Multi-area focusing visual navigation and reconnaissance system based on polarization correlated imaging and super-structure lens

Abstract

The invention discloses a multi-area focusing visual navigation and reconnaissance system based on polarization correlated imaging and a super-structure lens. The multi-area focusing visual navigation and reconnaissance system comprises a polarization imaging module, a super-structure lens module, a feature extraction module and a navigation and reconnaissance control module, the polarization imaging module is used for acquiring light intensity images of a to-be-reconnaissance target at different polarization angles, and processing the light intensity images by using a linear polarization degree method to obtain polarization images; the super-structure lens module is used for performing dynamic focusing to obtain a focused initial image, and performing contrast enhancement on the initial image to obtain a focused image; the feature extraction module is used for performing feature extraction on the polarized image and the focused image to obtain feature information of the to-be-reconnaissance target; and the navigation and reconnaissance control module performs path planning based on the feature information, and performs real-time optimization on the planned path according to environmental changes to obtain an optimized path, thereby completing navigation.

Description

Technical Field

[0001] The present invention belongs to the technical field of visual reconnaissance, and specifically relates to a multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens. Background Art

[0002] With the booming development of autonomous driving technology, drone applications, and robot intelligent systems, visual navigation technology has become the key to promoting the high autonomy and reliability of these systems. Especially in complex and variable environments such as urban blocks, the navigation system needs to have extremely high environmental perception and adaptation capabilities. However, traditional navigation technologies often appear powerless when faced with challenges such as line-of-sight occlusion caused by high-rise buildings, interference from variable lighting conditions, and target recognition in dynamic traffic conditions, resulting in a decline in perception accuracy and limited decision-making ability.

[0003] As an emerging visual perception method, polarization imaging technology can reveal the subtle differences between targets and backgrounds by capturing and analyzing the polarization state of light waves, providing richer scene information than traditional imaging. This technology has demonstrated its unique advantages and broad application potential in multiple fields such as remote sensing monitoring, precise target recognition, and biomedical imaging. Especially in complex environments, polarization imaging can effectively suppress the interference of reflected and scattered light, improve the contrast and clarity of images, and thus enhance the environmental perception ability of the system.

[0004] At the same time, as the latest breakthrough in the field of optics, metasurface lens technology realizes precise control of the light wave propagation path through nanoscale structural design, enabling the lens to achieve dynamic focusing on different regions. The application of this technology not only improves the resolution and contrast of the imaging system but also provides a new technical means for target detection and recognition in complex environments. Through the focusing ability of the metasurface lens, the system can achieve clear imaging of targets at different distances, further enhancing the performance of the navigation and reconnaissance system.

[0005] In view of this, the present invention patent proposes a multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens, aiming to overcome the limitations of traditional technologies in complex environments, improve the navigation and reconnaissance efficiency, and provide strong technical support for applications in fields such as autonomous driving vehicles, drones, and security monitoring. Summary of the Invention

[0006] The present invention aims to solve the deficiencies of the existing technology and provides the following solutions:

[0007] A multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens, comprising: a polarization imaging module, a metasurface lens module, a feature extraction module, and a navigation and reconnaissance control module;

[0008] The polarization imaging module is used to obtain intensity images of a target to be reconnoitered at different polarization angles, and processes the intensity images by using the degree of linear polarization method to obtain polarization images;

[0009] The metasurface lens module is used to perform dynamic focusing to obtain an initial focused image, and enhances the contrast of the initial image to obtain a focused image;

[0010] The feature extraction module is used to extract features from the polarization image and the focused image to obtain the feature information of the target to be reconnoitered;

[0011] The navigation and reconnaissance control module performs path planning based on the feature information, and optimizes the planned path in real time according to environmental changes to obtain an optimized path, thereby completing navigation.

[0012] Preferably, the polarization imaging module includes: a linear polarization grid, a photosensitive element, an imaging target surface, a camera housing, and a first image processing unit;

[0013] The linear polarization grid and the photosensitive element are both disposed on the imaging target surface, and the imaging target surface is disposed inside the camera housing;

[0014] The linear polarization grid takes different-direction grids of a 2×2 array as a unit, and each unit corresponds to a 2×2 array of photosensitive elements;

[0015] The incident light of the target to be reconnoitered is converted into linearly polarized light with polarization information in two directions after passing through the linear polarization grid, and is incident on the photosensitive element to obtain the intensity image;

[0016] The first image processing unit processes the intensity image by using the degree of linear polarization method to obtain the polarization image.

[0017] Preferably, the degree of linear polarization method includes:

[0018]

[0019] Among them, DOLP represents the degree of linear polarization, AOP represents the polarization angle, I0 represents the intensity image of horizontally polarized light, I 45 represents the intensity image of 45° polarized light, I 90 represents the intensity image of vertically polarized light, I 135 represents the intensity image of 135° polarized light.

[0020] Preferably, the metasurface lens module includes: a focusing adjustment unit, a metasurface lens, and a second image processing unit;

[0021] The focusing adjustment unit is used to adjust the focusing area of the metasurface lens according to the reconnaissance task;

[0022] The meta-lens is used to obtain an initial image after focusing;

[0023] The second image processing unit is used to focus and contrast enhance the initial image, and remove image blur using an image processing algorithm to obtain the focused image.

[0024] Preferably, the feature extraction module includes: an image recognition unit and a feature extraction unit;

[0025] The image recognition unit is used to perform target recognition on the polarization image and the focused image using a pre-trained deep learning model to obtain a recognition result of the reconnaissance target;

[0026] The feature extraction unit extracts the feature information of the target to be detected based on the recognition result.

[0027] Preferably, the navigation and reconnaissance control module includes: a path planning unit and a strategy optimization unit;

[0028] The path planning unit locates and tracks the target to be reconnaissance based on the feature information, and performs path planning to obtain a navigation path;

[0029] The strategy optimization unit optimizes the planned navigation path in real time according to environmental changes, obtains the optimized path, and completes navigation.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] (1) The present invention combines polarization imaging and meta-lens technology to effectively suppress reflected and scattered light in urban block environments, significantly improving the contrast and clarity of imaging. Polarization imaging technology distinguishes targets from backgrounds by analyzing the polarization state of light waves, while meta-lens achieves clear focus on targets at different distances by precisely controlling the light wavefront, thereby providing high-quality images under complex lighting and weather conditions.

[0032] (2) The present invention uses polarization imaging technology to extract information about the material and surface features of an object, combined with the multi-region focusing capability of the meta-lens, to achieve accurate imaging of targets at different distances and near distances. This combination of technologies improves the accuracy of target detection and recognition, allowing the system to more effectively identify and locate targets during visual navigation and reconnaissance, and maintain high accuracy even in urban environments with more visual interference.

[0033] (3) The present invention is equipped with a high-performance image processing unit, and by adopting optimized algorithms and hardware acceleration technologies, it can process and analyze a large amount of polarization image data in real time. This real-time data processing ability ensures the rapid response of the system in a dynamic environment, provides timely navigation and reconnaissance decision-making support for autonomous driving vehicles, drones, etc., and enhances the practicality and application value of the system. Brief Description of the Drawings

[0034] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0035] Figure 1 Schematic diagram of the system structure of the embodiment of the present invention;

[0036] Figure 2 Schematic diagram of the structure of the polarization imaging module of the embodiment of the present invention;

[0037] Figure 3 Schematic diagram of the code of the polarization imaging module of the embodiment of the present invention;

[0038] Figure 4 Schematic diagram of the code of the metasurface lens module of the embodiment of the present invention;

[0039] Figure 5 Schematic diagram of the code of the feature extraction module of the embodiment of the present invention;

[0040] Figure 6 Schematic diagram of the code of the navigation and reconnaissance control module of the embodiment of the present invention;

[0041] Explanation of the reference numerals in the drawings:

[0042] 1, incident light; 2, linearly polarized light; 3, linear polarization grid; 4, photosensitive element; 5, imaging target surface; 6, camera housing. Detailed Embodiments

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0044] To make the above-mentioned objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below in conjunction with the drawings and specific embodiments.

[0045] Embodiment

[0046] In this embodiment, as Figure 1 shown, the multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens includes: a polarization imaging module, a metasurface lens module, a feature extraction module, and a navigation and reconnaissance control module.

[0047] The polarization imaging module is used to obtain the intensity images of the target to be reconnoitered at different polarization angles, and process the intensity images by using the degree of linear polarization method to obtain the polarization image.

[0048] The polarization imaging module, as Figure 2 shown, includes: a linear polarization grid 3, a photosensitive element 4, an imaging target surface 5, a camera housing 6, and an image processing unit. The linear polarization grid 3 and the photosensitive element 4 are both arranged on the imaging target surface 5, and the imaging target surface 5 is arranged inside the camera housing 6; the linear polarization grid 3 takes the different-direction grids of a 2×2 array as a unit, and each unit corresponds to the photosensitive element 4 of the 2×2 array; the incident light 1 of the target to be reconnoitered is converted into linearly polarized light 2 with polarization information in four directions after passing through the linear polarization grid 3, and is incident on the photosensitive element 4 to obtain the intensity image; the first image processing unit processes the intensity image by using the degree of linear polarization method to obtain the polarization image. Among them, the intensity images include: the intensity image of horizontal polarization, the intensity image of 45° polarization, the intensity image of vertical polarization, and the intensity image of 135° polarization.

[0049] In this embodiment, assuming the polarization light conditions in the urban environment, the degree of linear polarization method includes:

[0050]

[0051] Among them, DOLP represents the degree of linear polarization, AOP represents the polarization angle, I0 represents the intensity image of horizontal polarization, I 45 represents the intensity image of 45° polarization, I 90 represents the intensity image of vertical polarization, I 135 represents the intensity image of 135° polarization.

[0052] As Figure 3 shown, this code defines a class PolarizedImagingModule for polarization imaging processing. The main function of this class is to load polarization images at different angles and calculate polarization-related parameters, specifically including the degree of linear polarization and the polarization angle.

[0053] The metasurface lens module is used to perform dynamic focusing to obtain the initial focused image, and enhance the contrast of the initial image to obtain the focused image.

[0054] The metasurface lens module includes: a focusing adjustment unit, a metasurface lens, and a second image processing unit. The focusing adjustment unit is used to adjust the focusing area of the metasurface lens according to the reconnaissance mission; the metasurface lens is used to obtain the initial image after focusing; the second image processing unit is used to perform focusing and contrast enhancement on the initial image, and use image processing algorithms to remove image blur to obtain a focused image.

[0055] As Figure 4 shown, the image processing algorithms include: using median filtering and Gaussian filtering to eliminate noise. Calculating the target distance according to information such as image dispersion, calculating the Laplacian operator of the image, and sharpening and gray-level stretching to enhance the image contrast.

[0056] The feature extraction module is used to extract features from the polarization image and the focused image to obtain the feature information of the target to be reconnoitered.

[0057] The feature extraction module includes: an image recognition unit and a feature extraction unit. The image recognition unit is used to perform target recognition on the polarization image and the focused image using a pre-trained deep learning model to obtain the recognition result of the reconnaissance target. In this embodiment, the deep learning model can be a convolutional neural network, and the model can recognize the target to be reconnoitered in the image, the edges, textures, etc. of the target; the feature extraction unit extracts the feature information of the target to be reconnoitered based on the recognition result. In this embodiment, the feature information includes: the shape, size, color, texture, etc. of the target to be reconnoitered.

[0058] As Figure 5 shown, the main function of the convolutional neural network is to perform image processing and feature extraction using a pre-trained VGG16 model. The convolutional neural network includes: 13 convolutional layers, using 3*3 convolutional kernels stacked to extract features. Each convolutional layer is equipped with a ReLU activation function to introduce non-linearity into the network learning. After every 2, 2, 3, 3 convolutional layers, a pooling layer is equipped to downsample the image and reduce the data volume. Then a fully connected layer is equipped to comprehensively process the feature information by custom neurons.

[0059] The navigation and reconnaissance control module performs path planning based on the feature information, and real-time optimizes the planned path according to environmental changes to obtain the optimized path and complete navigation.

[0060] The navigation and reconnaissance control module includes: a path planning unit and a strategy optimization unit. The path planning unit locates and tracks the target to be reconnoitered based on the feature information and performs path planning to obtain a navigation path; the strategy optimization unit real-time optimizes the planned navigation path according to environmental changes (such as weather, light, obstacles, etc.) to obtain the optimized path and complete navigation.

[0061] As Figure 6As shown, the navigation and control unit class is used for automatic navigation control. Combining imaging technology, target recognition, and path planning, it can guide the system to navigate safely and efficiently from the current position to the target position. First, through image processing and feature extraction, the target position is recognized and target tracking is performed to ensure that the system always moves towards the target. By inputting the current position and the target position, the plan_path method will plan a navigation path from the current position to the target, providing a reference for subsequent movement. In addition, the system will also continuously monitor the obstacles on the path through imaging and make detour decisions based on the situation of the obstacles to avoid collisions and ensure safety during driving. It not only realizes simple path planning but also can dynamically adjust the path to cope with real-time environmental changes, and is applicable to application scenarios that require intelligent navigation and obstacle avoidance.

[0062] The embodiments described above are only descriptions of the preferred embodiments of the present invention and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens, characterized in that Including: A polarization imaging module, a metasurface lens module, a feature extraction module, and a navigation and reconnaissance control module; The polarization imaging module is used to obtain intensity images of a target to be reconnoitered at different polarization angles, and process the intensity images by using the degree of linear polarization method to obtain polarization images; The metasurface lens module is used to perform dynamic focusing to obtain an initial focused image, and enhance the contrast of the initial image to obtain a focused image; The feature extraction module is used to extract features from the polarization image and the focused image to obtain feature information of the target to be reconnoitered; The navigation and reconnaissance control module performs path planning based on the feature information, and real-time optimizes the planned path according to environmental changes to obtain an optimized path and complete navigation.

2. The multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens according to claim 1, wherein The polarization imaging module includes: a linear polarization grid, a photosensitive element, an imaging target surface, a camera housing, and a first image processing unit; The linear polarization grid and the photosensitive element are both arranged on the imaging target surface, and the imaging target surface is arranged inside the camera housing; The linear polarization grid takes different-direction grids of a 2×2 array as a unit, and each unit corresponds to a 2×2 array of photosensitive elements; The incident light of the target to be reconnoitered is converted into linearly polarized light with polarization information in a certain direction after passing through the linear polarization grid, and is incident on the photosensitive element to obtain the intensity image; The first image processing unit processes the intensity image by using the degree of linear polarization method to obtain the polarization image.

3. The multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens according to claim 2, wherein The degree of linear polarization method includes: Among them, DOLP represents the degree of linear polarization, AOP represents the polarization angle, I0 represents the light intensity image of horizontal polarization, I 45 represents the light intensity image of 45° polarization, I 90 represents the light intensity image of vertical polarization, I 135 represents the light intensity image of 135° polarization.

4. The multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens according to claim 1, characterized in that The metasurface lens module includes: a focusing adjustment unit, a metasurface lens, and a second image processing unit; The focusing adjustment unit is used to adjust the focusing area of the metasurface lens according to the reconnaissance task; The metasurface lens is used to obtain an initial focused image; The second image processing unit is used to perform focusing and contrast enhancement on the initial image, and remove image blurring by using an image processing algorithm to obtain the focused image.

5. The multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens according to claim 1, characterized in that The feature extraction module includes: an image recognition unit and a feature extraction unit; The image recognition unit is used to perform target recognition on the polarization image and the focused image by using a pre-trained deep learning model to obtain the recognition result of the reconnaissance target; The feature extraction unit extracts the feature information of the target to be reconnoitered based on the recognition result.

6. The multi-region focusing visual navigation and reconnaissance system based on polarization correlation imaging and metasurface lens according to claim 1, characterized in that The navigation and reconnaissance control module includes: a path planning unit and a strategy optimization unit; The path planning unit performs positioning and tracking on the target to be reconnoitered based on the feature information and performs path planning to obtain a navigation path; The strategy optimization unit real-time optimizes the planned navigation path according to environmental changes to obtain an optimized path and complete navigation.

Images