A passive optical based target imaging three-dimensional positioning system calibration method

Abstract

The application discloses a kind of target imaging three-dimensional positioning system calibration methods based on passive optics, comprising the following steps S1, point deployment;S2, photoelectric hardware installation;S3, global networking and time synchronization configuration;S4, single equipment inside participates in global external parameter calibration;S5, AI algorithm and three-dimensional solution algorithm deployment;S6, whole system joint operation.The application adopts multi-device regional networking+visible light / infrared dual light passive imaging architecture, relies on large baseline triangulation, subpixel feature extraction and multi-source image fusion algorithm, and can realize millimeter-level three-dimensional positioning of flying targets.Compared with traditional laser ranging positioning scheme, the imaging positioning method breaks through the active ranging accuracy bottleneck, has the comprehensive advantages of high precision, all-weather, strong concealment, multi-target parallel processing, etc., and simultaneously fuses AI visual recognition capability, integrates positioning, imaging, recognition, tracking, etc., greatly improves the wide-area airspace target perception capability, and is suitable for low-altitude security, airspace control and other scenes.

Description

Technical Field

[0001] This invention relates to the fields of target localization and image processing technology, specifically to a calibration method for a three-dimensional target imaging localization system based on passive optics. Background Technology

[0002] Currently, the imaging and positioning system for airspace targets mainly adopts a technical solution of "laser ranging combined with monocular photoelectric imaging". This type of system typically relies on an electro-optical gimbal, equipped with a visible light or infrared camera to complete the imaging observation of the target. Simultaneously, it integrates a pulsed laser module to actively measure the target's slant range. The system obtains the azimuth and pitch angles of the gimbal in real time, and combines the ranging data and imaging information to calculate the target's three-dimensional spatial coordinates (longitude, latitude, and altitude). In actual operation, the operator manipulates the gimbal to scan the airspace; after detecting the target, they switch to lock-on tracking mode, and the system simultaneously outputs the imaging image and target positioning data. This existing technology has advantages such as simple deployment and mature technology, and has been widely used in many fields.

[0003] However, it still has some drawbacks in practical use. For example, pulsed laser ranging is currently an active detection mode, and its emitted laser signals are easily detected by the enemy or optical early warning systems on the target. In military, security and other application scenarios with high requirements for concealment, there is a risk of exposure. At the same time, laser ranging is greatly affected by atmospheric environment (such as fog, rain, snow and strong sunlight interference), and its all-weather working capability is limited. In the current technology, the accuracy of laser ranging is closely related to the target distance and the measurement error of the gimbal angle. At a typical detection range of 1,000 meters, the overall positioning accuracy of the system is usually only at the level of 2 meters, which is difficult to meet the requirements of high-precision tasks. Moreover, traditional systems rely on laser point-to-point ranging, which can only lock and measure the slant range and angle data of one target at a time. If multiple targets appear in the airspace, the system needs to switch to track them in turn, and cannot obtain the three-dimensional position information of multiple targets at the same time, which severely limits the ability of multi-target monitoring, tracking and situational awareness. Summary of the Invention

[0004] The purpose of this invention is to provide a calibration method for a target imaging 3D positioning system based on passive optics. By setting up a multi-device regional network + visible light / infrared dual-light passive imaging architecture, and relying on large baseline triangulation, sub-pixel feature extraction and multi-source image fusion algorithms, millimeter-level 3D positioning of flying targets can be achieved. Compared with traditional laser ranging positioning schemes, this imaging positioning method breaks through the accuracy bottleneck of active ranging and has comprehensive advantages such as high precision, all-weather operation, strong concealment, and multi-target parallel processing. At the same time, it integrates AI visual recognition capabilities, combining positioning, imaging, recognition and tracking into one, which greatly improves the ability to perceive targets in a large airspace. It is applicable to scenarios such as low-altitude security and airspace control, so as to solve the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a calibration method for a target imaging three-dimensional positioning system based on passive optics, comprising the following steps: S1, point deployment: according to the requirements of a planned baseline of 500m and a maximum detection distance of 1000m, select multiple photoelectric station installation locations to ensure that the field of view overlap rate of each station is ≥70% and there is no obstruction in the entire area. Use a total station and GPS equipment to accurately measure the geodetic three-dimensional coordinates of each station, record the original location data, and mark the equipment installation base, cable routing, and the placement locations of the switch and processing host to complete the layout of the entire area. S2. Photoelectric hardware installation: At the points marked in step S1, install multiple dual-light photoelectric devices in sequence, adjust the lens orientation to monitor the airspace, fix the device posture, connect power to each device, lay gigabit network cables, ensure that the physical links of all front-end terminals are connected, and perform a power-on self-test on each device to confirm that the visible light imaging, infrared thermal imaging, and image acquisition functions are normal. S3. Global Networking and Time Synchronization Configuration: Connect all front-end dual-optical devices to the PTP time synchronization switch, and then connect the switch to the back-end core processing host to complete the global network networking; connect the GPS time signal, enable the IEEE 1588PTP time synchronization protocol, and uniformly synchronize the time of all optoelectronic terminals and processing host, calibrate to nanosecond-level synchronization accuracy, test the data transmission link, and ensure that the image data stream and timestamp data of each device can be stably uploaded to the processing host in real time; S4. Single device participation in global extrinsic calibration: S4.1 Single device internal parameter calibration: The visible light channel and infrared channel of each dual-light optoelectronic device are calibrated one by one. The Zhang Zhengyou calibration method is used to collect multi-view calibration images, calculate the internal parameters of lens focal length, principal point, and distortion coefficient, and complete the single device accuracy calibration. S4.2 Joint calibration of external parameters across the entire field of view: Multiple high-precision three-dimensional control points are set up within the monitoring field of view. All photoelectric devices synchronously acquire images of the control points. Combined with the physical coordinates of the stations measured in step S1, the spatial position, attitude angle and other external parameters of each device are uniformly calculated through the bundle adjustment algorithm. S4.3 saves all internal and external parameters of the devices to the system configuration file as a fixed benchmark for 3D calculation, enabling all optoelectronic devices in the sensing layer to complete unified parameter calibration, forming a unified spatial coordinate system among the devices, and ensuring the effectiveness of multi-axis triangulation; calibration data is synchronously sent to the algorithm module to provide a calculation basis for coordinate calculation; S5. Deployment of AI and 3D solution algorithms: Deploy the complete set of algorithm programs in the processing host, including image preprocessing, dual-light image fusion, AI target recognition, sub-pixel feature extraction, multi-view triangulation, and error compensation. Load the device calibration parameters saved in step S4 and the time synchronization parameters in step S3 to complete the matching of algorithm and hardware parameters, and configure the target type library, positioning output format, video encoding rules, and data storage rules. S6. System-wide Joint Operation: Upon system power-up, the system enters automatic operation mode. The optoelectronic equipment continuously acquires visible light and infrared dual-light images and transmits them back in real time. Standard test flight targets are deployed to simulate the movement of targets such as drones and birds within an airspace of 100–1000m. The system automatically completes target detection, tracking, and feature extraction. The background observes the 3D positioning results, imaging video, and recognition results. For minor errors in long-distance and complex environments, the error compensation model and sub-pixel extraction parameters are fine-tuned. The system is continuously tested for 24 hours to verify its stability, latency, and multi-target processing capabilities.

[0006] To further optimize this technical solution, in step S1, the equipment used includes: a high-precision GPS locator, a laser rangefinder, a total station, a measuring tape, and site marking tools. In step S2, the equipment used includes: a dual-light optoelectronic integrated machine, a fixed bracket, a waterproof junction box, a power supply module, and a network cable; In step S3, the equipment used includes: a PTP precision time synchronization switch, a core processing host, and a GPS timing module. In step S4, the equipment used includes: a high-precision calibration board, standard three-dimensional control points, and a core processing host. In implementing step S5, the equipment used includes: the core processing host (GPU computing unit) and the system operation terminal.

[0007] To further optimize this technical solution, the camera of the dual-light optoelectronic device in step S2 has high resolution and low distortion characteristics, following the following trigonometric measurement error formula:

[0008] Definitions of each symbol: σx: Spatial lateral positioning error Z: Target slant range / detection range b: Camera network baseline length f: Camera equivalent focal length σu: Subpixel extraction accuracy of the target image.

[0009] To further optimize this technical solution, the global network deployment in step S3 should meet the following requirements: Number of devices: ≥4 units; Baseline length: 500m; Deployment height: 5-10m; Field of view overlap rate: ≥70%; Coverage range: 120° horizontally × 60° vertically, distance 100-500m.

[0010] To further optimize this technical solution, the intrinsic parameter calibration in step S4 includes: focal length, principal point coordinates, and distortion coefficient, with a calibration accuracy of ≤0.01 pixels; Global extrinsic parameter calibration includes the three-dimensional position and attitude of each camera, requiring a position calibration accuracy of ≤0.1mm and an attitude calibration accuracy of ≤0.001°.

[0011] To further optimize this technical solution, the error compensation model described in step S6 dynamically fine-tunes the sub-pixel extraction parameters and multi-view triangulation results based on measured error data under long-distance and complex environments, in order to compensate for positioning deviations caused by factors such as atmospheric disturbances and lens distortion.

[0012] Beneficial effects

[0013] This invention provides a calibration method for a three-dimensional positioning system for target imaging, which has the following advantages: Employing a multi-device regional networking architecture combined with a visible / infrared dual-light passive imaging system, and relying on large baseline triangulation, sub-pixel feature extraction, and multi-source image fusion algorithms, this system achieves millimeter-level 3D positioning of flying targets under conditions of a 500m baseline and a maximum detection distance of 1000m. Compared to traditional laser ranging and positioning schemes, this imaging positioning method breaks through the accuracy bottleneck of active ranging and combines the comprehensive advantages of high precision, all-weather operation, strong concealment, and parallel processing of multiple targets. It also integrates AI visual recognition capabilities, combining positioning, imaging, recognition, and tracking into one system, significantly improving the ability to perceive targets in a wide airspace. It is suitable for scenarios such as low-altitude security, airspace control, and anti-flying vehicle monitoring. It also has the ability to perform multi-target parallel three-dimensional positioning. By networking multiple dual-optical devices and combining them with multi-view triangulation algorithms, it can simultaneously detect, track, and calculate the three-dimensional coordinates of multiple targets (such as drones, birds, and aircraft) in the monitored airspace. This breaks through the limitation of traditional laser ranging systems that can only lock onto one target at a time, and greatly improves the efficiency of multi-target situational awareness and monitoring. This invention enables the integration of system calibration and algorithm deployment. It integrates the calibration of internal and external parameters of the equipment, the configuration of time synchronization, and the deployment of AI algorithms and 3D solution algorithms into a complete calibration and operation process. The calibration data is directly sent to the algorithm module, realizing the automatic matching of hardware parameters and software algorithms, reducing the complexity of system debugging, and improving the efficiency of system deployment and operation and maintenance. With strong scalability and versatility, the method of this invention is applicable to networking different numbers and models of dual-optical-electrical devices. The number and layout of stations can be flexibly adjusted according to the actual monitoring area size and accuracy requirements. It has good scalability and versatility and is suitable for various application scenarios such as military reconnaissance, border monitoring, airport airspace protection, and security of important facilities. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the overall method flow of the present invention; Figure 2 This is a schematic diagram of the overall architecture of the target imaging three-dimensional positioning system of the present invention; Figure 3 This is a schematic diagram showing the location distribution of multiple photoelectric stations in step S1 of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] As attached Figure 1 To be continued Figure 3 The calibration method of a target imaging three-dimensional positioning system based on passive optics includes the following steps: S1, Point deployment: According to the requirements of a planned baseline of 500m and a maximum detection distance of 1000m, select multiple photoelectric station installation locations to ensure that the field of view overlap rate of each station is ≥70% and there is no obstruction in the entire area. Use a total station and GPS equipment to accurately measure the geodetic three-dimensional coordinates of each station, record the original location data, and mark the equipment installation base, cable routing, and the placement of the switch and processing host to complete the layout of the entire area. Equipment used: high-precision GPS locator, laser rangefinder, total station, measuring tape, and site marking tools; S2. Optoelectronic Hardware Installation: At the points marked in step S1, install multiple dual-light optoelectronic devices sequentially, uniformly adjust the lens orientation to the monitored airspace, fix the device posture, connect power to each device, lay gigabit network cables, ensure the physical link of all front-end terminals is connected, and perform a power-on self-test on each device to confirm that the visible light imaging, infrared thermal imaging, and image acquisition functions are normal. Equipment used: dual-light optoelectronic integrated machine (visible light + infrared thermal imaging), fixed bracket, waterproof junction box, power supply module, network cable. This step completes the hardware construction of the front-end sensing layer. All optoelectronic terminals are unified imaging acquisition units of the system, deployed independently and with overlapping fields of view, which is the basis for subsequent image data and field of view intersection calculation. S3. Global Networking and Time Synchronization Configuration: Connect all front-end dual-optical devices to a PTP time synchronization switch, which then connects to the back-end core processing host to complete the global network setup. Integrate GPS timing signals, enable the IEEE 1588PTP time synchronization protocol, and provide unified time synchronization for all optoelectronic terminals and the processing host, calibrating to nanosecond-level synchronization accuracy. Test the data transmission link to ensure that the image data stream and timestamp data of each device can be stably uploaded to the processing host in real time. Equipment used: PTP precision time synchronization switch, core processing host, GPS timing module. This step establishes the data transmission layer, which is the core hub for the coordinated operation of the entire system. All optoelectronic devices rely on a unified time reference to achieve frame-by-frame imaging, eliminating positioning errors caused by timing deviations. The network architecture enables data communication between the front-end sensing layer and the back-end processing layer. S4. Single device participation in global extrinsic calibration: S4.1 Single device internal parameter calibration: The visible light channel and infrared channel of each dual-light optoelectronic device are calibrated one by one. The Zhang Zhengyou calibration method is used to collect multi-view calibration images, calculate the internal parameters of lens focal length, principal point, and distortion coefficient, and complete the single device accuracy calibration. S4.2 Joint calibration of external parameters across the entire field of view: Multiple high-precision three-dimensional control points are set up within the monitoring field of view. All photoelectric devices synchronously acquire images of the control points. Combined with the physical coordinates of the stations measured in step S1, the spatial position, attitude angle and other external parameters of each device are uniformly calculated through the bundle adjustment algorithm. S4.3 saves all internal and external parameters of the devices to the system configuration file as a fixed benchmark for 3D calculation, enabling all optoelectronic devices in the sensing layer to complete unified parameter calibration, forming a unified spatial coordinate system among the devices, and ensuring the effectiveness of multi-axis triangulation; calibration data is synchronously sent to the algorithm module to provide a calculation basis for coordinate calculation; Equipment used: High-precision calibration board, standard 3D control points, core processing host; S5. Deployment of AI and 3D Solving Algorithms: Deploy the complete set of algorithm programs in the processing host, including image preprocessing, dual-light image fusion, AI target recognition, sub-pixel feature extraction, multi-view triangulation, and error compensation. Load the device calibration parameters saved in step S4 and the time synchronization parameters in step S3 to complete the matching of algorithm and hardware parameters. Configure the target type library (such as drones, birds, aircraft, etc.), positioning output format, video encoding rules, and data storage rules. The equipment used is the core processing host (GPU computing unit) and the system operation terminal. This step builds the core processing layer. The backend algorithm module receives dual-light image data and timestamp data uploaded by the perception layer in real time and performs calculations based on unified calibration parameters. The three major algorithm modules of AI recognition, feature extraction, and 3D solving work in parallel and collaboratively to simultaneously complete target recognition and coordinate solving. S6. System-wide Joint Operation: The system is powered on and enters automatic operation mode. The photoelectric equipment continuously collects visible light and infrared dual-light images and transmits them back in real time. Standard test flight targets are deployed to simulate the movement of targets such as drones and birds in the airspace of 100-1000m. The system automatically completes target detection, tracking, and feature extraction. The background observes the 3D positioning results, imaging video, and recognition results. For small errors in long-distance and complex environments, the error compensation model and sub-pixel extraction parameters are fine-tuned. The system is continuously tested for 24 hours to verify its stability, latency, and multi-target processing capabilities. This step completes the construction of the application output layer.

[0017] The three-dimensional positioning system adopts a four-layer distributed architecture: Front-end sensing layer: Multiple high-precision optoelectronic devices are networked to form overlapping field-of-view coverage. Data transmission layer: Gigabit Ethernet + PTP time synchronization to ensure data real-time performance and synchronization. Core processing layer: AI visual recognition + passive imaging 3D solution dual engine Application output layer: Real-time 3D coordinates, high-definition video, and comprehensive display of target attribute information.

[0018] Each layer of the architecture is implemented sequentially according to the steps in the calibration method. The two complement each other and ultimately complete the operation process of imaging and locating the target.

[0019] In step S2, the camera of the dual-light optoelectronic device has high resolution and low distortion characteristics, and follows the following trigonometric measurement error formula:

[0020] Definitions of each symbol: σx: Spatial lateral positioning error (unit / m) Z: Target slant range / detection range (unit / m) b: Camera network baseline length (unit / m) f: Camera equivalent focal length (pixel focal length) σu: Subpixel extraction precision of the target image.

[0021] The global network deployment in step S3 shall comply with the following requirements: Number of devices: ≥4 units; Baseline length: 500m; Deployment height: 5-10m; Field of view overlap rate: ≥70%; Coverage range: 120° horizontally × 60° vertically, distance 100-500m.

[0022] The intrinsic parameter calibration in step S4 includes: focal length, principal point coordinates, and distortion coefficients, with a calibration accuracy of ≤0.01 pixels. Global extrinsic parameter calibration includes the three-dimensional position and attitude of each camera, requiring a position calibration accuracy of ≤0.1mm and an attitude calibration accuracy of ≤0.001°.

[0023] The error compensation model described in step S6 dynamically fine-tunes the sub-pixel extraction parameters and multi-view triangulation results based on measured error data under long distance and complex environments, in order to compensate for positioning deviations caused by factors such as atmospheric disturbances and lens distortion.

[0024] The calculation of the target data includes: Using a pinhole camera model, the three-dimensional world coordinates (X, Y, Z) are converted into two-dimensional image coordinates (u, v):

[0025] in: fx, fy: Focal length (in pixels) in the x and y directions, respectively. cx and cy: x and y principal point coordinates (in pixels), respectively. Multi-axis triangulation: Ideally, multiple lines of sight would converge at a single point, which represents the target's true 3D coordinates. However, due to errors, the lines of sight do not converge precisely. Therefore, the least squares method is used to find the optimal solution.

[0026] For n cameras, the objective function is: .

[0027] Finally, the following points should be noted: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installation", "connection", and "linkage" should be interpreted broadly, and can be mechanical or electrical connections, or internal connections between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may change. Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A calibration method for a target imaging three-dimensional positioning system based on passive optics, characterized in that: The process includes the following steps: S1, Site Deployment: Based on the planned baseline of 500m and the maximum detection distance of 1000m, select multiple photoelectric station installation locations to ensure that the field of view overlap rate of each station is ≥70% and there is no obstruction in the entire area. Use a total station and GPS equipment to accurately measure the three-dimensional geodetic coordinates of each station, record the original location data, and mark the equipment installation base, cable routing, and the placement of switches and processing hosts to complete the overall site layout. S2. Photoelectric hardware installation: At the points marked in step S1, install multiple dual-light photoelectric devices in sequence, adjust the lens orientation to monitor the airspace, fix the device posture, connect power to each device, lay gigabit network cables, ensure that the physical links of all front-end terminals are connected, and perform a power-on self-test on each device to confirm that the visible light imaging, infrared thermal imaging, and image acquisition functions are normal. S3. Global Networking and Time Synchronization Configuration: Connect all front-end dual-optical devices to the PTP time synchronization switch, and then connect the switch to the back-end core processing host to complete the global network networking. Connect the GPS time signal, enable the IEEE1588PTP time synchronization protocol, and provide unified time synchronization for all optoelectronic terminals and the processing host. The synchronization accuracy is calibrated to the nanosecond level. Test the data transmission link to ensure that the image data stream and timestamp data of each device can be stably uploaded to the processing host in real time. S4. Single device participation in global extrinsic calibration: S4.1 Single device internal parameter calibration: The visible light channel and infrared channel of each dual-light optoelectronic device are calibrated one by one. The Zhang Zhengyou calibration method is used to collect multi-view calibration images, calculate the internal parameters of lens focal length, principal point, and distortion coefficient, and complete the single device accuracy calibration. S4.2 Joint calibration of external parameters across the entire field of view: Multiple high-precision three-dimensional control points are set up within the monitoring field of view. All photoelectric devices synchronously acquire images of the control points. Combined with the physical coordinates of the stations measured in step S1, the external parameters of the spatial position and attitude angle of each device are uniformly calculated through the bundle adjustment algorithm. S4.3 saves all internal and external parameters of the devices to the system configuration file as a fixed benchmark for 3D calculation, enabling all optoelectronic devices in the sensing layer to complete unified parameter calibration, forming a unified spatial coordinate system between devices, ensuring the effectiveness of multi-view axis triangulation, and synchronously sending calibration data to the algorithm module to provide a calculation basis for coordinate calculation; S5. Deployment of AI and 3D solution algorithms: Deploy the complete set of algorithm programs in the processing host, including image preprocessing, dual-light image fusion, AI target recognition, sub-pixel feature extraction, multi-view triangulation, and error compensation. Load the device calibration parameters saved in step S4 and the time synchronization parameters in step S3 to complete the matching of algorithm and hardware parameters, and configure the target type library, positioning output format, video encoding rules, and data storage rules. S6. System-wide Joint Operation: The system is powered on and enters automatic operation mode. The photoelectric equipment continuously collects visible light and infrared dual-light images and transmits them back in real time. Standard test flight targets are deployed to simulate the movement of UAVs and birds in the airspace of 100-1000m. The system automatically completes target detection, tracking, and feature extraction. The background observes the 3D positioning results, imaging video, and recognition results. For small errors in long-distance and complex environments, the error compensation model and sub-pixel extraction parameters are fine-tuned. The system is continuously tested for 24 hours to verify its stability, latency, and multi-target processing capabilities.

2. The calibration method for a target imaging three-dimensional positioning system based on passive optics according to claim 1, characterized in that: In step S1, the equipment used includes: a high-precision GPS locator, a laser rangefinder, a total station, a measuring tape, and site marking tools. In step S2, the equipment used includes: a dual-light optoelectronic integrated machine, a fixed bracket, a waterproof junction box, a power supply module, and a network cable; In step S3, the equipment used includes: a PTP precision time synchronization switch, a core processing host, and a GPS timing module. In step S4, the equipment used includes: a high-precision calibration board, standard three-dimensional control points, and a core processing host. In step S5, the equipment used includes: a core processing host and a system operation terminal.

3. The calibration method for a target imaging three-dimensional positioning system based on passive optics according to claim 1, characterized in that: In step S2, the camera of the dual-light optoelectronic device has high resolution and low distortion characteristics, and follows the following trigonometric measurement error formula: Definitions of each symbol: σx: Spatial lateral positioning error Z: Target slant range / detection range b: Camera network baseline length f: Camera equivalent focal length σu: Subpixel extraction accuracy of the target image.

4. The calibration method for a target imaging three-dimensional positioning system based on passive optics according to claim 1, characterized in that: The global network deployment in step S3 shall comply with the following requirements: Number of devices: ≥4 units; Baseline length: 500m; Deployment height: 5-10m; Field of view overlap rate: ≥70%; Coverage range: 120° horizontally × 60° vertically, distance 100-500m.

5. The calibration method for a target imaging three-dimensional positioning system based on passive optics according to claim 1, characterized in that: The intrinsic parameter calibration in step S4 includes: focal length, principal point coordinates, and distortion coefficients, with a calibration accuracy of ≤0.01 pixels. Global extrinsic parameter calibration includes the three-dimensional position and attitude of each camera, requiring a position calibration accuracy of ≤0.1mm and an attitude calibration accuracy of ≤0.001°.

6. The calibration method for a target imaging three-dimensional positioning system based on passive optics according to claim 1, characterized in that: The error compensation model described in step S6 dynamically fine-tunes the sub-pixel extraction parameters and multi-view triangulation results based on measured error data under long distance and complex environments, in order to compensate for positioning deviations caused by atmospheric disturbances and lens distortion.

Images