Method, device and equipment for collecting billion-pixel multi-scene dynamic image and storage medium

Abstract

The invention discloses a hundred-million-pixel multi-scene dynamic image acquisition method and device, equipment and a storage medium, and the method comprises the steps: a dynamic polarization modulation module generates phase retardation according to the intensity of a modulation electric field, and adjusts the polarization state of an incident light path to suppress the multi-path interference; the pulse synchronization controller generates a trigger period according to the time change and the space gradient of the scene light intensity, and controls exposure synchronization; and by combining the two parameters, the multi-lens optical array is accurately regulated and controlled to acquire images, and finally a hundred-million-pixel-level high-definition image is generated. According to the method, through real-time polarization adjustment and intelligent exposure control, low-distortion and real-time high-definition imaging in multiple scenes is realized, and quality and speed requirements in a dynamic environment are effectively balanced.

Description

Technical Field

[0001] This application belongs to the field of image acquisition, and in particular relates to a method, apparatus, device and storage medium for acquiring multi-scene dynamic images with a resolution of hundreds of millions of pixels. Background Technology

[0002] In the current field of image acquisition technology, especially for image acquisition in multi-scene dynamic environments at the megapixel level, traditional methods face numerous challenges and limitations. Traditional megapixel image acquisition systems mostly rely on fixed array sensors or mechanical scanning mechanisms. While these designs can maintain certain performance in static or low-dynamic-change scenes, their limitations become increasingly apparent in dynamic scenes, especially in complex environments requiring simultaneous high resolution, low distortion, and real-time adaptability. The demands of multiple scenes, such as rapid switching between near and far perspectives and sudden changes in lighting conditions, significantly increase the system's processing burden and complexity. Traditional solutions often compensate by increasing hardware redundancy or relying on software post-processing; however, these methods not only introduce response latency but also significantly increase resource consumption and exhibit insufficient scene generalization capabilities, making it difficult to meet the real-time and high-quality imaging requirements in dynamic environments. Summary of the Invention

[0003] The purpose of this application is to overcome the deficiencies in the prior art and provide a 100-megapixel drone-borne multi-scale camera system and method.

[0004] This application provides a method for acquiring multi-scene dynamic images with a resolution of 100 million pixels, including:

[0005] The dynamic polarization modulation module generates a phase delay based on the modulation electric field intensity. The phase delay is used to adjust the polarization state of the incident light path to suppress multipath interference.

[0006] The pulse synchronization controller generates a trigger cycle based on the temporal changes and spatial gradients of the scene light intensity distribution. The trigger cycle is used to control the synchronization of exposure events.

[0007] Based on the phase delay and the trigger period, the image acquisition of the multi-lens optical array is controlled, and a megapixel-level image is generated based on the acquired image data.

[0008] Optionally, the generation of the phase delay by the dynamic polarization modulation module based on the modulation electric field intensity includes:

[0009] The phase delay is calculated based on the modulation electric field strength using square root operations.

[0010] Optionally, the dynamic polarization modulation module includes a liquid crystal polarizer, and the phase delay of the liquid crystal polarizer is controlled by voltage to generate the phase delay amount.

[0011] Optionally, the step of generating a trigger cycle by the pulse synchronization controller based on the temporal changes and spatial gradients of the scene light intensity distribution includes:

[0012] The triggering period is calculated based on the temporal differential term and spatial Laplace term of the scene light intensity distribution.

[0013] Optionally, the triggering period is a non-uniform timestamp, generated by the pulse synchronization controller based on feedback from the light intensity distribution.

[0014] Optionally, generating a megapixel-level image based on the acquired image data includes:

[0015] The gradient product of the image data is processed using the hyperbolic tangent function to perform image fusion.

[0016] Optionally, the multi-lens optical array includes a main lens and an auxiliary lens, wherein the main lens and the auxiliary lens are a combination of heterogeneous curved surface lenses, the main lens is used for wide-area acquisition, and the auxiliary lens is used for dynamically adjusting the focal plane according to the scene.

[0017] This application provides a 100-megapixel multi-scene dynamic image acquisition device, including:

[0018] The phase module generates a phase delay based on the modulation electric field intensity by the dynamic polarization modulation module. The phase delay is used to adjust the polarization state of the incident light path to suppress multipath interference.

[0019] The periodic module, generated by the pulse synchronization controller based on the temporal changes and spatial gradients of the scene light intensity distribution, is used to control the synchronization of exposure events.

[0020] The adjustment module controls the image acquisition of the multi-lens optical array according to the phase delay and the trigger period, and generates a megapixel-level image based on the acquired image data.

[0021] This application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the system as described above.

[0022] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the system described above.

[0023] The beneficial effects of this application are:

[0024] Invention Point 1: Dynamic Polarization Modulation

[0025] Invention Point 2: Pulse Synchronization Trigger Function

[0026] Invention Point 3: Nonlinear Image Fusion

[0027] This application provides a method for acquiring 100-megapixel multi-scene dynamic images, comprising: a dynamic polarization modulation module generating a phase delay based on the modulation electric field intensity, wherein the phase delay is used to adjust the polarization state of the incident light path to suppress multipath interference; a pulse synchronization controller generating a trigger period based on the temporal variation and spatial gradient of the scene light intensity distribution, wherein the trigger period is used to control the synchronization of exposure events; controlling the image acquisition of a multi-lens optical array based on the phase delay and the trigger period; and generating a 100-megapixel-level image based on the acquired image data. This application achieves high-definition, low-distortion, and real-time output of 100-megapixel images in multiple scenes by using a dynamic polarization modulation module to adjust the polarization state of the light path in real time to suppress interference, and a pulse synchronization controller to intelligently control exposure synchronization based on changes in scene light intensity. Combined with precise acquisition by a multi-lens array, this method effectively resolves the contradiction between imaging quality and response speed in dynamic environments by enabling 100-megapixel images in multiple scenes with high definition, low distortion, and real-time output. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the 100-megapixel multi-scene dynamic image acquisition process in this application;

[0029] Figure 2 This is a schematic diagram of the 100-megapixel multi-scene dynamic image acquisition device in this application. Detailed Implementation

[0030] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is to be understood that various forms of implementation of the present disclosure are intended and should not be limited to the embodiments set forth herein. Rather, the embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0031] Please refer to Figure 1 As shown, this application provides a method for acquiring dynamic images in multiple scenes with a resolution of 100 million pixels. It is applied to the field of image acquisition technology for dynamic environments in multiple scenes and is suitable for vision systems that require high resolution and multi-scale adaptation. It is used to solve problems such as dynamic range limitation, scene distortion (low distortion), real-time contradiction, response delay, high resource consumption and insufficient scene generalization ability in 100 million pixel acquisition, especially when dealing with the needs of multiple scenes such as switching between near and far viewpoints and sudden changes in illumination.

[0032] The system involved in this application consists of a multi-lens optical array, a dynamic polarization modulation module, a pulse synchronization controller, and a nonlinear fusion unit. The core of this application lies in the deep coupling of the physical characteristics of optical devices with signal processing algorithms to form an inseparable technical whole, the synergistic effect of which far exceeds the sum of the individual workings.

[0033] The method includes:

[0034] S101. The dynamic polarization modulation module generates a phase delay based on the modulation electric field intensity. The phase delay is used to adjust the polarization state of the incident light path to suppress multipath interference.

[0035] The dynamic polarization modulation module is a device that dynamically changes optical properties through external signals, and in this application, it specifically refers to a unit that includes a liquid crystal polarizer.

[0036] The modulated electric field strength is a voltage signal that varies with time and is applied to the liquid crystal polarizer.

[0037] The polarization state is a property that describes the direction of vibration of the electric field of a light wave.

[0038] Multipath interference refers to the phenomenon where light travels through different paths before reaching the sensor, resulting in ghosting, blurring, or decreased contrast in the image.

[0039] The dynamic polarization modulation module includes a liquid crystal polarizer, and the phase delay of the liquid crystal polarizer is controlled by voltage to generate the phase delay amount.

[0040] When generating the phase delay, the phase delay is calculated using a square root operation based on the modulation electric field strength. Specifically, the dynamic polarization modulation module controls the phase delay of the liquid crystal polarizer via voltage control, with the voltage range between 0 and 10V, to adjust the polarization state of the incident light path in real time, thereby suppressing multipath interference. The phase delay is calculated using the dynamic polarization modulation model formula:

[0041]

[0042] Among them, the This represents the phase delay, where λ is the incident light wavelength. The time-dependent modulation electric field strength, the It is the field-induced variation term of the refractive index, where d is the thickness of the liquid crystal layer, and the... The dynamic incident angle is κ, which is the material nonlinear coefficient with dimensions m² / V².

[0043] The above formula describes the nonlinear relationship between polarization phase delay and electric field strength.

[0044] The formula introduces a square root term in the denominator, which physically represents the non-ohmic response caused by carrier migration under strong fields. This makes the phase delay sublinearly related to the electric field, rather than in the classical linear model. This design avoids saturation distortion under high electric fields and adapts to varying lighting conditions. For example, in strong light environments, the modulation electric field intensity increases, but the phase delay growth is slowed down through square root calculations, thus preventing optical saturation and ensuring that the dynamic range of image acquisition is improved to 120dB. Logically, this step deeply couples the physical characteristics of optical devices with electrical signal processing, achieving dynamic polarization modulation through a nonlinear control mechanism, unlike the traditional linear model, thereby effectively suppressing multipath interference and scene distortion.

[0045] S102. The pulse synchronization controller generates a trigger cycle based on the temporal changes and spatial gradient of the scene light intensity distribution. The trigger cycle is used to control the synchronization of exposure events.

[0046] The pulse synchronization controller is a signal generation unit that generates non-uniform electronic pulse signals based on real-time calculated scene characteristics.

[0047] The scene light intensity distribution refers to the light intensity information received from the scene, which is a three-dimensional function that varies with two-dimensional spatial coordinates (x, y) and time (t).

[0048] The time variation is the rate of change of scene light intensity distribution over time, which is obtained through the time derivative term.

[0049] The spatial gradient is the degree of change in scene light intensity in space, obtained through the Laplacian operator.

[0050] The exposure event refers to the complete process of an image sensor capturing an image once.

[0051] Synchronization refers to matching the timing of the exposure event with the dynamic changes in the scene in time.

[0052] When generating the trigger period, it is calculated based on the temporal derivative and spatial Laplacian term of the scene's light intensity distribution. The trigger period is a non-uniform timestamp, generated by the pulse synchronization controller based on feedback from the light intensity distribution. Specifically, the pulse synchronization controller generates a non-uniform timestamp signal, and its trigger period is adjusted by feedback from scene motion characteristics to ensure that the exposure event is synchronized with dynamic changes.

[0053] The trigger period is calculated using the pulse synchronization trigger function formula:

[0054]

[0055] in, The triggering period at time t is... The base period is set to 10ms. α is the intensity change sensitivity factor, with dimensions s² / cd. β is the spatial gradient weight, with dimensions m²·s. I(x,y,t) represents the scene light intensity distribution. For the Laplace operator.

[0056] This formula defines the non-uniform time interval for exposure triggering.

[0057] The formula combines an exponential term with a Laplacian operator: the exponential term captures instantaneous changes (such as sudden movement), while the Laplacian term senses spatial inhomogeneities (such as shadow boundaries). Traditional synchronizers rely solely on temporal differentiation, but this introduces spatial second-order differentiation, enabling triggering events to respond simultaneously to spatiotemporal abrupt changes, thus avoiding motion blur caused by global exposure. For example, when a fast-moving object appears in the scene, the temporal differentiation term increases, the triggering period shortens, and exposures become more frequent, thereby reducing motion blur; in areas of abrupt lighting changes, the spatial Laplacian term adjusts the triggering timing to ensure exposure synchronization. Logically, this step, through spatiotemporal joint control, suppresses motion blur to below 0.1 pixels. Traditional synchronizers rely solely on temporal differentiation, but this introduces spatial second-order differentiation, enhancing adaptability to dynamic scenes.

[0058] S103. Based on the phase delay and the trigger period, control the image acquisition of the multi-lens optical array, and generate a megapixel-level image based on the acquired image data.

[0059] The multi-lens optical array is a data acquisition system composed of multiple optical lenses arranged in a specific layout.

[0060] The image data refers to the set of raw images captured by each lens in a multi-lens optical array.

[0061] The generation refers to the process of synthesizing multi-source image data into a single, ultra-high-resolution image through a nonlinear fusion algorithm.

[0062] The term "megapixel-level image" refers to a composite image with a resolution of over 100 million pixels.

[0063] The multi-lens optical array includes a main lens and an auxiliary lens, which are a combination of heterogeneous curved surface lenses. The main lens is used for wide-area acquisition, and the auxiliary lens is used to dynamically adjust the focal plane according to the scene.

[0064] When generating megapixel-level images, the gradient product of the image data is processed using the hyperbolic tangent function for image fusion. Specifically, the multi-lens optical array uses a combination of heterogeneous curved lenses, where the ratio of the curvature radius of the main lens to the auxiliary lens is 1.5:1. The main lens is responsible for wide-area acquisition, while the auxiliary lens dynamically adjusts the focal plane according to the scene.

[0065] After image acquisition, the nonlinear fusion unit integrates multi-source image data through a coupling function, outputting a megapixel-level image. Image fusion employs a nonlinear image fusion equation:

[0066]

[0067] Wherein, Ifusion is the fused image, N is the number of lenses, and the... γ is the weighting factor for the k-th lens, adaptively allocated based on the lens signal-to-noise ratio. γ is the gradient coupling coefficient, with a value of 0.05 and dimensions of m² / cd. and These are the gradients of the image in the x and y directions, respectively.

[0068] The above formula achieves pixel-level fusion of multi-lens images.

[0069] The unconventional aspect of this formula lies in its use of a hyperbolic tangent transform of gradient products: traditional fusion employs linear weighting or maximum selection, while here a nonlinear function is used to emphasize edge intersections (such as texture boundaries) and suppress noise in flat areas.

[0070] gradient product term and Reflecting the anisotropic characteristics of pixels, the hyperbolic tangent function compresses them to the [-1,1] interval to avoid oversaturation of high gradient values.

[0071] For example, in image edge regions where gradient products are large, the hyperbolic tangent function compresses them to the [-1,1] interval, avoiding oversaturation and thus enhancing detail preservation. Logically, through deep coupling of optical modulation and signal processing, a gain exceeding the superposition effect is synergistically generated, applicable only to specific data features such as high gradient changes and non-uniform exposure acquired from multi-scene, high-resolution images. System calibration requires initial phase calibration under standard lighting conditions and does not require additional experimental data.

[0072] Through the above steps, this application achieves three major beneficial effects: First, dynamic polarization modulation improves the dynamic range; second, the pulse synchronization mechanism suppresses motion blur; and finally, the nonlinear fusion algorithm achieves a signal-to-noise ratio superior to linear methods. These effects are due to the deep synergy between optical nonlinearity, the triggering mechanism, and the fusion function, and are only applicable to specific data features acquired from multi-scene, high-resolution images (such as high gradient changes and non-uniform exposure). This application achieves high resolution, low distortion, and real-time adaptability through multimodal optical modulation and dynamic coupling, addressing multi-scene requirements such as switching between near and far viewing angles and sudden changes in illumination.

[0073] This application provides a 100-megapixel multi-scene dynamic image acquisition device, including:

[0074] The phase module 201 generates a phase delay amount based on the modulation electric field intensity by the dynamic polarization modulation module. The phase delay amount is used to adjust the polarization state of the incident light path to suppress multipath interference.

[0075] The period module 202 generates a trigger period based on the temporal changes and spatial gradients of the scene light intensity distribution by the pulse synchronization controller. The trigger period is used to control the synchronization of exposure events.

[0076] The adjustment module 203 controls the image acquisition of the multi-lens optical array according to the phase delay and the trigger period, and generates a megapixel-level image based on the acquired image data.

[0077] This application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the system as described above.

[0078] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the system described above.

[0079] The above description of the embodiments is provided to enable those skilled in the art to understand and apply this application. Those skilled in the art will readily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without inventive effort. Therefore, this application is not limited to the above embodiments, and any improvements and modifications made to this application based on the disclosure thereof should be within the scope of protection of this application.

Claims

1. A method for acquiring dynamic images of multiple scenes with hundreds of millions of pixels, characterized in that, include: The dynamic polarization modulation module generates a phase delay based on the modulation electric field intensity. The phase delay is used to adjust the polarization state of the incident light path to suppress multipath interference. The pulse synchronization controller generates a trigger cycle based on the temporal changes and spatial gradients of the scene light intensity distribution. The trigger cycle is used to control the synchronization of exposure events. Based on the phase delay and the trigger period, the image acquisition of the multi-lens optical array is controlled, and a megapixel-level image is generated based on the acquired image data.

2. The method according to claim 1, characterized in that, The generation of the phase delay by the dynamic polarization modulation module based on the modulation electric field intensity includes: The phase delay is calculated based on the modulation electric field strength using square root operations.

3. The method according to claim 1, characterized in that, The dynamic polarization modulation module includes a liquid crystal polarizer, and the phase delay of the liquid crystal polarizer is controlled by voltage to generate the phase delay amount.

4. The method according to claim 1, characterized in that, The triggering cycle generated by the pulse synchronization controller based on the temporal changes and spatial gradients of the scene light intensity distribution includes: The triggering period is calculated based on the temporal differential term and spatial Laplace term of the scene light intensity distribution.

5. The method according to claim 1, characterized in that, The triggering period is a non-uniform timestamp, generated by the pulse synchronization controller based on feedback from the light intensity distribution.

6. The method according to claim 1, characterized in that, The step of generating a megapixel-level image based on the collected image data includes: The gradient product of the image data is processed using the hyperbolic tangent function to perform image fusion.

7. The method according to claim 1, characterized in that, The multi-lens optical array includes a main lens and an auxiliary lens, which are a combination of heterogeneous curved surface lenses. The main lens is used for wide-area acquisition, and the auxiliary lens is used to dynamically adjust the focal plane according to the scene.

8. A multi-scene dynamic image acquisition device with a resolution of 100 million pixels, characterized in that, include: The phase module generates a phase delay based on the modulation electric field intensity by the dynamic polarization modulation module. The phase delay is used to adjust the polarization state of the incident light path to suppress multipath interference. The periodic module, generated by the pulse synchronization controller based on the temporal changes and spatial gradients of the scene light intensity distribution, is used to control the synchronization of exposure events. The adjustment module controls the image acquisition of the multi-lens optical array according to the phase delay and the trigger period, and generates a megapixel-level image based on the acquired image data.

9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method as described in claim 8.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed in a computer, causes the computer to perform the method of claim 8.

Images