel n, b represents the color attribute, X represents red, Y represents green, and Z represents blue.

5. A method for detecting the presence of personnel in an area as claimed in claim 1, characterized in that: The vibration model of the human body is expressed as: F(i)=F1+f(i) Among them, F(i) represents the vibration of the human body, F1 represents the unit distance, and f(i) represents the amplitude of small movement; f(i)=f e (i)+f g (i)+h(i) Among them, f e (i) represents the human breathing signal, f g (i) represents the human heartbeat signal, h(i) represents the comprehensive noise signal; Among them, f1 represents the vibration amplitude of human breathing, f2 represents the vibration amplitude of human heartbeat, j1 represents the vibration frequency of human breathing, j2 represents the vibration frequency of human heartbeat, Indicates the initial phase of human breathing, Indicates the initial phase of the human heartbeat signal.

6. A method for detecting the presence of personnel in an area as claimed in claim 1, characterized in that: The exponential average adaptive background subtraction method is expressed as: H x [y]=θ x [y]×H x-1 [y]+(1-θ x [y])×I x [and] Among them, H x [y] represents the current background estimate, H x-1 [y] represents the real background at the previous moment, I x [y] represents the received radar echo, θ x [y] represents the weight factor.

7. A method for detecting the presence of people in an area as claimed in claim 1, characterized in that: The target enhancement expression is: I KL (y K ,y L ,x′)=P{J K (y K ,x n )J L (y L ,x m )} Among them, J K (y K , x n ) represents the radar echo matrix No. 1, J L (y L , x m ) represents the radar echo matrix No. 2, P represents the cross-correlation operation, J KL (y K ,y L , x′) represents the value of the two radar echo matrices after cross-correlation operation, FFT ω represents the Fourier transform operation, Represents the value of the cross-correlation operation of two radar echo matrices in the frequency domain.

8. A method for detecting the presence of personnel in an area as claimed in claim 1, characterized in that: The local energy ratio is expressed as: Q v =K r / (K-K j ) Among them, Q v represents the local energy ratio, K r represents the inner window energy, K j represents the warning window energy, and K represents the outer window energy.

9. A method for detecting the presence of people in an area as claimed in claim 1, characterized in that: The coordinates of the personnel are expressed as: Among them, M R Indicates the distance between the human body and radar No. 1, M O Indicates the distance between the human body and radar No. 2, i R The horizontal axis represents the distance between the human body and radar No. 1, j R The vertical coordinate representing the distance between the human body and radar No. 1, i O The horizontal axis represents the distance between the human body and radar No. 2, j O The vertical coordinate represents the distance between the human body and radar No. 2, i represents the horizontal coordinate of the human body position, and j represents the vertical coordinate of the human body position.

Images