An air-space reconfigurable polarization-aware interactive system and a recognition method thereof

Abstract

The application discloses a kind of space reconfigurable polarization perception interaction systems and its identification method, belong to circularly polarized photon perception and machine vision collaborative application technical field.System includes space programmable circularly polarized light field coding chip, unmanned aerial vehicle platform, communication and collaborative control unit, ground photon beacon identification terminal and photon coding analysis unit.Chip contains left-handed, right-handed circularly polarized light emitting pixel area and achiral reference area.Ground terminal is through double channel acquisition left-handed and right-handed polarized image, analysis unit carries out double channel brightness fusion calculation with achiral reference area as benchmark, constructs structured chiral light field feature map, realizes chip identification and spatial orientation.The application constructs "air rotation direction signal emission-ground polarization reception-intelligent analysis feedback" full closed loop perception link, utilizes circularly polarized signal rotation direction orthogonal characteristics to carry out background suppression and feature enhancement, improves target identification precision in complex optical environment.

Description

Technical Field

[0001] This invention relates to the intersection of circularly polarized photon sensing technology and machine vision collaborative application technology, and in particular to a reconfigurable polarization interactive sensing system based on spatial deployment of circularly polarized light. Background Technology

[0002] Due to their high mobility and flexible deployment capabilities in three-dimensional space, unmanned aerial vehicle (UAV) platforms have shown broad application potential in complex scene perception, target recognition, and human-computer interaction. However, the optical marking systems deployed by UAVs have long relied on intensity imaging and linear polarization marking. In complex optical interference environments such as low visibility fog and haze, and strong scattering backgrounds, the superposition of non-chiral background scattering and target signals leads to a sharp deterioration in the signal-to-noise ratio, resulting in a target recognition failure rate of over 70%, which severely restricts the reliable deployment of UAVs in high-value perception scenarios.

[0003] Spatial programmable circularly polarized light field encoding chips have a natural advantage over intensity imaging in the perception of rotational brightness features due to the intrinsic chirality of rotational light emission: the depolarization effect of circularly polarized light in atmospheric scattering media is significantly weaker than that of linearly polarized light, the rotational nature does not change with the target rotation angle, and the non-chiral background scattering can be effectively filtered out by the fusion calculation of left / right rotational dual-channel rotational brightness features, and the target signal-to-noise ratio can be improved by 3 to 5 times.

[0004] However, applying high-performance spatially programmable circularly polarized light field encoding chips to practical sensing scenarios faces fundamental constraints: (1) the observation geometry of fixed platform systems is fixed and cannot adapt to dynamic targets and occluded terrain; (2) the field of view of ground systems is limited and cannot achieve omnidirectional rotational light field coverage in three-dimensional space; (3) in scenarios with light interference and occlusion, the line of sight between the fixed ground photonic beacon identification terminal and the ground-borne chip is often cut off, causing the system to fail. The above constraints cannot be overcome simply by increasing the chip's rotational anisotropy factor. The fundamental reason is the lack of a motion platform that can dynamically control the chip's spatial position and radiation direction.

[0005] Deeply integrating a UAV platform with a space-programmable circularly polarized optical field encoding chip, where the UAV provides the motion platform, the chip provides high-performance rotational coded optical field signals, and a ground-based photonic beacon identification terminal performs precise rotational analysis, the three work together to form a complete perception closed loop. This is the fundamental way to overcome all the aforementioned constraints. Currently, no such system integration scheme has been reported in existing literature and patents. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a spatially reconfigurable polarization sensing interactive system and its recognition method based on a mobile chiral light field. It aims to solve technical problems such as fixed observation geometry, limited field of view, and easy obstruction of the line of sight in fixed ground systems, and to realize dynamic rotation sensing with reconfigurable line of sight in three-dimensional space.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A spatially reconfigurable polarization sensing interactive system, the system comprising: a spatially programmable circularly polarized light field encoding chip, an unmanned aerial vehicle platform, a communication and collaborative control unit, a ground photonic beacon identification terminal, and a photonic encoding and parsing unit;

[0009] The spatially programmable circularly polarized light field encoding chip includes a left-handed circularly polarized emission pixel region, a right-handed circularly polarized emission pixel region, and a non-chiral reference region with a rotational anisotropy factor of zero. The rotational anisotropy factor of each pixel is |g lum Not less than 0.5;

[0010] Furthermore, in the spatially programmable circularly polarized light field encoding chip, each pixel has a globally vertically ordered nanostructure, with the interlayer spacing corresponding to half the helical pitch of the chiral liquid crystal; the lateral aperture of the pixel is 50 to 129 μm, the dome height is approximately 17.60 μm, and the aspect ratio is approximately 0.29; the helical anisotropy factor |g lum | Up to 1.01.

[0011] Preferably, the light field encoding chip includes liquid crystal unit LC013, crosslinking agent EDDET, polymerization inhibitor BHT, photoinitiator I-651, toluene, chiral dopant R-5011 or S-5011, and chromophores 4-(dicyanomethylene)-2-methyl-6-(4-dimethylaminostyryl)-4H-pyran, coumarin 6, and 7-hydroxycoumarin.

[0012] Furthermore, in the spatially programmable "warning" circularly polarized light field encoding chip, the left-handed region is located in the warning pattern area on the outer periphery of the chip (g). lum > 0), the right-handed region is located in the triangular region at the center of the chip (g lum < 0), the non-chiral reference mark region is located in the triangular region in the upper right corner of the chip (the rotation anisotropy factor is zero); the three regions emit the same color, only the rotation is different; the effective area of ​​the chip is not less than 3 cm × 3 cm.

[0013] Furthermore, the UAV platform mounts the directional coding circularly polarized photonic chip on its fuselage and combines it with an ultraviolet excitation module, so that the photonic chip serves as a mobile directional photonic signal source for maneuvering flight.

[0014] Preferably, the drone is selected from the Potenslc ATOM 2 Drone DSDR23A.

[0015] Furthermore, the communication and coordination control unit is used to coordinate the flight trajectory of the UAV platform and the pointing control of the remote observation terminal. The UAV's maneuvering flight overcomes terrain obstructions, providing a line-of-sight reconfigurable mobile rotation perception capability that is impossible with fixed ground systems, enabling dynamic photon beacon tracking, target confirmation, and real-time information exchange in complex environments;

[0016] Preferably, the communication and collaboration control unit is selected from Potenalc PT1 Remote Controller DSRC23A.

[0017] Furthermore, the ground-based photonic beacon identification terminal includes a first polarization sensing channel (equipped with a left-handed circular polarization analyzer) and a second polarization sensing channel (equipped with a right-handed circular polarization analyzer), which respectively acquire left-handed polarization images I from the photonic chip. L With right-handed polarization image I R ;

[0018] Furthermore, the two polarization sensing channels of the ground photonic beacon identification terminal are equipped with synchronous image sensors;

[0019] Preferably, the first and second polarization image sensors are luminance meters, which are CCD / CMOS cameras used to capture light intensity signals in the direction of the chip plane.

[0020] Preferably, the CMOS / CCD camera has a resolution greater than 1920×1080.

[0021] Preferably, the operating wavelength of the left-hand and right-hand circular polarization analyzer is selected from 350-1500 nm.

[0022] Furthermore, the photonic coding and parsing unit uses the non-chiral reference region signal as the real-time background normalization benchmark to fuse and calculate C(x,y)=I from optical data from multiple polarization response channels. L -I R A structured photonic feature map with background suppression capability is constructed to achieve high-fidelity recognition and spatial positioning of the axially coded circularly polarized photonic chip. This refers to the image intensity value at the left-handed corresponding pixel position of the polarization sensing channel. This represents the image intensity value at the right-handed corresponding pixel position of the polarization sensing channel.

[0023] Preferably, the photonic coding system employs a pseudo-color coding method to map brightness values ​​into a visual color image.

[0024] Furthermore, the photonic coding and parsing unit has a depth information extraction function. Using the epipolar-corrected left-eye view as a spatial reference, the photonic coding and parsing unit performs stereo matching and fusion calculations on the left and right image signals from the binocular sensors. We construct a parallax feature map containing depth information to achieve high-fidelity 3D reconstruction and spatial positioning of the target scene. Let (x, y) be the horizontal coordinate of the point in the right image that matches the pixel (x, y) in the left image. The horizontal coordinates of the matching point itself;

[0025] Preferably, the photonic coding system employs a pseudo-color coding method.

[0026] Furthermore, the photonic coding and parsing unit uses the image signal intensity of the non-chiral reference region (where the rotational anisotropy factor is zero) as the real-time background normalization benchmark to eliminate the influence of ambient light fluctuations and detector response drift on polarization imaging quality; polarization operations, in principle, filter out the contribution of non-chiral background scattering ( 背景 ).

[0027] Preferably, data transmission is performed using one or more of the following interfaces: USB 3.0 and above, high-speed Thunderbolt, and GigE.

[0028] Furthermore, the UAV-cooperative aerial rotation identification method of the system includes the following steps:

[0029] S1. Deploy the rotation-encoded circularly polarized photonic chip on a drone platform;

[0030] S2. Initiate the drone to maneuver and fly over the target scene, providing a line-of-sight reconfigurable mobile rotation photon signal source;

[0031] S3, Ground Photon Beacon Identification Terminal Synchronous Acquisition L with I R ;

[0032] S4. Perform normalization correction based on the non-chiral reference region signal, and execute the characteristic brightness calculation C(x,y)=I. L -I R To obtain a rotation-selective polarization image;

[0033] S5. Based on the spatial distribution of the left-handed spiral region (C < 0), the right-handed spiral region (C > 0), and the non-chiral reference region (|C|≈0) in the polarization image, the spiral direction encoding information carried by the decoding chip is used to realize target spiral direction recognition and positioning.

[0034] Furthermore, the system utilizes a deep learning target perception model to perform deep localization of the chip region in the polarization image, enabling high-precision detection of the target region in complex optical interference environments such as urban building background scattering and low visibility.

[0035] Furthermore, in search and rescue operations with low visibility and light interference, the target personnel or equipment carry the aforementioned rotation-coded circularly polarized photonic chip as a rotation-coded identification identifier. The UAV can achieve target rotation identification and positioning in scenarios with visibility as low as 8 to 12 m through dual-channel polarization rotation sensing.

[0036] This invention has the following advantages and beneficial effects:

[0037] 1. Reconfigurable dynamic perception of airspace: The maneuvering flight of UAVs breaks through the line-of-sight obstruction of fixed ground systems, realizing reconfigurable dynamic perception of three-dimensional airspace, and achieving deep integration of airspace maneuvering platform and circular polarization signal perception.

[0038] 2. Robustness to strong light interference: The dual-channel rotational brightness feature fusion calculation suppresses non-chiral background scattering in principle, effectively suppressing interference from natural light, artificial light sources and other non-circularly polarized light in the environment, improving the target signal-to-noise ratio by 3 to 5 times compared to single-channel intensity imaging.

[0039] 3. High-fidelity real-time reconstruction of photon spin information: Based on pixel-by-pixel spin orientation feature extraction, full-field polarization decoding, and pseudo-color mapping algorithms, real-time visualization and high-precision spatial reconstruction of circular polarization information are achieved. Combined with the dynamic scanning capabilities of the UAV platform, a large-scale, high-resolution, spin-resolved three-dimensional polarization information map can be constructed.

[0040] 4. Complete Perception Interaction Closed Loop: Establishing an intelligent perception closed loop of "transmission-transmission-perception-decoding-feedback". A spatially programmable circularly polarized optical field encoding chip serves as the active information transmitter, while the UAV acts as a mobile perception node in the airspace. Combined with a deep learning spin pattern recognition network, it achieves autonomous perception, intelligent decoding, and real-time decision-making of circularly polarized information. Even under conditions of high background noise, dynamic occlusion, and complex urban environments, it can still achieve high-accuracy target identification and positioning, providing a new technological path for intelligent unmanned systems, spin photonics communication, and next-generation human-machine collaborative perception. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the reconfigurable polarization sensing interactive system based on a mobile chiral light field for spatial deployment, obtained in Embodiment 1 of the present invention.

[0042] Figure 2 This is a planar layout diagram of the spatial rotation programming pattern of the rotation-encoded circularly polarized photonic chip provided in Embodiment 1 of the present invention;

[0043] Figure 3 A real-world photograph of a UAV deploying a rotation-coded circularly polarized photonic chip, as provided in Embodiment 1 of the present invention.

[0044] Figure 4 The following is a city background experimental verification diagram for the UAV flight experiment provided in Embodiment 2 of the present invention: Figure 4 In the image, 'a' represents the left-hand circular polarization channel image (I). L ), Figure 4 In the image, b represents the right-hand circular polarization channel image (I). R ), Figure 4 c in the figure represents the normalized characteristic brightness (rotation map), showing the high-contrast rotation resolution results of the left-rotational warning area, the right-rotational triangular area, and the non-chiral reference area;

[0045] Figure 5 This is a result image of depth-sensing imaging provided in Embodiment 2 of the present invention. Detailed Implementation

[0046] Several embodiments of this application will now be described in more detail with reference to the accompanying drawings to enable those skilled in the art to implement this application. This application may be embodied in many different forms and for various purposes and should not be limited to the embodiments set forth herein. These embodiments are provided to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. The embodiments described do not limit this application.

[0047] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having a meaning consistent with their meaning in the relevant field and / or the context of this specification, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0048] Example 1: Construction of a Spatial Reconfigurable Perceptual Interaction System Based on a Mobile Chiral Light Field

[0049] This invention provides a method for constructing a reconfigurable polarization-sensing interactive system for spatial deployment of mobile chiral light fields. The system includes the following components: Figure 1 ):

[0050] (1) Spatial programmable circular polarization light field encoding chip: Spatial rotation programming "warning" photonic chip (left-handed central warning area) Right-handed outer triangular area The chiral reference region has a zero spin anisotropy factor; the chip is 3 cm × 3 cm; and the emission peak is approximately 600 nm. Figure 3 ).

[0051] (2) Ultraviolet excitation module: an airborne miniature 365 nm UV LED module (power 80mW, mass about 5 g) is fixed next to the chip fixing frame to perform near-field excitation on the chip.

[0052] (3) Unmanned Aerial Vehicle Platform: Commercial quadcopter UAV (wheelbase 218 mm). The aircraft camera can collect aerial flight field of view. The spatially programmable circularly polarized light field encoding chip is fixed under the fuselage, with the light-emitting surface facing the ground photonic beacon identification terminal.

[0053] (4) Communication and Coordination Control Unit (Potenalc PT1 Remote Controller): Used to coordinate the flight trajectory of the UAV platform and the pointing control of the remote observation terminal. The UAV's maneuvering flight breaks through terrain obstruction and provides a line-of-sight reconfigurable mobile rotation perception capability that cannot be achieved by fixed ground systems, enabling dynamic photon beacon tracking, target confirmation, and real-time information exchange in complex environments.

[0054] (5) Photonic beacon identification terminal: The tripod is installed about 10 m from the center of the flight area. The left-hand circular polarization analysis channel uses a Thorlabs CP1L532 left-hand circular polarizer, and the right-hand circular polarization analysis channel uses a Thorlabs CP1R532 right-hand circular polarizer. The two-channel imaging sensors (CMOS, 1920 × 1200 pixels, 50 mm focal length lens) are configured for synchronous exposure.

[0055] In this embodiment, the specific steps of the photonic coding parsing unit are as follows:

[0056] A) Camera Calibration: 20 sets of checkerboard images (10×7 corner points) were acquired using OpenCV's stereoCalibrate function. The intrinsic matrix (focal length approximately 667 pixels, principal point coordinates approximately (320, 180)), extrinsic matrix (rotation matrix R≈I, translation vector T=[10, 0, 0]T mm), and distortion coefficients (radial distortion) were calculated for the left and right channels. , ).

[0057] B) Noise reduction: Adaptive Gaussian filtering (kernel size 5×5, σ=1.5) is used to adjust the filtering intensity according to the local gradient of the image, preserving edge details.

[0058] C) Grayscale conversion: Convert the RGB image into a double-type grayscale matrix (weights: R: 0.299, G: 0.587, B: 0.114).

[0059] D) Image registration: Feature points are extracted using the SIFT algorithm and matched using the ORB algorithm (number of key points ≥ 1,000 points / image). Pixel-level alignment of the left and right channels is achieved through affine transformation (horizontal displacement Δx ≈ 15 pixels, rotation angle θ ≈ 0.5°).

[0060] In this embodiment, the rotation characteristic brightness is calculated as: C(x,y)=I L -I R Background normalization correction is performed using the non-chiral reference region image signal as a real-time reference.

[0061] In this embodiment, the rotation pattern is reconstructed by mapping the rotation feature brightness values ​​to the Jet color table (deep blue: C). min Crimson:C max ), generate a circular polarization characteristic brightness map.

[0062] Communication and Collaborative Control: The digital image transmission system (delay < 80 ms) transmits the real-time position and attitude information of the UAV to the ground station. The ground station adjusts the aiming direction of the sensing station in real time according to the position of the UAV. The data transmission is via USB 3.0 or higher interface.

[0063] like Figure 1 , Figure 3 As shown, the above components work together to form a complete airborne rotation sensing system.

[0064] Example 2

[0065] Unmanned aerial vehicle (UAV) collaborative aerial rotation recognition flight experiment verification

[0066] Flight parameters: Flight altitude approximately 1-30 m, flight speed approximately 5 m / s, flight area is urban building area (surrounding buildings are 10 to 30 m high, glass curtain walls form complex background scattering); dual-channel acquisition frame rate 30 fps, exposure time 20ms.

[0067] like Figure 4 As shown, the left and right spiral directions in the spiral pattern show a significant difference in brightness compared to the non-chiral reference region. Spiral direction was identified frame-by-frame at flight speeds ≤ 5 m / s; as... Figure 5 As shown, under the conditions of non-chiral background light intensity and urban background scattering, the system can only resolve circularly polarized light, but cannot resolve interference light, which verifies the robust rotational orientation recognition capability of the system under complex urban optical backgrounds.

[0068] Example 3

[0069] Verification of Relative Distance and Depth Perception between UAVs and Ground Terminals Based on Imaging Geometry

[0070] This embodiment verifies the ability of the depth perception module in the photonic coding and parsing unit to calculate the relative distance between the UAV and the ground photonic beacon identification terminal in real time.

[0071] The ground-based photonic beacon identification terminal completes calibration according to step A of Example 1, calibrating the focal length parameters. The physical dimensions of the spatially programmable circularly polarized optical field encoding chip are 3 cm × 3 cm.

[0072] The drone equipped with the photonic chip was hovered 5 meters above the ground-based photonic beacon identification terminal. The ground-based photonic beacon identification terminal synchronously collected I... L with I R The image, after being processed by an automatic rotation pattern decoding algorithm, identifies the imaging boundary of the photonic chip in the polarization image, and measures the chip imaging side length and number of pixels w; the depth perception module, based on... Calculate relative distance .

[0073] like Figure 5 As shown, when the drone hovers at a distance of 5m, the depth of the photonic chip imaging is inconsistent with that of the drone and the background, and is displayed in pseudo-color.

[0074] The depth perception module works in conjunction with the rotation pattern automatic decoding algorithm: the rotation pattern automatic decoding algorithm analyzes the identity and orientation information carried by the photonic chip, and the depth perception module simultaneously provides the relative distance between the UAV and the ground terminal. Together, they constitute an integrated perception result of the UAV's spatial orientation (orientation, identity) and spatial distance, which can be applied to UAV hovering distance monitoring, close-range human-machine collaborative avoidance, and integrated positioning feedback of target orientation and distance in search and rescue scenarios.

[0075] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A spatially reconfigurable polarization sensing interactive system, characterized in that, include: A spatially programmable circularly polarized light field encoding chip includes a left-handed circularly polarized emission pixel region, a right-handed circularly polarized emission pixel region, and a non-chiral reference region with a zero rotational anisotropy factor. The rotational anisotropy factor of each pixel is |g lum Not less than 0.5; The unmanned aerial vehicle (UAV) platform is used to mount the spatially programmable circularly polarized light field encoding chip on the fuselage and combine it with an ultraviolet excitation module, so that the spatially programmable circularly polarized light field encoding chip can be used as a mobile directional photon signal source for maneuvering flight. A communication and coordination control unit is used to coordinate the flight trajectory of the UAV platform with the pointing control of the remote observation terminal; The ground-based photonic beacon identification terminal includes a first polarization sensing channel equipped with a left-handed circular polarization analyzer and a second polarization sensing channel equipped with a right-handed circular polarization analyzer, which respectively acquire left-handed polarization images I from the spatially programmable circular polarization optical field encoding chip. L With right-handed polarization image I R ; The photonic coding and parsing unit is used to perform fusion calculation C(x,y)=I on optical data from the first polarization sensing channel and the second polarization sensing channel, using the signal from the non-chiral reference region as a real-time background normalization reference. L -I R A structured photon feature map is constructed to enable the identification and spatial positioning of the spatially programmable circularly polarized light field encoding chip.

2. The system according to claim 1, characterized in that, The spatially programmable circularly polarized optical field encoding chip is composed of pixels, each pixel having a globally vertically ordered nanostructure, with interlayer spacing corresponding to half the helical pitch of a chiral liquid crystal; the lateral aperture of the pixel ranges from 50 μm to 129 μm, the dome height is 17.60 μm, and the aspect ratio is 0.29; the helical anisotropy factor |g lum | Up to 1.

01.

3. The system according to claim 1, characterized in that, The pixels of the spatially programmable circularly polarized light field encoding chip can be edited into any pattern; the effective area of ​​the chip is not less than 3cm × 3cm, and the total mass does not exceed 5g.

4. The system according to claim 3, characterized in that, The spatially programmable circularly polarized light field encoding chip is shaped like a warning triangle; the left-handed circularly polarized light-emitting pixel area is located in the warning graphic area on the periphery of the chip, the right-handed circularly polarized light-emitting pixel area is located in the central triangular area of ​​the chip, and the non-chiral reference area is located in the triangular area at the upper right corner of the chip; the three areas emit the same light color, only differing in rotation.

5. The system according to claim 1, characterized in that, The communication and coordination control unit is used to coordinate the real-time flight attitude of the UAV platform with the observation direction of the ground photonic beacon identification terminal, so that the UAV platform can overcome terrain obstruction during maneuvering flight and provide a line-of-sight reconfigurable motion rotation perception capability.

6. The system according to claim 1, characterized in that, The first polarization sensing channel and the second polarization sensing channel of the ground photonic beacon identification terminal are equipped with synchronous image sensors; the synchronous image sensors are CCD cameras or CMOS cameras, used to capture light intensity signals in the direction of the chip plane.

7. The system according to claim 1, characterized in that, The photonic coding and parsing unit uses the image signal intensity of the non-chiral reference region as a real-time background normalization benchmark to eliminate the influence of ambient light fluctuations and detector response drift on polarization imaging quality; non-chiral background scattering contribution 背景 ≈0; The photonic coding and parsing unit also has a depth calculation function; the photonic coding and parsing unit uses the epipolar-corrected left-eye view as a spatial reference to perform stereo matching and fusion calculation on the left and right image signals from the binocular sensor. A disparity feature map is constructed to achieve 3D reconstruction and spatial positioning of the target scene; among which, Let (x, y) be the horizontal coordinate of the matching point in the right image. The horizontal coordinates of the matching point itself; The photonic coding and parsing unit is also equipped with an automatic rotation pattern decoding algorithm, which is used to automatically identify the left-handed circularly polarized light emission pixel region, the right-handed circularly polarized light emission pixel region, and the non-chiral reference region of the spatially programmable circularly polarized light field coding chip in the polarization image, and display them with color patterns of different brightness.

8. A method for UAV cooperative aerial rotation identification based on the system described in any one of claims 1 to 7, characterized in that, Includes the following steps: S1. Deploy the spatially programmable circularly polarized optical field encoding chip on the UAV platform; S2. Initiate the drone to maneuver and fly over the target scene, providing a line-of-sight reconfigurable mobile rotation photon signal source; S3, Ground photon beacon identification terminal synchronously acquires left-handed polarization image I L With right-handed polarization image I R ; S4. Perform normalization correction based on the non-chiral reference region signal, and execute the luminance feature operation C(x,y)=I. L -I R To obtain a rotation-selective polarization image; S5. Based on the spatial distribution of the left-hand circularly polarized emission pixel region, the right-hand circularly polarized emission pixel region, and the non-chiral reference region in the direction-selective polarization image, the direction encoding information carried by the decoding chip is used to realize target direction recognition and positioning.

9. The method according to claim 8, characterized in that, Step S4 also includes using a deep learning target perception model to detect and locate the chip region in the polarization image.

10. The application of the method according to claim 8 in search and rescue operations under low visibility and light interference conditions, characterized in that, The target personnel or equipment carry the spatially programmable circularly polarized light field encoding chip as a rotational encoding identification. The UAV can achieve target rotational identification and positioning in scenarios with visibility as low as 8m to 12m through dual-channel polarization rotational perception.

Images