Low, small and slow target identification and positioning system and method based on Leiyu fusion

Abstract

The invention discloses a low, small and slow target identification and positioning system and method based on Leighting fusion, and relates to the technical field of low, small and slow target identification and positioning, and the system comprises a Leighting detection module, a target identification module and a target positioning module. The features of the target are deeply analyzed through the neural network model, the analyzed target is continuously judged, on the basis of improving the low, small and slow target recognition efficiency, the situation of low, small and slow target recognition errors is avoided, a target positioning module is arranged, and after the low, small and slow target is recognized, the low, small and slow target recognition efficiency is improved. All devices in the positioning auxiliary range are subjected to positioning analysis, and proper devices are quickly judged and selected to perform auxiliary positioning on the low, small and slow targets, so that the positioning accuracy of the low, small and slow targets can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of low, small and slow target recognition and positioning, and more specifically, to a low, small and slow target recognition and positioning system and method based on radar-vision fusion. Background Art

[0002] Low, small and slow targets usually refer to those with a flight altitude of ≤1000m, a speed of ≤200km / h and a radar cross-section (RCS) of <2m. 2 These targets are usually difficult to detect through single radar detection or visual sensor detection, so the current detection method for low, small and slow targets is more radar-visual fusion. Although radar-visual fusion can detect low, small and slow targets, it is often difficult to ensure rapid identification of low, small and slow targets. At the same time, it is impossible to accurately locate low, small and slow targets due to their flight status and environmental factors. Summary of the invention

[0003] In view of the deficiencies in the prior art, the object of the present invention is to provide a low, small and slow target recognition and positioning system and method based on radar and vision fusion.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A low, small and slow target recognition and positioning system based on radar and vision fusion, comprising a radar and vision detection module, a target recognition module and a target positioning module;

[0006] The radar detection module is used to control the device to detect the target through the radar technology and obtain the radar detection record of the target;

[0007] The target recognition module is used to obtain the low, small, and slow recognition value of the target in real time, and determine whether the target is a low, small, and slow target based on the comparison result of the low, small, and slow recognition value and its threshold;

[0008] After the target is marked as a low, small and slow target, the target positioning module analyzes the remaining devices within the positioning assistance range, determines the device that assists in positioning the low, small and slow target, and locates the low, small and slow target through multiple devices.

[0009] Furthermore, the target is detected by using the radar vision technology to obtain the target's radar vision detection record, specifically: the device is controlled to shoot a video toward the monitoring position, a monitoring video is generated in real time, and the monitoring video is subjected to frame extraction processing to obtain a monitoring image;

[0010] When a target appears in a monitoring image, the device is marked as an identification device, the target is numbered, and a radar detection record is generated while the monitoring image is generated.

[0011] Furthermore, the radar detection record includes the target number, low, small and slow analysis value, and detection time.

[0012] Furthermore, the low, small and slow analysis values ​​recorded by the radar detection are obtained in the following manner: the image features and radar features of the target are obtained, the image features and radar features are packaged and processed to obtain a target feature set, the target feature set is used as the input data of the low, small and slow analysis model, and the low, small and slow analysis values ​​are obtained as output.

[0013] Furthermore, the image features and radar features of the target are obtained in the following manner: extracting features of the target in the monitoring image to obtain the image features, and simultaneously obtaining the radar features of the target.

[0014] Furthermore, the low, small and slow analysis model is obtained in the following manner: multiple target feature sets are collected, different low, small and slow analysis values ​​are assigned to different target feature sets, the target feature sets are divided into a training set and a validation set, and the neural network is iteratively trained on the training set and the validation set to obtain the low, small and slow analysis model. The larger the low, small and slow analysis value, the more the target meets the target characteristics of low, small and slow.

[0015] Furthermore, the low, small, and slow identification value of the target is obtained in the following manner: obtain all the radar detection records of the numbered target before the current time of the system, sort all the radar detection records in the order of detection time, calculate the difference between the two adjacent low, small, and slow analysis values ​​after sorting, and take the absolute value to obtain the low, small, and slow analysis change value, set the low, small, and slow analysis change threshold, and when the low, small, and slow analysis change value is greater than or equal to the low, small, and slow analysis change threshold, increase the number of analysis changes by one, sum up all the analysis change times of the numbered target before the current time of the system, obtain the total number of analysis changes, and mark it as MST. Sum and average the low, small and slow analysis values ​​of all the radar detection records to get the low, small and slow analysis mean, and mark it as CSK. Set the low, small and slow analysis threshold. When the low, small and slow analysis value of the radar detection record is greater than or equal to the low, small and slow analysis threshold, mark the radar detection record as a low, small and slow detection record. Sort all low, small and slow detection records in the order of detection time. Calculate the time difference between the detection times of two adjacent low, small and slow detection records after sorting to get the low, small and slow detection interval. Sum and average all low, small and slow detection intervals to get the average low, small and slow detection interval and mark it as WZB. Use the formula The low, small and slow recognition value DXM of the target is obtained, where a1 is the total number of analysis changes coefficient, a2 is the low, small and slow analysis mean coefficient, and a3 is the average low, small and slow detection interval coefficient.

[0016] Furthermore, based on the comparison result between the low, small, and slow recognition value and its threshold, it is determined whether the target is a low, small, and slow target, specifically: a low, small, and slow recognition threshold is set, and when the low, small, and slow recognition value of the target is greater than or equal to the low, small, and slow recognition threshold, the target is marked as a low, small, and slow target.

[0017] Furthermore, after the target is marked as a low, small, and slow target, a circle is drawn with the position of the recognition device as the center and a preset radius to obtain a positioning assistance range, and the remaining devices within the positioning assistance range are marked as potential auxiliary devices. The recognition device locates the low, small, and slow target every t. While the recognition device locates the low, small, and slow target, the potential auxiliary device locates the low, small, and slow target. A positioning record is generated for each positioning, and the positioning record includes the device ID, positioning coordinates, and positioning time.

[0018] The identification device is compared with each potential auxiliary device to obtain measurement reference values ​​of the identification device and the potential auxiliary device, and a measurement reference threshold is set. When the measurement reference value is greater than or equal to the measurement reference threshold, the potential auxiliary device no longer locates the low, small, and slow targets. When the measurement reference value is less than the measurement reference threshold, the potential auxiliary device is marked as a positioning auxiliary device. The positioning auxiliary device and the identification device simultaneously locate the low, small, and slow targets, and the positioning coordinates of the positioning auxiliary device are used as auxiliary references for the positioning coordinates of the identification device.

[0019] The measurement reference values ​​of the identification device and the potential auxiliary device are obtained by: obtaining the positioning variable stabilization value of the identification device and the positioning variable stabilization value of the potential auxiliary device, performing difference calculation on the positioning variable stabilization value of the identification device and the positioning variable stabilization value of the potential auxiliary device and taking the absolute value, thereby obtaining the measurement reference values ​​of the identification device and the potential auxiliary device;

[0020] The positioning variable stability value is obtained by the following method: obtain all positioning records of the device within n time periods before the current system time, sort all positioning records in the order of positioning time, mark the positioning coordinates of the adjacent and previous positioning records after sorting as (X, Y, Z), and mark the positioning coordinates of the adjacent and subsequent positioning records after sorting as (X / , Y / , Z / ), using the formula Get the positioning coordinate difference FGS, set the positioning coordinate threshold difference, when the positioning coordinate difference is greater than or equal to the positioning coordinate threshold difference, increase the number of positioning changes by one, when the positioning coordinate difference is less than the positioning coordinate threshold difference, mark the positioning coordinate difference as a stable coordinate difference, sum up all stable coordinate differences and take the average to get the average stable coordinate difference, and mark it as Ks. Sum the number of positioning changes to get the total number of positioning changes, and mark it as Tu. Use the formula The positioning stability value Gm of the equipment is obtained, where b1 is the average stable coordinate difference coefficient and b2 is the total number of positioning changes coefficient.

[0021] Furthermore, a method for identifying and locating low, small and slow targets based on radar and vision fusion includes the following steps:

[0022] Step 1: The device detects the target through the radar technology and obtains the radar detection record of the target;

[0023] Step 2: Obtain the low, small, and slow recognition value of the target in real time, and determine whether the target is a low, small, and slow target based on the comparison result between the low, small, and slow recognition value and its threshold;

[0024] Step 3: After the target is marked as a low, small, and slow target, analyze the remaining devices within the positioning assistance range to determine the device that assists in positioning the low, small, and slow target, and locate the low, small, and slow target through multiple devices.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] 1. The positioning method of the present invention can improve the recognition efficiency of low, small and slow targets and avoid the occurrence of low, small and slow target recognition errors;

[0027] 2. Set up the radar detection module and the target recognition module, conduct in-depth analysis of the target's features through the neural network model, and make continuous judgments on the analyzed targets. Set up the target positioning module, and after identifying low, small, and slow targets, conduct positioning analysis on all devices within the positioning assistance range, quickly judge and select appropriate equipment to assist in the positioning of low, small, and slow targets, which can improve the positioning accuracy of low, small, and slow targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flowchart of a method for identifying and locating low, small and slow targets based on radar and visual fusion;

[0029] Figure 2 The flowchart is for determining whether a target is a low, small and slow target. DETAILED DESCRIPTION

[0030] Example 1

[0031] Reference Figure 1 , a method for identifying and locating low, small and slow targets based on radar and vision fusion, comprising the following steps:

[0032] Step 1: The device detects the target through the radar technology and obtains the radar detection record of the target;

[0033] Step 2: Obtain the low, small, and slow recognition value of the target in real time, and determine whether the target is a low, small, and slow target based on the comparison result between the low, small, and slow recognition value and its threshold;

[0034] Step 3: After the target is marked as a low, small, and slow target, analyze the remaining devices within the positioning assistance range to determine the device that assists in positioning the low, small, and slow target, and locate the low, small, and slow target through multiple devices.

[0035] Example 2

[0036] Reference Figure 2 A low, small and slow target recognition and positioning system based on radar and vision fusion includes a radar and vision detection module, a target recognition module and a target positioning module.

[0037] The radar and vision detection module is used to control the device (the device is a radar and vision integrated device that can realize radar and vision fusion) to shoot video towards the monitoring position, generate monitoring video in real time, extract frames from the monitoring video, and obtain monitoring images.

[0038] When a target appears in the monitoring image (the visual contour recognition technology can be used to analyze whether a target appears in the monitoring image. The target is an object that appears outside the set scene and is not necessarily a low, small, and slow target), the device is marked as an identification device, and the target is numbered (each target corresponds to a unique number). When the monitoring image is generated, a radar detection record is generated. The radar detection record includes the target number, low, small, and slow analysis value, and detection time (i.e., the generation time of the radar detection record).

[0039] The low, small and slow analysis values ​​recorded by radar detection are obtained in the following way: the image features and radar features of the target are obtained, the image features and radar features are packaged and processed to obtain a target feature set, the target feature set is used as the input data of the low, small and slow analysis model, and the low, small and slow analysis values ​​are obtained as output.

[0040] The image features and radar features of the target are obtained in the following manner: feature extraction is performed on the target in the monitoring image to obtain image features (image features include but are not limited to texture features, shape features, and spatial relationship features of the target), and (through the radar function of the device) radar features of the target are simultaneously obtained (radar features include but are not limited to the speed of the target, the position of the target, and the moving direction of the target).

[0041] The low, small and slow analysis model is obtained by the following method: collecting multiple target feature sets, the features in the target feature set can be virtual settings or real features, assigning different low, small and slow analysis values ​​to different target feature sets, dividing the target feature set into a training set and a validation set, and performing neural network iterative training on the training set and the validation set to obtain the low, small and slow analysis model. The value of the low, small and slow analysis value is greater than 0. The larger the value of the low, small and slow analysis value, the more the target meets the target characteristics of low, small and slow.

[0042] Target recognition module: obtain the low, small, and slow recognition value of each target in real time, set the low, small, and slow recognition threshold, which is the system setting threshold and can be modified according to actual needs. When the low, small, and slow recognition value of the target is greater than or equal to the low, small, and slow recognition threshold, the target is marked as a low, small, and slow target. When the low, small, and slow recognition value of the target is less than the low, small, and slow recognition threshold, no corresponding processing is performed.

[0043] The low, small, and slow identification value of the target is obtained in the following way: obtain all the radar detection records of the numbered target before the current time of the system, sort all the radar detection records in the order of detection time, calculate the difference between the two adjacent low, small, and slow analysis values ​​after sorting and take the absolute value to obtain the low, small, and slow analysis change value, set the low, small, and slow analysis change threshold, when the low, small, and slow analysis change value is greater than or equal to the low, small, and slow analysis change threshold, increase the number of analysis changes once, when the low, small, and slow analysis change value is less than the low, small, and slow analysis change threshold, do not do the corresponding processing, sum up all the analysis change times of the numbered target before the current time of the system, obtain the total number of analysis changes, and mark it as MST, and perform the low, small, and slow analysis value of all the radar detection records. Sum and average the values ​​to get the low, small and slow analysis mean value, and mark it as CSK. Set the low, small and slow analysis threshold. The low, small and slow analysis threshold is the system setting threshold, which can be modified according to actual needs. When the low, small and slow analysis value of the radar detection record is greater than or equal to the low, small and slow analysis threshold, the radar detection record is marked as a low, small and slow detection record. When the low, small and slow analysis value of the radar detection record is less than the low, small and slow analysis threshold, no corresponding processing is performed. All low, small and slow detection records are sorted in the order of detection time. The detection time of the two adjacent low, small and slow detection records after sorting is calculated to obtain the low, small and slow detection interval. All low, small and slow detection intervals are summed and averaged to obtain the average low, small and slow detection interval and marked as WZB. Use the formula The low, small and slow recognition value DXM of the target is obtained, where a1 is the total number of analysis changes coefficient, a2 is the mean coefficient of low, small and slow analysis, and a3 is the average low, small and slow detection interval coefficient. The value of a1 is 0.45, the value of a2 is 0.81, and the value of a3 is 0.69.

[0044] A radar detection module and a target recognition module are set up to conduct in-depth analysis of the target's features through a neural network model, and to make continuous judgments on the analyzed targets. This will improve the efficiency of identifying low, small, and slow targets and avoid errors in identifying low, small, and slow targets.

[0045] Target positioning module: After the target is marked as a low, small and slow target, a circle is drawn with the position of the recognition device as the center and the preset radius to obtain the positioning assistance range. The remaining devices (the remaining radar-vision integrated devices) within the positioning assistance range are marked as potential auxiliary devices. The recognition device locates the low, small and slow targets every t. While the recognition device locates the low, small and slow targets, the potential auxiliary devices locate the low, small and slow targets (the positioning of low, small and slow targets can be completed only by the radar function of the potential auxiliary device, without the need to use the visual function). A positioning record is generated for each positioning, which includes the device ID, positioning coordinates and positioning time.

[0046] The identification device is compared with each potential auxiliary device to obtain the measurement reference values ​​of the identification device and the potential auxiliary device, and a measurement reference threshold is set. The measurement reference threshold is a threshold preset by the system and can be modified according to actual needs. When the measurement reference value is greater than or equal to the measurement reference threshold, the potential auxiliary device will no longer locate low, small and slow targets. When the measurement reference value is less than the measurement reference threshold, the potential auxiliary device will be marked as a positioning auxiliary device. The positioning auxiliary device and the identification device simultaneously locate low, small and slow targets, and the positioning coordinates located by the positioning auxiliary device are used as a reference.

[0047] The measurement reference values ​​of the identification device and the potential auxiliary device are obtained in the following manner: obtain the positioning stabilization value of the identification device and the positioning stabilization value of the potential auxiliary device, calculate the difference between the positioning stabilization value of the identification device and the positioning stabilization value of the potential auxiliary device and take the absolute value to obtain the measurement reference values ​​of the identification device and the potential auxiliary device.

[0048] The positioning variable stability value is obtained by the following method: obtain all positioning records of the device within n time periods before the current system time (n time period is greater than t time period), sort all positioning records in the order of positioning time, mark the positioning coordinates of the adjacent and previous positioning records after sorting as (X, Y, Z), and mark the positioning coordinates of the adjacent and subsequent positioning records after sorting as (X / , Y / , Z / ), using the formula Get the positioning coordinate difference FGS, set the positioning coordinate threshold difference, when the positioning coordinate difference is greater than or equal to the positioning coordinate threshold difference, increase the number of positioning changes by one, when the positioning coordinate difference is less than the positioning coordinate threshold difference, mark the positioning coordinate difference as a stable coordinate difference, sum up all stable coordinate differences and take the average to get the average stable coordinate difference, and mark it as Ks. Sum the number of positioning changes to get the total number of positioning changes, and mark it as Tu. Use the formula The positioning stability value Gm of the equipment is obtained, where b1 is the average stable coordinate difference coefficient, b2 is the total number of positioning changes coefficient, the value of b1 is 0.59, and the value of b2 is 0.48.

[0049] By setting up the target positioning module, after identifying low, small and slow targets, all devices within the positioning assistance range can be analyzed for positioning, and the appropriate device can be quickly determined and selected to assist in the positioning of low, small and slow targets, thereby improving the positioning accuracy of low, small and slow targets.

[0050] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0051] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0052] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0053] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0054] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0055] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0056] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0057] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A low, small and slow target recognition and positioning system based on radar and vision fusion, characterized in that: It includes radar detection module, target recognition module and target positioning module; The radar detection module is used to control the device to detect the target through the radar technology and obtain the radar detection record of the target; The target recognition module is used to obtain the low, small, and slow recognition value of the target in real time, and determine whether the target is a low, small, and slow target based on the comparison result of the low, small, and slow recognition value and its threshold; After the target is marked as a low, small and slow target, the target positioning module analyzes the remaining devices within the positioning assistance range, determines the device that assists in positioning the low, small and slow target, and locates the low, small and slow target through multiple devices.

2. According to claim 1, a low, small and slow target recognition and positioning system based on radar and visual fusion is characterized in that: Detect the target through the radar vision technology and obtain the target's radar vision detection record, specifically: control the device to shoot video toward the monitoring position, generate monitoring video in real time, extract frames from the monitoring video, and obtain monitoring images; When a target appears in a monitoring image, the device is marked as an identification device, the target is numbered, and a radar detection record is generated while the monitoring image is generated.

3. The low, small and slow target recognition and positioning system based on radar and visual fusion according to claim 2 is characterized in that: The radar detection record includes the target number, low, small and slow analysis value and detection time.

4. The low, small and slow target recognition and positioning system based on radar and visual fusion according to claim 3 is characterized in that: The low, small and slow analysis values ​​recorded by radar detection are obtained in the following way: the image features and radar features of the target are obtained, the image features and radar features are packaged and processed to obtain a target feature set, the target feature set is used as the input data of the low, small and slow analysis model, and the low, small and slow analysis values ​​are obtained as output.

5. According to claim 4, a low, small and slow target recognition and positioning system based on radar and visual fusion is characterized in that: The image features and radar features of the target are obtained in the following manner: feature extraction is performed on the target in the monitoring image to obtain the image features, and the radar features of the target are simultaneously obtained.

6. The low, small and slow target recognition and positioning system based on radar and visual fusion according to claim 4 is characterized in that: The low, small and slow analysis model is obtained in the following way: collect multiple target feature sets, assign different low, small and slow analysis values ​​to different target feature sets, divide the target feature sets into training sets and verification sets, perform neural network iterative training on the training sets and verification sets, and obtain the low, small and slow analysis model. The larger the low, small and slow analysis value, the more the target meets the low, small and slow target characteristics.

7. The low, small and slow target recognition and positioning system based on radar and visual fusion according to claim 1 is characterized in that: The low, small, and slow identification value of the target is obtained in the following way: obtain all the radar detection records of the numbered target before the current time of the system, sort all the radar detection records in the order of detection time, calculate the difference between the two adjacent low, small, and slow analysis values ​​after sorting and take the absolute value to obtain the low, small, and slow analysis change value, set the low, small, and slow analysis change threshold, when the low, small, and slow analysis change value is greater than or equal to the low, small, and slow analysis change threshold, increase the number of analysis changes by one, sum up all the analysis change times of the numbered target before the current time of the system, obtain the total number of analysis changes, and mark it as MST, and The low, small and slow analysis values ​​of the radar vision detection records are summed and averaged to obtain the low, small and slow analysis mean, which is marked as CSK. The low, small and slow analysis threshold is set. When the low, small and slow analysis value of the radar vision detection record is greater than or equal to the low, small and slow analysis threshold, the radar vision detection record is marked as a low, small and slow detection record. All low, small and slow detection records are sorted in the order of detection time. The detection time of the two adjacent low, small and slow detection records after sorting is calculated to obtain the low, small and slow detection interval. All low, small and slow detection intervals are summed and averaged to obtain the average low, small and slow detection interval and marked as WZB. The formula is used The low, small and slow recognition value DXM of the target is obtained, where a1 is the total number of analysis changes coefficient, a2 is the low, small and slow analysis mean coefficient, and a3 is the average low, small and slow detection interval coefficient.

8. The low, small and slow target recognition and positioning system based on radar and visual fusion according to claim 1 is characterized in that: Based on the comparison result between the low, small, and slow recognition value and its threshold, determine whether the target is a low, small, and slow target. Specifically, set a low, small, and slow recognition threshold. When the low, small, and slow recognition value of the target is greater than or equal to the low, small, and slow recognition threshold, mark the target as a low, small, and slow target.

9. The low, small and slow target recognition and positioning system based on radar and visual fusion according to claim 1 is characterized in that: After the target is marked as a low, small and slow target, a circle is drawn with the position of the recognition device as the center and the preset radius to obtain the positioning assistance range. The remaining devices within the positioning assistance range are marked as potential auxiliary devices. The recognition device locates the low, small and slow target every t. While the recognition device locates the low, small and slow target, the potential auxiliary device locates the low, small and slow target. A positioning record is generated for each positioning, which includes the device ID, positioning coordinates and positioning time. The identification device is compared with each potential auxiliary device to obtain measurement reference values ​​of the identification device and the potential auxiliary device, and a measurement reference threshold is set. When the measurement reference value is greater than or equal to the measurement reference threshold, the potential auxiliary device no longer locates the low, small, and slow targets. When the measurement reference value is less than the measurement reference threshold, the potential auxiliary device is marked as a positioning auxiliary device. The positioning auxiliary device and the identification device simultaneously locate the low, small, and slow targets, and the positioning coordinates of the positioning auxiliary device are used as auxiliary references for the positioning coordinates of the identification device. The measurement reference values ​​of the identification device and the potential auxiliary device are obtained by: obtaining the positioning variable stabilization value of the identification device and the positioning variable stabilization value of the potential auxiliary device, performing difference calculation on the positioning variable stabilization value of the identification device and the positioning variable stabilization value of the potential auxiliary device and taking the absolute value, thereby obtaining the measurement reference values ​​of the identification device and the potential auxiliary device; The positioning variable stability value is obtained by the following method: obtain all positioning records of the device within n time periods before the current system time, sort all positioning records in the order of positioning time, mark the positioning coordinates of the adjacent and previous positioning records after sorting as (X, Y, Z), and mark the positioning coordinates of the adjacent and subsequent positioning records after sorting as (X / , Y / , Z / ), using the formula Get the positioning coordinate difference FGS, set the positioning coordinate threshold difference, when the positioning coordinate difference is greater than or equal to the positioning coordinate threshold difference, increase the number of positioning changes by one, when the positioning coordinate difference is less than the positioning coordinate threshold difference, mark the positioning coordinate difference as a stable coordinate difference, sum up all stable coordinate differences and take the average to get the average stable coordinate difference, and mark it as Ks. Sum the number of positioning changes to get the total number of positioning changes, and mark it as Tu. Use the formula The positioning stability value Gm of the equipment is obtained, where b1 is the average stable coordinate difference coefficient and b2 is the total number of positioning changes coefficient.

10. A method for identifying and locating low, small and slow targets based on radar and visual fusion, applied to a system for identifying and locating low, small and slow targets based on radar and visual fusion as claimed in claim 1, characterized in that: The steps include: Step 1: The device detects the target through the radar technology and obtains the radar detection record of the target; Step 2: Obtain the low, small, and slow recognition value of the target in real time, and determine whether the target is a low, small, and slow target based on the comparison result between the low, small, and slow recognition value and its threshold; Step 3: After the target is marked as a low, small, and slow target, analyze the remaining devices within the positioning assistance range to determine the device that assists in positioning the low, small, and slow target, and locate the low, small, and slow target through multiple devices.

Images