Infrared scattering signal-based muzzle position identification method and system

Abstract

The invention relates to a muzzle position identification method and system based on infrared scattering signals, and belongs to the technical field of muzzle identification. The method comprises the following steps: acquiring a target infrared image and a target active infrared image; calculating to obtain an enhanced target infrared image and an enhanced target active infrared image according to the target infrared image and the target active infrared image, and calculating to obtain an active infrared recognition area according to the enhanced target infrared image and the enhanced target active infrared image; calculating a shooting area center point according to the active infrared identification area, collecting target image data, preprocessing and identifying the target image data to obtain a target lamp identification result, and calculating a target lamp pixel center point; a pixel point distance threshold value is preset, a target lamp pixel center point is determined according to the pixel point distance threshold value, the shooting area center point and the target lamp pixel center point, and the actual space coordinates of the target lamp are calculated according to the target lamp pixel center point. The muzzle position identification is realized through the infrared scattering signal.

Description

Technical Field

[0001] The invention belongs to the technical field of muzzle recognition, and in particular relates to a muzzle position recognition method and system based on infrared scattering signals. Background Art

[0002] In traditional shooting training, laser shooting systems are widely used. This system determines the shooting position by emitting a laser beam to the target and using a receiver to capture the reflected signal of the laser beam. However, laser shooting systems have some inherent limitations and problems: 1. Safety issues: Laser beams may cause damage to eyes in certain situations, especially at high power output or improper use.

[0003] 2. Environmental interference: Laser beams are easily affected in bright light or complex environments (such as haze, dust), resulting in signal loss or misjudgment.

[0004] 3. High environmental requirements: When traditional laser shooting systems are used for shooting training at night, due to insufficient lighting conditions, users cannot determine the target, resulting in insufficient gaming experience. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a muzzle position recognition method and system based on infrared scattering signals. The purpose of the present invention can be achieved through the following technical solutions: The infrared image of the target is collected by the built-in infrared thermal imager of the AR camera, and an active infrared signal is emitted by the infrared transmitter to obtain a trigger pull signal. The intelligent terminal controls the infrared thermal imager to collect the active infrared image of the target according to the trigger pull signal; According to the target infrared image and the target active infrared image, an enhanced target infrared image and an enhanced target active infrared image are calculated by a preprocessing model, and according to the enhanced target infrared image and the enhanced target active infrared image, an active infrared recognition area is calculated by an analysis model; Calculate the center point of the shooting area according to the active infrared recognition area, collect target image data through an AR camera, pre-process the target image data to obtain target pre-processing data, calculate the target light recognition result through a recognition model according to the target pre-processing data, and calculate the target light pixel center point according to the target light recognition result; A pixel point distance threshold is preset, and the pixel center point of the target target light is determined according to the pixel point distance threshold, the center point of the shooting area, and the pixel center point of the target light. The actual spatial coordinates of the target target light are calculated through a coordinate conversion model according to the pixel center point of the target target light. The intelligent terminal generates an adaptive signal according to the actual spatial coordinates of the target target light to control the corresponding target light to turn off.

[0006] Specifically, the preprocessing model includes: Performing filtering processing on the target infrared image and the target active infrared image to obtain a noise-reduced target infrared image and a noise-reduced target active infrared image; The enhanced target infrared image and the enhanced target active infrared image are obtained by performing image enhancement on the denoised target infrared image and the denoised target active infrared image through a Retinex algorithm.

[0007] Specifically, the analysis model includes: According to the enhanced target infrared image and the enhanced target active infrared image, a target infrared image edge and a target active infrared image edge are obtained by wavelet multi-scale detection; Extracting edge feature points of the target infrared image and the target active infrared image by using a canny operator to obtain edge feature points of the target infrared image and edge feature points of the target active infrared image; Obtaining a matching pair by calculating similarity based on the edge feature points of the target infrared image and the edge feature points of the target active infrared image; The active infrared recognition area is obtained by eliminating the matching pairs through a background difference method according to the active infrared image of the target.

[0008] Specifically, the preprocessing includes: A conversion image is obtained by performing color space conversion on the target image data, a correction image is obtained by gray correction on the conversion image, a denoised image is obtained by filtering on the correction image, and the target preprocessing data is obtained by histogram equalization on the denoised image.

[0009] Specifically, the recognition model includes: Performing scale normalization on the target preprocessed data to obtain normalized data, and performing two-dimensional mutation position detection using a Harris corner detector on the normalized data to obtain interest points; According to the interest point, a convex hull is obtained in advance by using the Graham algorithm to obtain a convex hull of the interest point, and a color feature is obtained by calculating the convex hull of the interest point; Performing translation normalization on the convex hull of the point of interest to obtain a translation convex hull, mapping the pixel coordinates of the translation convex hull into polar coordinates to obtain a mapping convex hull, calculating the zernike moment according to the mapping convex hull, and obtaining the reference target light color feature and the reference target light zernike moment; A similarity value is calculated based on the color feature, the zernike moment, the reference target light color feature, and the reference target light zernike moment, a similarity threshold is preset, and the similarity value is screened based on the similarity threshold to obtain the target light recognition result.

[0010] Specifically, the step of determining the pixel center point of the target light according to the pixel distance threshold, the shooting area center point, and the target light pixel center point specifically includes: Calculate the distance data based on the center point of the shooting area and the center point of the target light pixel, and determine whether the distance data is greater than the pixel point distance threshold. If so, count the number of points that meet the requirements and record the coordinates of the points that meet the requirements; if not, do not process; A judgment is made based on the number of points that meet the requirements. If the number of points that meet the requirements is greater than 1, a screening rule is preset, and the coordinates of the points that meet the requirements are screened according to the screening rule to obtain the pixel center point of the target target light; if the number of points that meet the requirements is equal to 1, the coordinates of the points that meet the requirements are used as the pixel center point of the target target light.

[0011] Specifically, the coordinate transformation model includes: The pixel coordinates of the center point of the pixel of the target target light are converted into NDC coordinates, a projection matrix is ​​obtained, the NDC coordinates are converted into clipping space coordinates according to the projection matrix, a view matrix is ​​obtained, and the clipping space coordinates are converted into world coordinates according to the view matrix to obtain the actual space coordinates of the target target light.

[0012] A muzzle position recognition system based on infrared scattering signals, comprising: a data acquisition module, a preprocessing module, a recognition module, and a control module; The data acquisition module is used to collect infrared images of the target through the built-in infrared thermal imager of the AR camera, emit active infrared signals through the infrared transmitter, and obtain a trigger pull signal. The smart terminal controls the infrared thermal imager to collect active infrared images of the target according to the trigger pull signal; The preprocessing module is used to calculate an enhanced target infrared image and an enhanced target active infrared image according to the target infrared image and the target active infrared image through a preprocessing model, and calculate an active infrared recognition area according to the enhanced target infrared image and the enhanced target active infrared image through an analysis model; The recognition module is used to calculate the center point of the shooting area according to the active infrared recognition area, collect target image data through an AR camera, pre-process the target image data to obtain target pre-processing data, calculate the target light recognition result through a recognition model according to the target pre-processing data, and calculate the target light pixel center point according to the target light recognition result; The control module is used to preset a pixel point distance threshold, determine the target target light pixel center point based on the pixel point distance threshold, the shooting area center point, and the target light pixel center point, calculate the target target light pixel center point through a coordinate transformation model to obtain the actual spatial coordinates of the target target light, and the intelligent terminal generates an adaptive signal based on the actual spatial coordinates of the target target light to control the corresponding target light to turn off.

[0013] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the muzzle position recognition method based on infrared scattering signals as described above is implemented.

[0014] A storage medium containing computer executable instructions, which are used to perform the muzzle position recognition method based on infrared scattering signals as described above when executed by a computer processor.

[0015] The beneficial effects of the present invention are: (1) An infrared scattering signal transmitter is provided to emit an infrared scattering signal, image data of a target light is obtained according to the infrared scattering signal, and aiming point recognition is performed according to the image data, thereby solving the problem of being unable to recognize the target point in dark conditions and the insecurity and instability of the laser shooting system.

[0016] (2) By setting up a preprocessing model and an analysis model, the target infrared image and the target active infrared image are cleaned and identified through the preprocessing model and the analysis model to obtain the active infrared recognition area, thereby solving the interference of the external infrared scattering signal to the recognition system.

[0017] (3) By setting up a recognition model, preprocessing is performed according to the active infrared recognition area to obtain target preprocessing data, the target light pixel center point is calculated through the recognition model according to the target preprocessing data, the coordinate conversion is performed according to the target light pixel center point to obtain the actual space coordinate, and the corresponding target light is turned off according to the actual space coordinate, thereby realizing the hit feedback of shooting through infrared scattering signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0019] Figure 1 It is a flow chart of a muzzle position recognition method based on infrared scattering signals of the present invention; Figure 2 It is a schematic diagram of the target device structure in the present invention. DETAILED DESCRIPTION

[0020] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0021] See also Figure 1 ,A muzzle position recognition method based on infrared scattering signals, S1: The infrared image of the target is collected by the built-in infrared thermal imager of the AR camera, and an active infrared signal is emitted by the infrared transmitter to obtain a trigger pull signal. The intelligent terminal controls the infrared thermal imager to collect the active infrared image of the target according to the trigger pull signal; S2: obtaining an enhanced target infrared image and an enhanced target active infrared image by calculating through a preprocessing model according to the target infrared image and the target active infrared image, and obtaining an active infrared recognition area by calculating through an analysis model according to the enhanced target infrared image and the enhanced target active infrared image; S3: Calculate the center point of the shooting area according to the active infrared recognition area, collect target image data through an AR camera, pre-process the target image data to obtain target pre-processing data, calculate the target light recognition result through a recognition model according to the target pre-processing data, and calculate the target light pixel center point according to the target light recognition result; S4: Preset a pixel point distance threshold, determine the target target light pixel center point according to the pixel point distance threshold, the shooting area center point, and the target light pixel center point, calculate the target target light pixel center point through a coordinate conversion model to obtain the actual space coordinates of the target target light, and the smart terminal generates an adaptive signal according to the actual space coordinates of the target target light to control the corresponding target light to turn off.

[0022] In this embodiment, the target is composed of a target light matrix, each target light has an independent number, and when performing shooting training, a target light is set up, and infrared scattering signals are emitted through the muzzle for shooting training.

[0023] Specifically, the preprocessing model includes: Performing filtering processing on the target infrared image and the target active infrared image to obtain a noise-reduced target infrared image and a noise-reduced target active infrared image; It should be noted that the filtering process includes median filtering, mean filtering, and Gaussian filtering; The enhanced target infrared image and the enhanced target active infrared image are obtained by performing image enhancement on the denoised target infrared image and the denoised target active infrared image through a Retinex algorithm.

[0024] Specifically, the analysis model includes: According to the enhanced target infrared image and the enhanced target active infrared image, a target infrared image edge and a target active infrared image edge are obtained by wavelet multi-scale detection; Extracting edge feature points of the target infrared image and the target active infrared image by using a canny operator to obtain edge feature points of the target infrared image and edge feature points of the target active infrared image; Obtaining a matching pair by calculating similarity based on the edge feature points of the target infrared image and the edge feature points of the target active infrared image; The similarity calculation formula is: , in, F is the similarity, f 1 (x, y) is the infrared image of the target, f 2 (x, y) is the active infrared image of the target, a 1 , b 1 The infrared image of the target is p The local mean and variance near the feature point, a 2 , b 2 Active infrared image of the target q The local mean and variance near the feature point, m To match the window size, x, y is a variable parameter, indicating the coordinates; The active infrared recognition area is obtained by eliminating the matching pairs through a background difference method according to the active infrared image of the target.

[0025] Specifically, the calculation formula for the center point of the shooting area is: , in, C x is the horizontal coordinate of the center point of the shooting area, C y is the ordinate of the center point of the shooting area, N is the number of pixels in the shooting area, i is a variable parameter, X i For the i The horizontal coordinate of the pixel point, Y i For the i The vertical coordinate of the pixel.

[0026] It should be noted that the calculation method of the target light pixel center point is the same as the calculation method of the center point of the involved area.

[0027] Specifically, the preprocessing includes: A conversion image is obtained by performing color space conversion on the target image data, a correction image is obtained by gray correction on the conversion image, a denoised image is obtained by filtering on the correction image, and the target preprocessing data is obtained by histogram equalization on the denoised image.

[0028] It should be noted that the gray correction includes gray level correction, grayscale change, and histogram correction, and the filtering process includes median filtering, mean filtering, and Gaussian filtering.

[0029] Specifically, the recognition model includes: Performing scale normalization on the target preprocessed data to obtain normalized data, and performing two-dimensional mutation position detection using a Harris corner detector on the normalized data to obtain interest points; According to the interest point, a convex hull is obtained in advance by using the Graham algorithm to obtain a convex hull of the interest point, and a color feature is obtained by calculating the convex hull of the interest point; The color feature calculation formula is: , Hk is the color feature, numk The convex hull color of the interest point is k The number of pixels, num is the total number of pixels of the convex hull of the interest point, k is a variable parameter, L is the number of color handles; Performing translation normalization on the convex hull of the point of interest to obtain a translation convex hull, mapping the pixel coordinates of the translation convex hull into polar coordinates to obtain a mapping convex hull, calculating the zernike moment according to the mapping convex hull, and obtaining the reference target light color feature and the reference target light zernike moment; The zernike moment calculation formula is: , in, Vnm(x, y) for zernike Moment, x, y is the pixel coordinate, Rnm(ρ) is a radial polynomial, j is an imaginary unit, m For repeatability, n is the order, ρ, θ For point (x, y) Polar coordinate representation of ; A similarity value is calculated based on the color feature, the zernike moment, the color feature of the reference target light, and the zernike moment of the reference target light, a similarity threshold is preset, and the similarity value is screened based on the similarity threshold to obtain the target light recognition result; The similarity value calculation formula is: , in, S is the similarity value, W c 、W z is the weight value, k is a variable parameter, L is the number of color handles, n is the number of zernike moments, H k （Q） is the color feature, H k （I） For reference target light color characteristics, Z k （Q） For the zernike moment, Z k （I） The zernike moment of the reference target light.

[0030] Specifically, the step of determining the pixel center point of the target light according to the pixel distance threshold, the shooting area center point, and the target light pixel center point specifically includes: Calculate the distance data based on the center point of the shooting area and the center point of the target light pixel, and determine whether the distance data is greater than the pixel point distance threshold. If so, count the number of points that meet the requirements and record the coordinates of the points that meet the requirements; if not, do not process; A judgment is made based on the number of points that meet the requirements. If the number of points that meet the requirements is greater than 1, a screening rule is preset, and the coordinates of the points that meet the requirements are screened according to the screening rule to obtain the pixel center point of the target target light; if the number of points that meet the requirements is equal to 1, the coordinates of the points that meet the requirements are used as the pixel center point of the target target light.

[0031] In this embodiment, the screening rule is to traverse the coordinates of the points that meet the requirements in the order of top, right, bottom, and left in the camera coordinate system. When the results are retrieved, the traversal is no longer performed and the retrieval results are output.

[0032] Specifically, the coordinate transformation model includes: The pixel coordinates of the center point of the pixel of the target target light are converted into NDC coordinates, a projection matrix is ​​obtained, the NDC coordinates are converted into clipping space coordinates according to the projection matrix, a view matrix is ​​obtained, and the clipping space coordinates are converted into world coordinates according to the view matrix to obtain the actual space coordinates of the target target light; The NDC coordinate calculation formula is: , in, Xndc , Yndc is the NDC coordinate, Xp , Yp is the pixel coordinate, width is the viewport width, height is the viewport height; The clipping space coordinate calculation formula is: , in, Xc, Yc, Zc, Wc is the clipping space coordinate, p is the projection matrix, Xndc, Yndc is the NDC coordinate; The world coordinate calculation formula is: , in, Xa, Ya, Za, Wa is the world coordinate, M, V is the view matrix, Xc, Yc, Zc, Wc is the clip space coordinate.

[0033] A muzzle position recognition system based on infrared scattering signals, comprising: a data acquisition module, a preprocessing module, a recognition module, and a control module; The data acquisition module is used to collect infrared images of the target through the built-in infrared thermal imager of the AR camera, emit active infrared signals through the infrared transmitter, and obtain a trigger pull signal. The smart terminal controls the infrared thermal imager to collect active infrared images of the target according to the trigger pull signal; The preprocessing module is used to calculate an enhanced target infrared image and an enhanced target active infrared image according to the target infrared image and the target active infrared image through a preprocessing model, and calculate an active infrared recognition area according to the enhanced target infrared image and the enhanced target active infrared image through an analysis model; The recognition module is used to calculate the center point of the shooting area according to the active infrared recognition area, collect target image data through an AR camera, pre-process the target image data to obtain target pre-processing data, calculate the target light recognition result through a recognition model according to the target pre-processing data, and calculate the target light pixel center point according to the target light recognition result; The control module is used to preset a pixel point distance threshold, determine the target target light pixel center point based on the pixel point distance threshold, the shooting area center point, and the target light pixel center point, calculate the target target light pixel center point through a coordinate transformation model to obtain the actual spatial coordinates of the target target light, and the intelligent terminal generates an adaptive signal based on the actual spatial coordinates of the target target light to control the corresponding target light to turn off.

[0034] In this embodiment, the steps for implementing the shooting game according to the muzzle position recognition method based on infrared scattering signals are as follows: Step 1: Emit infrared scattered signals through the muzzle, and obtain the target infrared image and target active infrared image through the AR camera carried on the gun; Step 2: The AR camera transmits the collected target infrared image and target active infrared image to the smart terminal, and the smart terminal calculates the aiming point according to steps S2-S3 and visualizes the aiming point; Step 3: The user triggers the trigger, and the intelligent terminal collects the last frame of the target active infrared image after receiving the trigger signal, and calculates the target light of the last shot according to steps S2-S4; Step 4: The intelligent terminal generates an adaptive signal according to the actual spatial coordinates of the target light to control the corresponding target light to turn off, thereby achieving shooting feedback.

[0035] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0036] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0037] The program code included on the computer readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0038] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A muzzle position recognition method based on infrared scattering signals, characterized in that: include: The infrared image of the target is collected by the built-in infrared thermal imager of the AR camera, and an active infrared signal is emitted by the infrared transmitter to obtain a trigger pull signal. The intelligent terminal controls the infrared thermal imager to collect the active infrared image of the target according to the trigger pull signal; According to the target infrared image and the target active infrared image, an enhanced target infrared image and an enhanced target active infrared image are calculated by a preprocessing model, and according to the enhanced target infrared image and the enhanced target active infrared image, an active infrared recognition area is calculated by an analysis model; Calculate the center point of the shooting area according to the active infrared recognition area, collect target image data through an AR camera, pre-process the target image data to obtain target pre-processing data, calculate the target light recognition result through a recognition model according to the target pre-processing data, and calculate the target light pixel center point according to the target light recognition result; A pixel point distance threshold is preset, and the pixel center point of the target target light is determined according to the pixel point distance threshold, the center point of the shooting area, and the pixel center point of the target light. The actual spatial coordinates of the target target light are calculated through a coordinate conversion model according to the pixel center point of the target target light. The intelligent terminal generates an adaptive signal according to the actual spatial coordinates of the target target light to control the corresponding target light to turn off.

2. The muzzle position recognition method based on infrared scattering signals according to claim 1 is characterized in that: The preprocessing model includes: Performing filtering processing on the target infrared image and the target active infrared image to obtain a noise-reduced target infrared image and a noise-reduced target active infrared image; The enhanced target infrared image and the enhanced target active infrared image are obtained by performing image enhancement on the denoised target infrared image and the denoised target active infrared image through a Retinex algorithm.

3. The muzzle position recognition method based on infrared scattering signals according to claim 1 is characterized in that: The analysis model includes: According to the enhanced target infrared image and the enhanced target active infrared image, a target infrared image edge and a target active infrared image edge are obtained by wavelet multi-scale detection; Extracting edge feature points of the target infrared image and the target active infrared image by using a canny operator to obtain edge feature points of the target infrared image and edge feature points of the target active infrared image; Obtaining a matching pair by calculating similarity based on the edge feature points of the target infrared image and the edge feature points of the target active infrared image; The active infrared recognition area is obtained by eliminating the matching pairs through a background difference method according to the active infrared image of the target.

4. The muzzle position recognition method based on infrared scattering signals according to claim 1 is characterized in that: The pre-processing comprises: A conversion image is obtained by performing color space conversion on the target image data, a correction image is obtained by gray correction on the conversion image, a denoised image is obtained by filtering on the correction image, and the target preprocessing data is obtained by histogram equalization on the denoised image.

5. The muzzle position recognition method based on infrared scattering signals according to claim 1 is characterized in that: The recognition model includes: Performing scale normalization on the target preprocessed data to obtain normalized data, and performing two-dimensional mutation position detection using a Harris corner detector on the normalized data to obtain interest points; According to the interest point, a convex hull is obtained in advance by using the Graham algorithm to calculate the convex hull of the interest point, and a color feature is obtained according to the convex hull of the interest point; Performing translation normalization on the convex hull of the point of interest to obtain a translation convex hull, mapping the pixel coordinates of the translation convex hull into polar coordinates to obtain a mapping convex hull, calculating the zernike moment according to the mapping convex hull, and obtaining the reference target light color feature and the reference target light zernike moment; A similarity value is calculated based on the color feature, the zernike moment, the reference target light color feature, and the reference target light zernike moment, a similarity threshold is preset, and the similarity value is screened based on the similarity threshold to obtain the target light recognition result.

6. The muzzle position recognition method based on infrared scattering signals according to claim 1 is characterized in that: Determining the pixel center point of the target light according to the pixel distance threshold, the shooting area center point, and the target light pixel center point specifically includes: Calculate the distance data based on the center point of the shooting area and the center point of the target light pixel, and determine whether the distance data is greater than the pixel point distance threshold. If so, count the number of points that meet the requirements and record the coordinates of the points that meet the requirements; if not, do not process; A judgment is made based on the number of points that meet the requirements. If the number of points that meet the requirements is greater than 1, a screening rule is preset, and the coordinates of the points that meet the requirements are screened according to the screening rule to obtain the pixel center point of the target target light; if the number of points that meet the requirements is equal to 1, the coordinates of the points that meet the requirements are used as the pixel center point of the target target light.

7. The muzzle position recognition method based on infrared scattering signals according to claim 1 is characterized in that: The coordinate transformation model includes: The pixel coordinates of the center point of the pixel of the target target light are converted into NDC coordinates, a projection matrix is ​​obtained, the NDC coordinates are converted into clipping space coordinates according to the projection matrix, a view matrix is ​​obtained, and the clipping space coordinates are converted into world coordinates according to the view matrix to obtain the actual space coordinates of the target target light.

8. A muzzle position recognition system based on infrared scattering signals, characterized in that: include: Data acquisition module, preprocessing module, identification module, control module; The data acquisition module is used to collect infrared images of the target through the built-in infrared thermal imager of the AR camera, emit active infrared signals through the infrared transmitter, and obtain a trigger pull signal. The smart terminal controls the infrared thermal imager to collect active infrared images of the target according to the trigger pull signal; The preprocessing module is used to calculate an enhanced target infrared image and an enhanced target active infrared image according to the target infrared image and the target active infrared image through a preprocessing model, and calculate an active infrared recognition area according to the enhanced target infrared image and the enhanced target active infrared image through an analysis model; The recognition module is used to calculate the center point of the shooting area according to the active infrared recognition area, collect target image data through an AR camera, pre-process the target image data to obtain target pre-processing data, calculate the target light recognition result through a recognition model according to the target pre-processing data, and calculate the target light pixel center point according to the target light recognition result; The control module is used to preset a pixel point distance threshold, determine the target target light pixel center point based on the pixel point distance threshold, the shooting area center point, and the target light pixel center point, calculate the target target light pixel center point through a coordinate transformation model to obtain the actual spatial coordinates of the target target light, and the intelligent terminal generates an adaptive signal based on the actual spatial coordinates of the target target light to control the corresponding target light to turn off.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the muzzle position recognition method based on infrared scattering signals as described in any one of claims 1-7 is implemented.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to execute the muzzle position recognition method based on infrared scattering signals as described in any one of claims 1 to 7 when executed by a computer processor.

Images