A photoelectric radar integrated detection method and system based on target autonomous identification

Abstract

The present invention relates to an integrated detection method and system for photoelectric radar based on autonomous target identification, and relates to the technical field of integrated detection of photoelectric systems and radar systems. The method comprises the following steps: Step 1: area selection; Step 2: conversion of preset photographing points; Step 3: fan scanning of azimuth components; Step 4: processing by a splicing module; Step 5: processing by an information fusion and recognition module; Step 6: processing by a display and decision module; Step 7: switching of a photoelectric video tracking mode. The integrated detection method and system for photoelectric radar based on autonomous target identification of the present invention combines the target detected by the radar with the target recognition results of the photoelectric system on the wide-width and wide-angle image captured by the radar to determine the authenticity of the target and the tracking priority, thereby improving the tracking control accuracy of the photoelectric system, reducing the radar false alarm rate, and improving the guidance efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated detection of optoelectronic systems and radar systems, and in particular to an optoelectronic radar integrated detection method and system based on autonomous target recognition. Background Art

[0002] With the increasing demand for security protection in China, various security measures are widely used in personal safety, security facilities, prisons, coastline and border control. Common security measures include human patrols, electronic fences, and photoelectric radar area detection. These security measures can meet the needs of security area early warning, target intrusion tracking, and threat disposal.

[0003] In the field of integrated optoelectronic and radar detection systems, the combined use of optoelectronic and radar systems effectively overcomes the limitations of all-day, all-weather applications, leveraging the strengths of individual devices while overcoming their limitations. Their performance surpasses that of manual patrols and electronic fences, and they are widely used in key areas, perimeters of important facilities, border defense, and coastal defense. Radar is responsible for detecting targets within the detection area and acquiring information such as their position and speed, which in turn guides the optoelectronic system to acquire, capture, and track the target's image, thereby achieving precise positioning and identification.

[0004] In response to the above needs, there are currently many relevant solutions at home and abroad.

[0005] Chinese patent document CN202410606505.6 discloses a target detection, tracking and identification method and system based on multi-data fusion. The invention uses infrared and radar for real-time infrared monitoring and radar detection, obtains infrared monitoring data and radar detection data, and performs data preprocessing to generate infrared processing data and radar processing data. Target detection is performed on the infrared processing data and the radar processing data to determine whether there is a suspicious target to be detected. When there is a suspicious target to be detected, the suspicious target's position is determined based on the infrared processing data and the radar processing data, and tracking, photographing and preprocessing are performed to obtain photographing processing data. Based on the photographing processing data, the suspicious target to be detected is confirmed and identified to determine whether there is a confirmed target to be detected. When there is a confirmed target to be detected, the infrared processing data, radar processing data, and photographing processing data are integrated to determine whether there is any abnormal behavior. If there is any abnormal behavior, an abnormal behavior alarm is issued. However, this invention is only applicable to multi-target capture at close range and with a narrow field of view, and its effectiveness is significantly reduced for long distances and wide-angle fields of view.

[0006] Chinese patent document CN202110227211.9 discloses a camera and lidar time synchronization control system. The system comprises a camera, a lidar, and an FPGA. The FPGA is equipped with an image acquisition module, a clock module, and a signal delay module. The clock module is used to receive the lidar clock output by the lidar and use the lidar clock as the camera's operating clock. The signal delay module includes a delay unit and an asynchronous FIFO unit. The delay unit is used to receive the field synchronization signal output by the camera and forward the field synchronization signal to the asynchronous FIFO unit after a preset delay interval. The asynchronous FIFO unit generates a frame synchronization signal based on the field synchronization signal. The frame synchronization signal is used to wake up the lidar and put it into operation. The image acquisition module is used to receive image data collected by the camera. This invention achieves high-precision clock and frame rate synchronization between the camera and lidar at a low cost. However, using the signal delay module to synchronize the clocks between the camera and lidar still results in communication signal delays, making it unsuitable for situations requiring real-time synchronization.

[0007] Generally speaking, the existing optoelectronic and radar integrated detection systems have the following problems in the process of detecting and identifying targets in actual applications:

[0008] (1) Regardless of whether mechanical radar or phased array radar is used, the radar system is used for large-angle detection (typical value ≥90°). Due to its own working mechanism, the scanning and processing speed leads to a low data refresh rate (1-2Hz). The target position information obtained by the radar directly guides the optoelectronic system to track and control with low accuracy, and the picture is shaking and not smooth.

[0009] (2) When used in complex scenarios, radar scanning will be affected by clutter, such as waves, trees and vegetation in windy conditions, which may cause radar detection to be misidentified. Adjusting radar parameters for specific scenarios is cumbersome. In order to ensure an extremely low missed alarm rate, it is impossible to completely eliminate clutter interference, resulting in false alarms in display alarms and guidance photoelectric systems.

[0010] (3) The system generally uses a long-focal-length, narrow-field-of-view optoelectronic imaging component and is equipped with a two-dimensional optoelectronic turntable to meet the needs of long-distance image recognition and tracking. Even when the radar detects multiple real targets, it is necessary to guide the optoelectronic system one by one to make judgments and tracks. At the same time, there is also the problem of repeated target switching and identification, which leads to low guidance efficiency. Summary of the Invention

[0011] The present invention aims to solve the technical problems in the prior art and provide a photoelectric radar integrated detection method and system based on autonomous target identification.

[0012] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0013] A photoelectric radar integrated detection method based on target autonomous identification includes the following steps:

[0014] Step 1: Region selection;

[0015] Calibrate the coordinate systems of the optoelectronic system and radar system, determine the sector scanning boundary of the optoelectronic system through the area type set on the electronic map, and convert the optical focal length and azimuth speed according to the image stitching overlap rate and image recognition pixel number requirements;

[0016] Step 2: Preset the photo point conversion;

[0017] Calculate the optical focal length, the field of view of the optoelectronic system, determine the number of times the optoelectronic system takes pictures, and determine the position where the optoelectronic system takes pictures;

[0018] Step 3: Azimuth component fan scanning;

[0019] Determine the scanning rotation speed of the photoelectric system and trigger the photo taking;

[0020] Step 4: Splicing module processing;

[0021] Based on the images obtained by taking photos, the image overlap rate parameter is used to remove overlapping images, and the parts outside the radar detection area are calculated and removed. Then, the processed images are aligned and spliced ​​to form a wide-angle image with a long-range view to meet the requirements of target detection.

[0022] Step 5: Information fusion and recognition module processing;

[0023] Target detection is performed on the stitched wide-angle images. The target type and position information within the field of view are integrated with the target position information provided by the radar for identification. Targets in the same direction and angle are considered common true targets, while other targets are considered false targets and eliminated. The integrated information is sent to the display and decision-making module.

[0024] Step 6: Display and decision module processing;

[0025] The radar's electronic map will be updated with real-world target type, coordinate position, and speed information, filtering out all invalid interference targets. If there are multiple real targets, the radar will comprehensively judge the target threat level based on the target's distance from the edge of the radar defense zone / protection area, the target's movement speed, and the target type. The most threatening target will be selected, and autonomous decision-making will be made to provide target priority guidance information for the optoelectronic video tracking mode.

[0026] Step 7: Switch the photoelectric video tracking mode;

[0027] The optoelectronic system switches to the target area and turns on the image tracker based on the guidance information of the target to be tracked, locking and continuously tracking the target.

[0028] In the above technical solution, the area types in step 1 include: the entire area and sensitive areas.

[0029] In the above technical solution, step 2 is specifically as follows:

[0030] First calculate the optical focal length:

[0031]

[0032] Among them, f is the optical focal length, PPT is the number of image recognition pixels required, S is the target distance, Pixel size is the pixel size of the camera sensor, and h is the target size;

[0033] Then calculate the field of view of the optoelectronic system:

[0034]

[0035] Among them, θ0 is the field of view of the optoelectronic system, P 宽 is the width of the camera sensor, and f is the optical focal length;

[0036] Determine the number of times the photoelectric system takes pictures:

[0037]

[0038] Where M is the number of times the optoelectronic system takes pictures, and R is the image stitching overlap rate. In the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area is replaced by θ0, which is the field of view of the photoelectric system.

[0039] Determine the photoelectric system's photographic position. In the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area will be replaced;

[0040] The position of the photoelectric system when scanning from left to right is:

[0041]

[0042] The position of the photoelectric system when scanning from right to left is:

[0043]

[0044] Where N is 0, 1, ..., M in sequence, R is the image stitching overlap rate, and θ0 is the field of view of the optoelectronic system.

[0045] In the above technical solution, step 3 is specifically as follows:

[0046] Determine the scanning rotation speed of the photoelectric system:

[0047]

[0048] Among them, v0 is the scanning rotation speed of the photoelectric system, t r is the radar refresh cycle; in the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area is replaced, and θ0 is the field of view angle of the photoelectric system.

[0049] In the above technical solution, step 4 is specifically as follows:

[0050] Based on the M images obtained by M shots, the image overlap ratio parameter R is used to remove overlapping images, and the formula M×θ0-θ r The parts beyond the radar detection area are calculated and eliminated, and then the processed images are aligned and spliced ​​to form a wide-angle image with a long-range view to meet the requirements of target detection.

[0051] A photoelectric radar integrated detection system based on target autonomous identification is applicable to the above-mentioned photoelectric radar integrated detection method, wherein the radar system and the photoelectric system are coaxially arranged.

[0052] The present invention has the following beneficial effects:

[0053] The present invention's optoelectronic radar integrated detection method and system based on autonomous target identification, and the combined use of optoelectronic systems and radar systems, effectively resolve the limitations of all-day and all-weather applications. The radar detects the target and obtains target position and speed information, further guiding the optoelectronic system to complete the acquisition, capture, and tracking of the target image, thereby comprehensively realizing the precise positioning and identification of the target, and is widely used in scenarios such as border and coastal defense and facility perimeter security.

[0054] The present invention provides an integrated detection method and system for optoelectronic radar based on autonomous target identification. By combining the target detected by the radar with the target identification results of the optoelectronic system on the wide-width and wide-angle images it captures, the authenticity of the target and the tracking priority are determined, thereby improving the tracking and control accuracy of the optoelectronic system, reducing the radar false alarm rate, and improving guidance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0056] Figure 1 The figure is a flow chart of the steps of the photoelectric radar integrated detection method based on autonomous target identification of the present invention.

[0057] Figure 2 Schematic diagram of the spatial coordinates of the optoelectronic system and radar system.

[0058] Figure 3 Schematic diagram of the field of view and trigger matching relationship between the optoelectronic system and the radar system in the full working range.

[0059] Figure 4 Schematic diagram of the field of view and trigger matching relationship between the optoelectronic system and the radar system to delineate the working range of sensitive areas.

[0060] Figure 5 Schematic diagram of target information fusion and false target elimination. DETAILED DESCRIPTION

[0061] The inventive concept of the present invention is:

[0062] The present invention's optoelectronic radar integrated detection method and system based on autonomous target identification combines the target detected by the radar system with the target identification results of the optoelectronic system on the wide-width and wide-angle images it captures, determines the authenticity of the target and the tracking priority, improves the tracking and control accuracy of the optoelectronic system, reduces the radar false alarm rate, and improves guidance efficiency.

[0063] The present invention will be described in detail below with reference to the accompanying drawings.

[0064] The photoelectric radar integrated detection method based on target autonomous identification of the present invention is mainly used in the inspection and identification mode of the photoelectric and radar integrated monitoring system, and completes the target discovery and identification by combining radar scanning with photoelectric sector scanning.

[0065] The optoelectronic radar integrated detection method based on target autonomous identification of the present invention is applicable to the situation where the radar system and the optoelectronic system are coaxially installed.

[0066] The process of the photoelectric radar integrated detection method based on target autonomous identification of the present invention is as follows: Figure 1 As shown. The optoelectronic imaging system has two operating modes: triggered photography and video preview. In the triggered photography mode, the optoelectronic imaging component operates in a long-focal-length fixed-focus state that meets the system's working distance, meeting the basic requirements for image recognition within the radar's entire range or the longest detection distance within a designated sensitive area. Both the camera and the radar have built-in independent GPS clock synchronization devices to handle the synchronization of the scanning frequency, synchronizing time information once per second to ensure that the camera's photography time accurately matches the radar's scanning cycle. Within a radar refresh cycle, the optoelectronic imaging component is triggered at a specific time and location to take a photo and acquire an image. In the video preview mode, the optoelectronic imaging component and the image tracking component are used in conjunction, mainly for tracking and collecting evidence of specific targets.

[0067] The optoelectronic system azimuth component fan-scans, and the fan-scan frequency is consistent with the radar scanning refresh frequency, that is, within one refresh cycle, the optoelectronic system azimuth component completes a unidirectional fan-scan motion consistent with the radar's horizontal working angle range or the set sensitive area boundary angle range; within two refresh cycles, the optoelectronic system azimuth component realizes one round-trip motion.

[0068] At the same time, during the azimuth fan scanning process, the imaging component of the optoelectronic system converts the angle of rotation of the optical axis according to a certain field of view overlap rate (such as a 1 / 4 overlap rate between the front and rear frames), triggers photography at relevant positions, and sends the images to the image stitching module for registration, stitching, and cropping to form a frame of wide-angle image.

[0069] The stitched single-frame image is fed into the information fusion target recognition module, where the target type and location are identified using a target detection algorithm and then compared and fused with the radar scanned target. The system then conducts a comprehensive hazard assessment based on target type, location, and speed. The target is then imported into subsequent modules, switching to pure optoelectronic video tracking mode for tracking and evidence collection.

[0070] The photoelectric radar integrated detection method based on target autonomous identification of the present invention comprises the following steps:

[0071] Step 1: Region selection;

[0072] This step involves the calibration of the coordinate systems of the optoelectronic system and the radar system. The sector scanning boundary of the optoelectronic system is determined by the area type (global / sensitive area) set on the electronic map. The optical focal length, azimuth speed, etc. are converted according to the image stitching overlap rate and the number of image recognition pixels.

[0073] According to the coaxial installation of the radar system and the optoelectronic system, it can be simplified as follows: Figure 2 coordinate system, where θ0 is the field of view of the optoelectronic system, θ r is the field of view of the radar system. The red area is the scanning area of ​​the optoelectronic system, and the blue area is the detection area of ​​the radar system.

[0074] exist Figure 3 In the full working range, the initial coverage area of ​​the optoelectronic system is FOV1, and the coverage area after the first rotation is FOV2. The overlapping area of ​​the two rotation fields of view is the field of view overlap rate (Overlap). After M rotations, the entire radar detection area is covered.

[0075] exist Figure 4 Within the working range of the sensitive area, the area that the photoelectric system needs to cover is the area enclosed by the red straight line AB. If there are multiple sensitive areas, they will rotate in sequence according to the number of sensitive areas and the refresh cycle.

[0076] Step 2: Preset the photo point conversion;

[0077] First calculate the optical focal length:

[0078]

[0079] Among them, f is the optical focal length, PPT is the number of image recognition pixels required, S is the target distance, Pixel size is the pixel size of the camera sensor, and h is the target size.

[0080] Then calculate the field of view of the optoelectronic system:

[0081]

[0082] Among them, θ0 is the field of view of the optoelectronic system, P 宽 is the width of the camera sensor and f is the optical focal length.

[0083] Determine the number of times the photoelectric system takes pictures:

[0084]

[0085] Where M is the number of times the optoelectronic system takes pictures, and R is the image stitching overlap rate. In the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area (such as Figure 4 ∠AB) in the figure, θ0 is the field of view of the optoelectronic system.

[0086] Determine the photoelectric system's photographic position. In the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area (such as Figure 4 is replaced by ∠AB) in .

[0087] The position of the photoelectric system when scanning from left to right is:

[0088]

[0089] The position of the photoelectric system when scanning from right to left is:

[0090]

[0091] R is the image stitching overlap ratio, and θ0 is the field of view of the optoelectronic system.

[0092] Step 3: Azimuth component fan scanning;

[0093] Determine the scanning rotation speed of the photoelectric system:

[0094]

[0095] Among them, v0 is the scanning rotation speed of the photoelectric system, t ris the radar refresh cycle. In the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area (such as Figure 4 ∠AB) in the figure, θ0 is the field of view of the optoelectronic system.

[0096] To ensure synchronization between the optoelectronic and radar scanning frequencies, both the camera and radar incorporate independent GPS clock synchronization devices and processors. The optoelectronic system's time is calibrated based on radar time. This synchronization occurs once per second during system operation, ensuring precise alignment between camera capture times and radar scanning cycles.

[0097] Step 4: Splicing module processing;

[0098] Based on the M images obtained by M shots, the image overlap ratio parameter R is used to remove overlapping images, and the formula M×θ0-θ r The parts beyond the radar detection area are calculated and eliminated, and then the processed images are aligned and spliced ​​to form a wide-angle image with a long-range view, which can meet the requirements of target detection.

[0099] Step 5: Information fusion and recognition module processing;

[0100] like Figure 5 As shown in the figure, target detection is performed on the stitched wide-angle image. Target type and position information within the field of view are integrated with the target position information provided by the radar for identification. Targets in the same direction and angle are considered common true targets, while other targets are considered false targets and are eliminated. The integrated information is then fed into the next level module.

[0101] Step 6: Display and decision module processing;

[0102] The radar electronic map will be refreshed synchronously (the refresh rate is the same as the radar refresh cycle t r The radar then automatically filters out all valid interference targets, including true target type, coordinate position, and speed, while maintaining consistency. If there are multiple true targets, the radar can comprehensively assess their threat level based on information such as their distance from the radar's defense zone edge / protection zone, their speed, and their type. The radar then selects the most threatening target, making its own decision and providing target priority guidance information for the optoelectronic video tracking mode.

[0103] Step 7: Switch the photoelectric video tracking mode;

[0104] The optoelectronic system switches to the target area and turns on the image tracker according to the target guidance information to be tracked, locking and continuously tracking the target. The content in step 7 is a typical common method and is not the focus of the description of this invention, so it will not be repeated here.

[0105] The present invention's optoelectronic radar integrated detection method and system based on autonomous target identification, and the combined use of optoelectronic systems and radar systems, effectively resolve the limitations of all-day and all-weather applications. The radar detects the target and obtains target position and speed information, further guiding the optoelectronic system to complete the acquisition, capture, and tracking of the target image, thereby comprehensively realizing the precise positioning and identification of the target, and is widely used in scenarios such as border and coastal defense and facility perimeter security.

[0106] The present invention provides an integrated detection method and system for optoelectronic radar based on autonomous target identification. By combining the target detected by the radar with the target identification results of the optoelectronic system on the wide-width and wide-angle images it captures, the authenticity of the target and the tracking priority are determined, thereby improving the tracking and control accuracy of the optoelectronic system, reducing the radar false alarm rate, and improving guidance efficiency.

[0107] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A photoelectric radar integrated detection method based on target autonomous identification, characterized in that: The following steps are involved: Step 1: Region selection; Calibrate the coordinate systems of the optoelectronic system and radar system, determine the sector scanning boundary of the optoelectronic system through the area type set on the electronic map, and convert the optical focal length and azimuth speed according to the image stitching overlap rate and image recognition pixel number requirements; Step 2: Preset the photo point conversion; Calculate the optical focal length, the field of view of the optoelectronic system, determine the number of times the optoelectronic system takes pictures, and determine the position where the optoelectronic system takes pictures; Step 3: Azimuth component fan scanning; Determine the scanning rotation speed of the photoelectric system and trigger the photo taking; Step 4: Splicing module processing; Based on the images obtained by taking photos, the image overlap rate parameter is used to remove overlapping images, and the parts outside the radar detection area are calculated and removed. Then, the processed images are aligned and spliced ​​to form a wide-angle image with a long-range view to meet the requirements of target detection. Step 5: Information fusion and recognition module processing; Target detection is performed on the stitched wide-angle images. The target type and position information within the field of view are integrated with the target position information provided by the radar for identification. Targets in the same direction and angle are considered common true targets, while other targets are considered false targets and eliminated. Send the fused information to the display and decision module; Step 6: Display and decision module processing; The radar's electronic map will be updated with real-world target type, coordinate position, and speed information, filtering out all invalid interference targets. If there are multiple real targets, the radar will comprehensively judge the target threat level based on the target's distance from the edge of the radar defense zone or the protection zone, the target's movement speed, and the target type. The most threatening target will be selected, and autonomous decision-making will be made to provide target priority guidance information for the optoelectronic video tracking mode. Step 7: Switch the photoelectric video tracking mode; The optoelectronic system switches to the target area and turns on the image tracker based on the guidance information of the target to be tracked, locking and continuously tracking the target.

2. The photoelectric radar integrated detection method based on target autonomous identification according to claim 1 is characterized in that: The area types in step 1 include: global area and sensitive area.

3. The photoelectric radar integrated detection method based on target autonomous identification according to claim 1 is characterized in that: Step 2 is as follows: First calculate the optical focal length: Where f is the optical focal length, PPT is the number of image recognition pixels required, S is the target distance, Pixel size is the pixel size of the camera sensor, and h is the target size; Then calculate the field of view of the optoelectronic system: Among them, θ0 is the field of view of the optoelectronic system, P 宽 is the width of the camera sensor, and f is the optical focal length; Determine the number of times the photoelectric system takes pictures: All incoming and outgoing orders Where M is the number of times the optoelectronic system takes pictures, and R is the image stitching overlap rate. In the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area is replaced by θ0, which is the field of view of the photoelectric system. Determine the photoelectric system's photographic position. In the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area will be replaced; The position of the photoelectric system when scanning from left to right is: The position of the photoelectric system when scanning from right to left is: Where N is 0, 1, ..., M in sequence, R is the image stitching overlap rate, and θ0 is the field of view of the optoelectronic system.

4. The photoelectric radar integrated detection method based on target autonomous identification according to claim 3 is characterized in that: Step 3 is as follows: Determine the scanning rotation speed of the photoelectric system: Among them, v0 is the scanning rotation speed of the photoelectric system, t r is the radar refresh cycle; in the case of sensitive area division, the radar field of view angle θ r The angle formed by the two points on the left and right boundaries of the sensitive area is replaced, and θ0 is the field of view angle of the photoelectric system.

5. The photoelectric radar integrated detection method based on target autonomous identification according to claim 4 is characterized in that: Step 4 is as follows: Based on the M images obtained by M shots, the image overlap ratio parameter R is used to remove overlapping images, and the formula M×θ0-θ r The parts beyond the radar detection area are calculated and eliminated, and then the processed images are aligned and spliced ​​to form a wide-angle image with a long-range view to meet the requirements of target detection.

6. A photoelectric radar integrated detection system based on target autonomous identification applicable to the photoelectric radar integrated detection method according to any one of claims 1 to 5, characterized in that: The radar system and the optoelectronic system are coaxially arranged.

Images