An embedded image enhancement method and system for a large target area image sensor

Abstract

The application provides an embedded image enhancement method and system of a large target surface image sensor. In the application, the temperature gradient distribution and the environmental humidity data of the large target surface image sensor are collected, and an environmental parameter matrix and a temperature noise mapping model are constructed in association. The optical distortion of the incident light signal is corrected by using a liquid crystal modulator array, and an original image is output. Then, through a double-path module, the main path extracts noise features, and the auxiliary path separates color shift features caused by humidity. Based on the thermodynamic equation, the physical consistency of the features is verified, and the effective data is screened. Finally, the embedded processor dynamically optimizes the image contrast, dynamic range and target details, and outputs an enhanced image. Through environmental parameter modeling, double-path feature separation and physical verification, the application effectively suppresses the temperature and humidity interference, and improves the detail enhancement and color fidelity of the large target surface image sensor.

Description

Technical Field

[0001] This application relates to the field of embedded image processing technology, and in particular to an embedded image enhancement method and system for a large target surface image sensor. Background Technology

[0002] In temperature-sensitive scenarios such as industrial high-temperature detection and aerospace remote sensing, large-area image sensors are prone to pixel response non-uniformity, optical distortion, and humidity-related color drift due to thermal expansion and drastic fluctuations in ambient temperature and humidity. Such scenarios require embedded systems to achieve dynamic non-uniformity correction with limited hardware resources, while decoupling temperature gradient noise from humidity-induced color distortion and ensuring that the enhanced image details conform to thermodynamic physical constraints.

[0003] The current mainstream approach adopts a hybrid framework based on pre-calibrated lookup tables (LUTs) and data-driven enhancement: noise distribution of sensors under different temperature and humidity combinations is collected in advance to generate a LUT library, and pixel-level gain compensation coefficients are calculated through temperature interpolation to suppress fixed-pattern noise; then the compensated image is input into a convolutional neural network model for detail enhancement, and image quality is improved through end-to-end learning.

[0004] This scheme relies heavily on static calibration data and is difficult to adapt to complex transient temperature changes, which can lead to compensation failure. Its single temperature variable control mechanism ignores the nonlinear color shift caused by the synergistic effect of temperature and humidity, resulting in residual color distortion. In addition, the pure data-driven enhancement process of the convolutional neural network lacks thermodynamic constraints and is prone to over-sharpening details in high-temperature regions, thereby destroying the authenticity of the thermal radiation energy distribution. Summary of the Invention

[0005] This application provides an embedded image enhancement method and system for a large target area image sensor to solve the problems of image color distortion, noise residue, and detail enhancement distortion caused by ignoring temperature and humidity cross-interference and lack of physical constraints in the prior art.

[0006] In a first aspect, this application provides an embedded image enhancement method for a large target area image sensor, comprising:

[0007] The temperature gradient distribution of each pixel of a large target surface image sensor is collected, and environmental humidity data is acquired simultaneously. The temperature gradient distribution and environmental humidity data are correlated with the environmental humidity data according to spatial coordinates to form an environmental parameter matrix. A temperature noise mapping model is constructed based on the environmental parameter matrix.

[0008] The incident light signal received by the large target image sensor is processed by an integrated liquid crystal modulator array. The polarization direction of each modulation unit of the liquid crystal modulator array is adjusted according to the temperature gradient distribution to eliminate optical distortion and output the distortion-corrected original image signal.

[0009] The original image signal is input into the dual-path feature extraction module. In the dual-path feature extraction module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features.

[0010] The noise distribution characteristics and the color shift characteristics are physically consistent. An energy conservation constraint function is constructed through thermodynamic equations to verify the matching degree between the noise distribution characteristics and the color shift characteristics and the physical response characteristics of the sensor, and feature data that conforms to physical laws are selected.

[0011] Based on the feature data, an image enhancement operation is performed. The embedded processor dynamically adjusts the contrast and dynamic range of the image and enhances the detail features of the target area, outputting an optimized enhanced image.

[0012] Optionally, the step of constructing an energy conservation constraint function through thermodynamic equations, verifying the matching degree between the noise distribution characteristics, the color shift characteristics, and the sensor's physical response characteristics, and filtering out feature data that conforms to physical laws includes:

[0013] An energy conservation constraint function is constructed based on a preset thermodynamic equation, which includes noise distribution characteristics, color shift characteristics, and sensor physical response parameters.

[0014] The feature energy value of the interaction between the noise distribution feature and the color shift feature is calculated pixel by pixel based on the energy conservation constraint function.

[0015] The feature energy value of each pixel is matched with the pre-stored threshold range allowed by the sensor's physical response characteristics.

[0016] Pixel feature data whose feature energy values ​​are within the allowable threshold range are selected as feature data that conform to physical laws.

[0017] Optionally, the step of dynamically adjusting the contrast and dynamic range of the image and enhancing the detail features of the target region through an embedded processor to output an optimized enhanced image includes:

[0018] The contrast adjustment parameters are dynamically generated by the embedded processor based on the noise distribution characteristic values ​​in the feature data.

[0019] The embedded processor dynamically generates dynamic range extension parameters based on the color offset feature values ​​in the feature data.

[0020] The embedded processor identifies the target region and calculates detail enhancement parameters based on the pixel detail feature values ​​of the target region.

[0021] The original image signal is processed synchronously in the embedded processor according to the contrast adjustment parameters, dynamic range expansion parameters, and detail enhancement parameters, and an optimized enhanced image is output.

[0022] Optionally, the step of processing the incident light signal received by the large target image sensor through an integrated liquid crystal modulator array, adjusting the polarization direction of each modulation unit of the liquid crystal modulator array according to the temperature gradient distribution, eliminating optical distortion, and outputting the distortion-corrected original image signal includes:

[0023] A liquid crystal modulator array is integrated on the surface of the optical components of the large target image sensor, and the positional correspondence between each modulation unit in the liquid crystal modulator array and the pixel of the image sensor is established to generate a modulation unit coordinate mapping table.

[0024] Based on the temperature data of each pixel in the temperature gradient distribution, the polarization adjustment parameters of the corresponding modulation unit are determined through the modulation unit coordinate mapping table.

[0025] The polarization adjustment parameters drive each modulation unit in the liquid crystal modulator array to perform polarization direction adjustment, so that the incident light signal is projected onto the optical component through the adjusted liquid crystal modulator array, forming a compensation optical path to eliminate optical distortion.

[0026] The original image signal, after distortion correction, is output through the compensation optical path.

[0027] Optionally, the step of driving each modulation unit in the liquid crystal modulator array to perform polarization direction adjustment according to the polarization adjustment parameters, so that the incident light signal is projected onto the optical component through the adjusted liquid crystal modulator array to form a compensation optical path to eliminate optical distortion, includes:

[0028] The polarization adjustment parameters are input into the driving circuit of the liquid crystal modulator array to generate a voltage control signal corresponding to each modulation unit;

[0029] The polarization direction is adjusted by changing the molecular arrangement direction of each modulation unit in the liquid crystal modulator array according to the voltage control signal.

[0030] The incident light signal is made to generate a corresponding polarization deflection by passing through the liquid crystal modulator array after the polarization direction is adjusted, thereby generating a polarization-corrected light signal.

[0031] The polarization-corrected light signal is projected onto the surface of the optical component to form a compensation optical path that eliminates optical distortion.

[0032] Optionally, the step of inputting the original image signal into a dual-path feature extraction module, wherein the main path in the dual-path feature extraction module extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features, includes:

[0033] The original image signal is input into the dual-path feature extraction module;

[0034] In the main path of the dual-path feature extraction module, the temperature noise mapping model is called to process the original image signal and extract the temperature-related noise distribution features of each pixel.

[0035] In the auxiliary path of the dual-path feature extraction module, the original image signal is received synchronously and combined with the environmental humidity data to separate the humidity-induced color shift features of each pixel.

[0036] The noise distribution features and color offset features maintain the same pixel coordinate mapping relationship.

[0037] Optionally, the step of acquiring the temperature gradient distribution of each pixel of the large target surface image sensor, simultaneously acquiring ambient humidity data, associating the temperature gradient distribution and ambient humidity data according to spatial coordinates to form an environmental parameter matrix, and constructing a temperature noise mapping model based on the environmental parameter matrix, includes:

[0038] Temperature gradient distribution of each pixel of a large target image sensor is collected by a temperature sensor array, and ambient humidity data is acquired synchronously by a humidity sensor.

[0039] The temperature gradient distribution data corresponding to each pixel is associated and bound with the ambient humidity data according to the two-dimensional spatial coordinates of the pixel. An environmental parameter matrix containing coordinate positions and their corresponding data is generated based on the associated data of all pixels.

[0040] A temperature noise mapping model is directly constructed based on the mapping relationship between the temperature gradient values ​​in the environmental parameter matrix and the image noise.

[0041] Secondly, this application provides an embedded image enhancement system for a large target area image sensor, comprising:

[0042] The acquisition module is used to acquire the temperature gradient distribution of each pixel of the large target surface image sensor, simultaneously acquire environmental humidity data, associate the temperature gradient distribution and environmental humidity data according to spatial coordinates to form an environmental parameter matrix, and construct a temperature noise mapping model based on the environmental parameter matrix.

[0043] The processing module is used to process the incident light signal received by the large target image sensor through an integrated liquid crystal modulator array, adjust the polarization direction of each modulation unit of the liquid crystal modulator array according to the temperature gradient distribution, eliminate optical distortion, and output the distortion-corrected original image signal.

[0044] The extraction module is used to input the original image signal into the dual-path feature extraction module. In the dual-path feature extraction module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features.

[0045] The verification module is used to verify the physical consistency between the noise distribution characteristics and the color shift characteristics. It constructs an energy conservation constraint function through thermodynamic equations, verifies the matching degree between the noise distribution characteristics and the color shift characteristics and the physical response characteristics of the sensor, and filters out feature data that conforms to physical laws.

[0046] The output module is used to perform image enhancement operations based on the feature data, dynamically adjust the contrast and dynamic range of the image through the embedded processor and enhance the detailed features of the target area, and output the optimized enhanced image.

[0047] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to implement an embedded image enhancement method for a large target area image sensor as described in the first aspect above.

[0048] Fourthly, this application provides a computer storage medium storing a computer program, which, when executed by a computer, implements an embedded image enhancement method for a large target surface image sensor as described in the first aspect.

[0049] In this application example, the temperature gradient distribution of each pixel of a large-area image sensor is acquired, and environmental humidity data is simultaneously obtained. The temperature gradient distribution and environmental humidity data are correlated by spatial coordinates to form an environmental parameter matrix, and a temperature noise mapping model is constructed based on the environmental parameter matrix. The incident light signal received by the large-area image sensor is processed by an integrated liquid crystal modulator array. The polarization direction of each modulation unit of the liquid crystal modulator array is adjusted according to the temperature gradient distribution to eliminate optical distortion, and the distortion-corrected original image signal is output. The original image signal is input into a dual-path feature extraction module. In the dual-path feature extraction module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path separates humidity-induced color shift features by combining the environmental humidity data. The physical consistency of the noise distribution features and the color shift features is checked. An energy conservation constraint function is constructed through thermodynamic equations to check the matching degree between the noise distribution features, the color shift features and the sensor's physical response characteristics, and feature data that conforms to physical laws is selected. Based on the feature data, an image enhancement operation is performed. The contrast and dynamic range of the image are dynamically adjusted by an embedded processor, and the detail features of the target area are enhanced, and the optimized enhanced image is output.

[0050] The technical solution of this application has the following beneficial effects:

[0051] This application constructs an environmental parameter matrix and a temperature noise mapping model related to temperature and humidity, and combines it with a liquid crystal modulator array to dynamically correct optical distortion; it uses a dual-path feature extraction module to simultaneously separate temperature noise features and humidity-induced color shift features, and performs physical consistency verification based on thermodynamic equations to screen valid data; finally, it achieves adaptive adjustment of contrast and dynamic range and enhancement of target details in an embedded processor, achieving a synergistic optimization effect of simultaneous suppression of temperature and humidity interference, preservation of image physical authenticity, and high-fidelity enhancement of key areas.

[0052] The physical consistency verification steps are further defined as follows: an energy conservation constraint function is constructed based on thermodynamic equations, including noise distribution characteristics, color shift characteristics, and sensor physical response parameters; the feature energy value of the interaction between the two types of features is calculated pixel by pixel; the energy value is matched and verified against the allowable threshold range of the sensor's physical response; and feature data within the threshold range is selected as valid output. Through pixel-by-pixel quantization verification of the energy conservation constraint function, distorting features that violate thermodynamic laws are accurately eliminated, ensuring that the enhancement process strictly follows the sensor's physical response characteristics. This fundamentally avoids image thermal radiation distortion and abnormal energy distribution caused by environmental interference, improving the physical credibility and scene adaptability of the enhancement results.

[0053] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A flowchart of an embedded image enhancement method for a large target surface image sensor provided in this application is shown;

[0056] Figure 2 The illustration shows a scene diagram of an embedded image enhancement method for a large target area image sensor provided in this application;

[0057] Figure 3 This invention provides a schematic diagram of the structure of an embedded image enhancement system for a large target surface image sensor.

[0058] Figure 4 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0059] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0060] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0061] Research indicates that current enhancement schemes for large-area image sensors in temperature-sensitive environments suffer from three major limitations due to their reliance on static temperature and humidity calibration lookup tables (LUTs) and purely data-driven convolutional neural network enhancement frameworks: First, the pre-stored calibration lookup tables struggle to cover dynamic temperature changes such as transient thermal shocks, leading to the failure of non-uniformity correction. Second, the single temperature variable control mechanism ignores the nonlinear color shift caused by the synergistic effect of temperature and humidity, resulting in residual regional color shifts in the corrected image. Third, the convolutional neural network enhancement process lacks thermodynamic constraints, and excessive sharpening of details in high-temperature regions can destroy the authenticity of thermal radiation energy distribution, causing distortion of physical information in industrial inspection and remote sensing scenarios.

[0062] To address the aforementioned issues, this application proposes an embedded image enhancement method for large-area image sensors. The core of this method lies in constructing a temperature and humidity-related environmental parameter matrix to drive a liquid crystal modulator to eliminate optical distortion, and utilizing dual-path feature extraction to decouple temperature noise and humidity color shift features. Furthermore, it combines thermodynamic energy conservation constraint functions to filter physically compliant data pixel-by-pixel, and finally, the embedded processor performs dynamic enhancement. This method replaces static LUTs with dynamic modeling to solve the problem of transient temperature change failure, eradicates residual color shift through dual-path separation, and maintains the authenticity of thermal radiation through physical verification. It achieves high-fidelity enhancement of key details while ensuring compliance with the physical laws of the image.

[0063] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0064] Figure 1 This application provides a flowchart of an embedded image enhancement method for a large target area image sensor, as shown in the following embodiment. Figure 1 As shown, the method includes:

[0065] 101. Collect the temperature gradient distribution of each pixel of the large target image sensor, and simultaneously acquire the ambient humidity data. Associate the temperature gradient distribution and the ambient humidity data according to spatial coordinates to form an environmental parameter matrix. Construct a temperature noise mapping model based on the environmental parameter matrix.

[0066] Optionally, step 101 may specifically include the following steps:

[0067] 1011. The temperature gradient distribution of each pixel of the large target surface image sensor is collected through a temperature sensor array, and the ambient humidity data is acquired synchronously through a humidity sensor.

[0068] 1012. Associate and bind the temperature gradient distribution data corresponding to each pixel with the ambient humidity data according to the two-dimensional spatial coordinates of the pixel, and generate an environmental parameter matrix containing the coordinate position and its corresponding data based on the association data of all pixels;

[0069] 1013. A temperature noise mapping model is directly constructed based on the mapping relationship between the temperature gradient value and image noise in the environmental parameter matrix.

[0070] In the above scheme, the temperature gradient distribution refers to the temperature change rate data of each pixel on the surface of the large target image sensor, including the temperature difference between adjacent pixels and the spatial change trend, which is used to quantify the pixel response differences caused by thermal expansion.

[0071] Ambient humidity data refers to the absolute humidity value of the sensor's environment, reflecting the intensity of interference from the concentration of water molecules in the air on the optical path. The environmental parameter matrix is ​​a structured dataset that binds the two-dimensional coordinates, temperature gradient value, and ambient humidity value H of each pixel, establishing a mapping relationship between spatial location and environmental parameters. The temperature noise mapping model is an expression describing the mathematical relationship between the temperature gradient and image noise intensity, used to predict pixel-level noise distribution caused by temperature changes.

[0072] In this embodiment, firstly, step 1011 uses a high-precision temperature sensor array, such as a micrometer-level thermocouple grid, to collect temperature data for each pixel on the surface of a large target image sensor to obtain the temperature gradient distribution. Simultaneously, a humidity sensor integrated within the sensor module acquires ambient humidity data. The temperature sensor array covers the sensor target surface in a grid pattern, with each sensing unit corresponding to a pixel position, directly measuring the temperature at that point. The humidity sensor continuously outputs ambient absolute humidity data. For example, pixel (10,20) measures 52.1℃, the adjacent pixel (10,21) measures 53.0℃, and the ambient absolute humidity is 65%RH.

[0073] Next, based on the temperature gradient distribution and ambient humidity data output in step 1011, step 1012 calculates the temperature change rate between each pixel and its neighboring pixels to generate temperature gradient values. The calculation formula is as follows: ,in For the current pixel Temperature value, Adjacent pixels The temperature value, and d, the physical distance between two pixels, are determined by the sensor design specifications. Then, the two-dimensional coordinates of each pixel are... , corresponding temperature gradient value The ambient humidity value H is bound to a structured data unit, and finally integrated to generate an environmental parameter matrix. Each row of this matrix stores the environmental parameters of a single pixel, establishing a precise mapping relationship between spatial location and environmental parameters. For example, pixel (10,20) corresponds to a temperature of 52.1℃, and pixel (10,21) corresponds to a temperature of 53.0℃; the temperature gradient value between these two pixels is... Combined with the corresponding ambient humidity value H of 65, the generated environmental parameter matrix is ​​[x=10,y=20,ΔT=0.9,H=65].

[0074] Finally, in step 1013, a correlation analysis is performed between the temperature gradient data in the environmental parameter matrix and the pre-stored historical noise database. The temperature gradient values ​​are then fitted using a linear regression algorithm. The mathematical relationship between noise intensity N and temperature is used to construct a temperature noise mapping model. Specifically, this involves extracting all temperature gradient values ​​from the matrix. And the historical noise measured values ​​at the corresponding locations, the coefficients are solved using the least squares method to obtain the form as follows: The equation is used to predict the temperature-induced noise intensity of any pixel. For example, the temperature gradient value. The historical noise intensity measured at that time was N=15dB. A noise mapping model was established by fitting the data using linear regression, yielding k=0.03 and b=0.5. .

[0075] In practical applications, during surface temperature monitoring of industrial equipment A, when step 1011 is executed, the temperature sensor measures a temperature of 51.8℃ for pixel (100, 200), a temperature of 54.3℃ for the adjacent pixel (100, 201), and the humidity sensor outputs 70%RH; step 1012 calculates the gradient at this location. Bind coordinates (100, 200) =2.5, H=70 to generate matrix entries; Step 1013 calls the historical database to record N=0.56dB when ΔT=2.0℃ / mm, and establishes a model through linear regression. Substitute =2.5, so the predicted noise intensity N = 0.03 × 2.5 + 0.5 = 0.575 dB.

[0076] The above-mentioned overall solution 101 achieves precise spatial correlation of temperature and humidity data through synchronous data acquisition of temperature sensor array and humidity sensor, pixel-level temperature gradient calculation and construction of environmental parameter matrix; further, based on regression modeling of historical noise data, a quantitative mapping model that can dynamically predict temperature noise is generated, providing an adaptive calculation basis for noise suppression in complex temperature change scenarios, and significantly improving the accuracy and robustness of non-uniformity correction.

[0077] 102. The incident light signal received by the large target image sensor is processed by an integrated liquid crystal modulator array, the polarization direction of each modulation unit of the liquid crystal modulator array is adjusted according to the temperature gradient distribution, optical distortion is eliminated, and the distortion-corrected original image signal is output.

[0078] Optionally, step 102 may specifically include the following steps:

[0079] 1021. Integrate a liquid crystal modulator array on the surface of the optical components of the large target image sensor, establish the positional correspondence between each modulation unit in the liquid crystal modulator array and the pixel of the image sensor, and generate a modulation unit coordinate mapping table.

[0080] 1022. Based on the temperature data of each pixel in the temperature gradient distribution, determine the polarization adjustment parameters of the corresponding modulation unit through the modulation unit coordinate mapping table;

[0081] 1023. Drive each modulation unit in the liquid crystal modulator array to perform polarization direction adjustment according to the polarization adjustment parameters, so that the incident light signal is projected onto the optical component through the adjusted liquid crystal modulator array to form a compensation optical path to eliminate optical distortion.

[0082] Step 1023 may specifically include the following processes: inputting the polarization adjustment parameters into the driving circuit of the liquid crystal modulator array to generate a voltage control signal corresponding to each modulation unit; changing the molecular arrangement direction of each modulation unit in the liquid crystal modulator array according to the voltage control signal to achieve polarization direction adjustment; causing the incident light signal to generate a corresponding polarization deflection amount through the polarization-adjusted liquid crystal modulator array to generate a polarization-corrected light signal; and projecting the polarization-corrected light signal onto the surface of the optical component to form a compensation optical path to eliminate optical distortion.

[0083] 1024. Output the original image signal after distortion correction through the compensation optical path.

[0084] In the above scheme, the liquid crystal modulator array refers to a tunable optical device integrated on the optical surface of the image sensor. It is composed of a micron-scale liquid crystal unit matrix. Each unit can change the internal molecular alignment direction through a voltage signal, thereby dynamically adjusting the polarization angle of the transmitted light to correct the optical path offset caused by thermal distortion. The modulation unit coordinate mapping table is an index dataset that records the one-to-one correspondence between each modulation unit in the liquid crystal array and the spatial position of the image sensor pixel. It is generated by a coordinate calibration algorithm to ensure the accurate association between temperature data and optical modulation units. The polarization adjustment parameter refers to the amount of liquid crystal unit polarization angle adjustment calculated by a physical model based on the pixel temperature gradient value. This parameter is used to drive the liquid crystal molecules to deflect in an oriented manner to counteract thermal expansion distortion. The compensation optical path refers to the corrective optical path formed after the incident light signal has been polarized, eliminating the geometric deformation caused by uneven temperature distribution on the sensor surface, and ultimately outputting the original image signal with a true geometric structure.

[0085] In this embodiment, a liquid crystal modulator array, consisting of tens of thousands of micron-sized liquid crystal cells, is first tightly bonded to the surface of the optical lens of the image sensor in step 1021. A laser positioning system scans the physical coordinates of each liquid crystal cell and sensor pixel, and a coordinate matching algorithm generates a modulation cell coordinate mapping table. For example, the physical location of liquid crystal cell M20 is precisely bound to the sensor pixel region (100, 200), and M21 is bound to (100, 201), forming a one-to-one location database of "liquid crystal cell - pixel".

[0086] Next, in step 1022, the liquid crystal cell number corresponding to each pixel is looked up according to the modulation unit coordinate mapping table, based on the temperature gradient value of each pixel in the temperature gradient distribution. The polarization adjustment parameters are obtained by adjusting the liquid crystal unit corresponding to each pixel using the formula. Calculation, where The calibration experiment yielded a value of 0.3° / ℃ / mm. This represents the temperature gradient value for each pixel. For example, the temperature gradient at pixel (100, 200). =2.5℃ / mm, corresponding to liquid crystal cell M20, calculate its polarization adjustment parameters as follows: .

[0087] Then, in step 1023, the calculated polarization adjustment parameters are input into the driving circuit of the liquid crystal modulator array, and the voltage control signal of the modulation unit is generated through a preset voltage conversion model. The conversion formula is as follows: ,in =5 represents the voltage conversion coefficient; the voltage applied to the liquid crystal cell electrodes causes the liquid crystal molecules inside the modulation cell to rotate oriented under the influence of the electric field, synchronously changing the polarization direction of the transmitted light; at this time, the incident light undergoes polarization deflection after passing through the rotated liquid crystal layer, precisely offsetting the optical path shift caused by thermal expansion due to the temperature gradient in this region. For example, the polarization adjustment parameter corresponding to pixel (100, 200) Voltage control signals are generated using formulas. =3.75, the modulation unit is rotated 0.75° to compensate for the temperature gradient in this region. The thermal expansion caused the optical path to shift. The corrected optical signal is then projected perpendicularly back onto the imaging surface of the optical component, forming a compensating optical path that eliminates geometric distortion, so that the metal edge that was originally bent due to thermal deformation appears as a straight, true shape on the sensor target surface.

[0088] Finally, the compensated light path generated by the liquid crystal modulator array is projected onto the target surface of the image sensor to eliminate optical distortion caused by temperature gradient. The sensor converts the corrected light signal into the original image electrical signal output without any enhancement processing. This signal completely preserves the geometric structure and radiation information of the real scene, providing a high-fidelity input basis for the subsequent dual-path feature extraction module.

[0089] In practical applications, during the detection of overheated components in industrial equipment B, the image sensor detected a temperature gradient ΔT = 3.2℃ / mm at pixel (150, 250) in the target area; the pre-generated modulation unit coordinate mapping table confirmed that this area was handled by the liquid crystal unit M75; and the polarization adjustment parameters were calculated based on the physical model. Input the θ value into the drive circuit, and calculate the required voltage using the voltage conversion formula: The light is applied to the M75 unit to rotate its liquid crystal molecules by 0.96°, and the polarization direction of the incident light is corrected simultaneously to counteract the equivalent optical path deflection caused by thermal expansion. Finally, the compensation optical path is projected onto the sensor and the corrected image is output. The edges of the heat sink fins, which were originally distorted due to high temperature, are restored to a straight shape, forming the original image signal with a true geometric structure.

[0090] The overall scheme of 102 described above uses temperature data-driven dynamic adjustment of liquid crystal polarization to counteract optical distortion caused by thermal expansion, ensuring that the large target sensor captures the original image of the true geometric structure in a high-temperature environment, providing high-fidelity input for subsequent feature extraction.

[0091] 103. Input the original image signal into the dual-path feature extraction module. In the dual-path feature extraction module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features.

[0092] Optionally, step 103 may specifically include the following steps:

[0093] 1031. Input the original image signal into the dual-path feature extraction module;

[0094] 1032. In the main path of the dual-path feature extraction module, the temperature noise mapping model is called to process the original image signal and extract the temperature-related noise distribution features of each pixel.

[0095] 1033. In the auxiliary path of the dual-path feature extraction module, the original image signal is received synchronously and combined with the environmental humidity data to separate the humidity-induced color shift features of each pixel.

[0096] 1034. The noise distribution features and color offset features maintain the same pixel coordinate mapping relationship.

[0097] In the above scheme, the dual-path feature extraction module refers to an embedded architecture that processes image signals in parallel, containing independent main and auxiliary paths to achieve physical decoupling of temperature and humidity interference. Noise distribution features refer to pixel-level signal distortion data directly caused by temperature gradients, including the intensity and spatial distribution pattern of charge migration due to thermal expansion, used to characterize sensor response non-uniformity. Color shift features refer to the color channel deviation caused by optical path differences due to humidity changes, including intensity shift values ​​and spatial distribution of specific bands, used to identify humidity-induced color distortion. Pixel coordinate mapping relationship refers to the spatial correspondence rule that strictly binds the feature data output by the main / auxiliary paths to the pixel positions of the original image, ensuring the consistency of positioning in subsequent physical verification.

[0098] In this embodiment, the corrected original image signal is first synchronously transmitted to the main path and auxiliary path of the dual-path feature extraction module via step 1031. This module adopts a hardware-level parallel processing architecture to ensure that the main and auxiliary paths simultaneously receive completely identical image data streams and retain the spatial coordinate information of the original pixels. For example, the RGB value (120, 85, 110) of pixel (150, 250) is synchronously acquired by both paths, providing a unified input basis for subsequent feature separation.

[0099] Next, step 1032 calls the temperature noise mapping model generated in step 101 through the main path of the dual-path feature extraction module, processing the original signal pixel by pixel. First, it locates the pixel and reads the temperature gradient value from the environmental parameter matrix; then, it substitutes the value into the formula of the temperature noise mapping model to calculate the noise intensity, which is used to characterize the intensity of thermally induced charge migration; finally, it outputs the noise distribution characteristic value of the pixel, 0.596dB. For example, the temperature gradient value corresponding to pixel (150, 250) is ΔT = 3.2℃ / mm, which is substituted into the formula to calculate the noise intensity of the pixel. The above operation is performed on all pixels in the original image signal to generate noise distribution features.

[0100] Then, in step 1033, the original image signal is received synchronously through the auxiliary path of the dual-path feature extraction module, and combined with global environmental humidity data, a preset humidity color shift model is invoked. ,in The humidity sensitivity coefficient is determined by calibrating the sensor individually. These are the original channel values. Pixel-by-pixel color shift analysis: First, the original value of the specified color channel for the current pixel is read. Then, the channel shift is calculated based on humidity data. The humidity-induced color shift feature value of this pixel is output, generating a global pixel-level color shift feature map for subsequent physical verification. For example, at pixel (300, 400), the original green channel value G=110, the ambient humidity value H=75%RH, and the humidity sensitivity coefficient is 0.01. Substituting these values ​​into the model, the color shift feature is calculated as follows: This indicates that the green channel experienced an 82.5-unit increase in humidity.

[0101] Finally, in step 1034, the noise distribution features output by the main path and the color shift features output by the auxiliary path are strictly bound to the same pixel coordinates, generating a feature pair with perfectly matched spatial location. For example, if the noise feature value obtained by the main path is 0.596dB and the color shift feature value obtained by the auxiliary path is ΔB=77, the two features are bound to pixel (150,250) to generate a matching feature pair {coordinates (150,250): noise=0.596dB, color shift=ΔB=77}. This process achieves full-domain pixel-level matching through a coordinate index table, ensuring that subsequent physical verification can accurately locate the temperature and humidity interference components at each location.

[0102] In practical applications, when extracting features from the original image during welding point monitoring of industrial equipment D, the RGB value (100, 120, 90) at pixel (200, 300) in the corrected original image is simultaneously input into the dual-path module; the main path queries the temperature gradient at that location. =3.5℃ / mm, call the noise model Calculate noise characteristics: Add 0.5 to get The auxiliary path, combined with an ambient humidity of H=80%RH, and based on the original value G=120 for the green channel, is determined according to the color shift model. Calculate: 0.01 × 80 = 0.8, then multiply by 120 to get... ; to use noise characteristic values With color deviation characteristic value Bind coordinates (200, 300), output feature pairs {coordinates (200, 300), noise:} Color deviation: This is for subsequent verification.

[0103] The overall solution described above (103) simultaneously extracts temperature noise distribution and humidity color shift features through a dual-path parallel architecture: the main path quantifies thermally induced signal distortion based on a temperature noise mapping model, while the auxiliary path analyzes humidity-induced channel imbalance using environmental humidity data, achieving physical separation of temperature and humidity interference; at the same time, it strictly maintains pixel-level spatial alignment of feature data, providing independent and accurate input data for subsequent thermodynamic constraint verification, fundamentally solving the feature confusion problem caused by temperature and humidity coupling in traditional solutions, and significantly improving the interference suppression accuracy in complex environments.

[0104] 104. Perform physical consistency verification on the noise distribution characteristics and the color shift characteristics, construct an energy conservation constraint function through thermodynamic equations, verify the matching degree between the noise distribution characteristics and the color shift characteristics and the physical response characteristics of the sensor, and select feature data that conforms to physical laws.

[0105] Optionally, step 104 may specifically include the following steps:

[0106] 1041. Construct an energy conservation constraint function based on a preset thermodynamic equation, which includes noise distribution characteristics, color shift characteristics, and sensor physical response parameters;

[0107] 1042. Calculate the feature energy value of the interaction between the noise distribution feature and the color shift feature pixel by pixel according to the energy conservation constraint function;

[0108] 1043. Verify the matching degree between the feature energy value of each pixel and the pre-stored allowable threshold range of sensor physical response characteristics;

[0109] 1044. Select the pixel feature data whose feature energy value is within the allowable threshold range as feature data that conforms to physical laws.

[0110] In the above scheme, the energy conservation constraint function is a mathematical expression based on the first law of thermodynamics, which includes noise distribution characteristics, color shift characteristics, and sensor physical response parameters. It quantifies the compliance of signal energy under temperature and humidity interference to assess whether the feature data conforms to physical laws. The feature energy value is the comprehensive energy evaluation value generated by the interaction of noise and color shift features within a thermodynamic framework. It is calculated pixel-by-pixel through the constraint function and reflects the physical rationality of temperature and humidity interference at that location. The allowable threshold range refers to the boundary interval of acceptable feature energy values ​​within the sensor's physical limits. It is determined by the sensor material's thermal capacity characteristics and photoelectric response laws and is used to filter out distorted data that violates energy conservation. Physically compliant feature data refers to the feature set that passes the threshold verification. Its noise distribution and color shift strictly follow the sensor's thermodynamic response mechanism, ensuring that subsequent image enhancement does not destroy physical authenticity.

[0111] In this embodiment of the application, firstly, an energy conservation constraint function is constructed based on the first law of thermodynamics and the physical characteristic parameters of the sensor in step 1041: ,in The noise energy conversion factor has a calibration value of , The color shift energy conversion coefficient has a calibration value of , The temperature and humidity weighting ratio is calibrated to be 0.5. These are noise characteristic values. This function quantifies the energy compliance of the interaction between the noise feature value and the color shift feature, transforming abstract physical laws into a calculable feature energy value E. For example, the interaction energy between the noise feature value 0.6dB and the color shift feature ΔB=75 will be quantified according to this formula.

[0112] Next, based on the energy conservation constraint function constructed in step 1041, the feature energy values ​​are processed pixel by pixel through step 1042: the noise distribution feature value and color shift feature value of the current pixel are read and substituted into the energy conservation constraint function. Numerical calculations were performed, and the calculation process strictly followed arithmetic precedence: first, the noise components were solved separately. With color cast Then, sum them to obtain the dimensionless feature energy value E, forming the overall image energy distribution map. For example, read the dual-path feature data of pixel (100, 200): noise feature value N = 0.6dB and color shift feature ΔB = 75. Substitute them into the function to calculate: .

[0113] Then, in step 1043, the characteristic energy value E calculated in step 1042 is compared with the pre-stored sensor physical response allowable threshold range. The comparison and verification process is as follows: If the feature energy value E is within the specified range, the pixel feature data is determined to conform to thermodynamic laws; if the feature energy value E exceeds the threshold range, it is determined to violate physical laws. The verification result generates a binary marker map, where 0 indicates violation and 1 indicates compliance, used to guide feature data selection. For example, if a pixel feature energy value E=6.75 is compared with the sensor's physical response allowable threshold range [5.0, 7.0], the result is compliant and marked as 1 in the marker map; if a pixel feature energy value E=8.6, it is marked as non-compliant because it exceeds the threshold and is marked as 0 in the marker map.

[0114] Finally, based on the marker map from step 1043, step 1044 filters the verified pixel feature data. For pixels marked as compliant, their noise and color shift features are retained and packaged into... Data pairs are processed; for pixels marked as violations, their feature data is directly removed. The final output is a full-image physically compliant feature dataset, which is then transferred to the image enhancement module for further processing.

[0115] In practical applications, for heat sink temperature monitoring of industrial equipment D, dual-path feature data is input for pixel (500, 600): noise distribution characteristics. Color shift characteristics Based on sensor physical parameters Constructing the energy function Substitute numerical calculations: First, calculate the noise components. Then calculate the color cast component. ,final ;Will The data is compared with the pre-stored thresholds [5.0, 7.0], and if it is deemed compliant, it is marked with 1 in the generated binary marker image; this feature data is then retained. For the abnormal pixel (600, 700) Calculated If the value exceeds the threshold [5.0, 7.0], feature removal is performed, and the value is marked as 0 in the binary marker map. The pixel feature data that passes the verification in the marker map is filtered and packaged into data pairs, outputting a full-image physical compliance feature dataset.

[0116] The above-mentioned 104 overall scheme verifies the physical compliance of noise and color shift features pixel by pixel through the thermodynamic energy conservation constraint function, accurately filters out data that violates the physical response law of the sensor, and ensures that subsequent image enhancement is performed only based on thermodynamically compatible feature data. This fundamentally eliminates image radiation distortion and geometric distortion caused by environmental interference, and significantly improves the physical authenticity and reliability of images in industrial inspection and remote sensing scenarios.

[0117] 105. Based on the feature data, perform image enhancement operations, dynamically adjust the contrast and dynamic range of the image through the embedded processor, enhance the detailed features of the target area, and output the optimized enhanced image.

[0118] Optionally, step 105 may specifically include the following steps:

[0119] 1051. The embedded processor dynamically generates contrast adjustment parameters based on the noise distribution characteristic values ​​in the feature data;

[0120] 1052. The embedded processor dynamically generates dynamic range extension parameters based on the color offset feature values ​​in the feature data;

[0121] 1053. The embedded processor identifies the target region and calculates detail enhancement parameters based on the pixel detail feature values ​​of the target region;

[0122] 1054. The original image signal is processed synchronously in the embedded processor according to the contrast adjustment parameters, dynamic range expansion parameters and detail enhancement parameters, and the optimized enhanced image is output.

[0123] In the above scheme, the contrast adjustment parameter refers to the image brightness difference adjustment coefficient dynamically generated based on noise distribution feature values. It includes the brightness compensation intensity and spatial distribution pattern of signal distortion areas, used to optimize image layer performance. The dynamic range expansion parameter refers to the brightness level expansion coefficient calculated based on color shift feature values. It includes the exposure compression rate and channel balance factor for high dynamic scenes, used to restore shadow / highlight details. The detail enhancement parameter refers to the texture enhancement coefficient for the target area, including the local gradient response intensity and sharpening convolution kernel weights, used to improve the recognition of key structures. The synchronous processing mechanism refers to the operational architecture of the embedded processor executing contrast adjustment, dynamic range expansion, and detail enhancement in parallel. Through hardware acceleration, the three types of parameters work synergistically to ensure that the output image enhances target information while suppressing environmental interference.

[0124] In this embodiment of the application, firstly, step 1051 reads the noise distribution feature value from the physical compliance feature data filtered in step 104 using an embedded processor, calls a preset noise contrast mapping model, and calculates and generates the contrast adjustment parameters for the region. The calculation formula is as follows: ,in Adjust the parameters for contrast. For adjustment coefficients, The upper limit of noise, This represents the characteristic value of the noise distribution. Contrast adjustment parameters will be used to improve the brightness difference in low-noise areas. For example, the noise upper limit. The noise distribution characteristic value at pixel (300, 400) is 1.0. Then the contrast adjustment parameter of the pixel is calculated. .

[0125] Next, based on the color shift feature value in the compliant feature data of the same pixel in step 1052, the embedded processor applies the color shift dynamic range model to calculate and generate the dynamic range extension parameter. The calculation formula is as follows: ,in Reference gain, This is the color offset threshold. For color shift feature values, This refers to the dynamic range expansion parameter. The dynamic range expansion parameter is used to recover tonal details in high-humidity areas. For example, the color shift characteristic value of the same pixel is... First, calculate the color bias percentage. Then calculate 1-0.7=0.3 to finally obtain the dynamic range extension parameter of the pixel. .

[0126] Then, in step 1053, the embedded processor scans the original image using the Sobel edge detection algorithm, locates the target region, and outputs pixel detail feature values, such as edge point pixel detail feature values. Subsequently, the detail enhancement formula is called. Detail enhancement parameters are generated based on pixel detail feature values, where k is the sharpening coefficient set to 0.02. These are the pixel detail feature values ​​for edge points. Detail enhancement parameters will specifically enhance the texture of key structures. For example, the pixel detail feature value for the edge gradient of a bolt is... Substituting into the formula, the detail enhancement parameters are calculated as follows: .

[0127] Finally, step 1054 initiates the parallel computing unit via the embedded processor, and the first unit adjusts the parameters according to the ratio. Perform histogram stretching to adjust contrast; the second unit expands parameters according to dynamic range. Layered compression of brightness extends dynamic range; the third unit applies detail enhancement parameters to the target area. The sharpening convolution kernel enhances details. The three processing results are fused to output an enhanced image that eliminates temperature and humidity interference and clearly shows key structures.

[0128] In practical applications, such as surface inspection of industrial valves, noise characteristics are considered for compliant pixels (350, 450). and color shift features Meanwhile, this region was identified as the target structure by Sobel edge detection, namely the thread edge gradient. Next, the noise characteristics will be... Input the corresponding formula for the noise-contrast model and calculate the contrast enhancement parameters. Then based on color shift features Dynamic range extension parameters are generated by calculating the color cast dynamic range model. Then, based on the thread edge gradient identified by edge detection... Sharpening intensity parameters are calculated based on the detail enhancement formula. Finally, the embedded processor synchronously performs image enhancement: contrast enhancement parameters. Stretch histogram, expand parameters according to dynamic range Layered compression of brightness, and based on sharpening intensity parameters The threaded area is sharpened at 1.9 times the intensity; the valve thread geometry is clear and distortion-free in the output image, and the microscopic details of the rust are visible.

[0129] The above-mentioned overall solution 105, based on physical compliance feature data, dynamically generates contrast adjustment, dynamic range expansion and target detail enhancement parameters through an embedded processor, and simultaneously performs adaptive image optimization processing. While suppressing temperature and humidity interference, it significantly improves the structural clarity and texture details of key areas, and outputs high-fidelity enhanced images that meet industrial inspection requirements.

[0130] The following is a complete example for steps 101-105, such as Figure 2 As shown, in the surface temperature monitoring of an industrial reactor, the temperature data of the pixels on the surface of the image sensor is first collected using a micron-level temperature sensor array. The temperature of pixel (200, 300) is 51.8℃, and the temperature of the adjacent pixel (200, 301) is 54.3℃. The ambient humidity of 70%RH is simultaneously acquired. The temperature gradient between the two pixels is then calculated. The coordinates (200, 300), ΔT=2.5, and H=70 are bound to generate environmental parameter matrix entries; and a temperature noise mapping model is fitted based on historical noise data. Predict the noise intensity at this location. .

[0131] The pixel (200, 300) corresponds to the liquid crystal cell M50 by using a pre-generated modulation unit coordinate mapping table; based on the temperature polarization model. Calculate the polarization adjustment angle The drive circuit is based on the voltage conversion formula. The M50 unit is driven to rotate 0.75° to correct the polarization direction of the incident light; after the compensation light path is projected, a corrected image is output to eliminate the deformation of the bolt threads on the surface of the reactor caused by high temperature.

[0132] Next, the corrected original image is input into the dual-path module, where feature extraction is performed on each pixel. For example, pixel (200, 300) corresponds to RGB=(95, 110, 130). When input into the dual-path module, the main path calls the noise model to obtain the corresponding noise feature values. The auxiliary path, combined with H=70%RH, yields the corresponding color shift feature value according to the color shift model. ; Bind pixel coordinates (200, 300) to output feature pair {noise: 0.575dB, color cast: ΔB=91}.

[0133] Then, construct the energy conservation function. Substitute eigenvalues ​​to calculate If the value is within the threshold range [5.0, 7.0], it is deemed compliant; retain the feature data {noise: 0.575dB, color bias: ΔB=91}.

[0134] Finally, contrast parameters are generated using an embedded processor. Dynamic range parameters Based on the gradient of the reactor flange edge Detailed parameters obtained The image enhancement operation is performed synchronously with reference to the above parameters. The histogram is stretched to increase the contrast by 21.25%, the dynamic range is expanded by 18% through layered compression, and the flange edge is sharpened by 1.8 times. The final enhanced image is then output. The microstructure of the bolt meshing surface is clearly visible in the enhanced image, and the high-temperature stains are completely eliminated.

[0135] Figure 3 This application provides a schematic diagram of the structure of an embedded image enhancement system using a large target area image sensor, as shown in the embodiment of the present application. Figure 3 As shown, the system includes:

[0136] The acquisition module 31 is used to acquire the temperature gradient distribution of each pixel of the large target surface image sensor, simultaneously acquire the ambient humidity data, associate the temperature gradient distribution and the ambient humidity data according to spatial coordinates to form an environmental parameter matrix, and construct a temperature noise mapping model based on the environmental parameter matrix.

[0137] The processing module 32 is used to process the incident light signal received by the large target image sensor through the integrated liquid crystal modulator array, adjust the polarization direction of each modulation unit of the liquid crystal modulator array according to the temperature gradient distribution, eliminate optical distortion, and output the distortion-corrected original image signal.

[0138] Extraction module 33 is used to input the original image signal into the dual-path feature extraction module. In the dual-path feature extraction module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features.

[0139] The verification module 34 is used to verify the physical consistency between the noise distribution features and the color shift features. It constructs an energy conservation constraint function through thermodynamic equations, verifies the matching degree between the noise distribution features and the color shift features and the physical response characteristics of the sensor, and filters out feature data that conforms to physical laws.

[0140] The output module 35 is used to perform image enhancement operations based on the feature data, dynamically adjust the contrast and dynamic range of the image through the embedded processor and enhance the detailed features of the target area, and output the optimized enhanced image.

[0141] Figure 3 The embedded image enhancement system of the large target area image sensor described above can perform... Figure 1 The implementation principle and technical effects of the embedded image enhancement method for a large target area image sensor described in the illustrated embodiment will not be repeated here. The specific operation methods of each module and unit in the embedded image enhancement system for a large target area image sensor in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0142] In one possible design, Figure 3 An embedded image enhancement system for a large target area image sensor, as shown in the embodiment, can be implemented as a computing device, such as... Figure 4 As shown, the computing device may include a storage component 41 and a processing component 42;

[0143] The storage component 41 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 42.

[0144] The processing component 42 is used for the above Figure 1 The embodiment describes an embedded image enhancement method for a large target surface image sensor.

[0145] The processing component 42 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0146] Storage component 41 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0147] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0148] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0149] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0150] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0151] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 An embedded image enhancement method for a large target area image sensor is shown in the embodiment.

[0152] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0153] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0154] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An embedded image enhancement method for a large target area image sensor, characterized in that, include: The temperature gradient distribution of each pixel of a large target surface image sensor is collected, and environmental humidity data is acquired simultaneously. The temperature gradient distribution and environmental humidity data are correlated with the environmental humidity data according to spatial coordinates to form an environmental parameter matrix. A temperature noise mapping model is constructed based on the environmental parameter matrix. The incident light signal received by the large target image sensor is processed by an integrated liquid crystal modulator array. The polarization direction of each modulation unit of the liquid crystal modulator array is adjusted according to the temperature gradient distribution to eliminate optical distortion and output the distortion-corrected original image signal. The original image signal is input into the dual-path feature extraction module. In the dual-path feature extraction module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features. The noise distribution characteristics and the color shift characteristics are physically consistent. An energy conservation constraint function is constructed through thermodynamic equations to verify the matching degree between the noise distribution characteristics and the color shift characteristics and the physical response characteristics of the sensor, and feature data that conforms to physical laws are selected. Based on the feature data, an image enhancement operation is performed. The embedded processor dynamically adjusts the contrast and dynamic range of the image and enhances the detailed features of the target area, outputting an optimized enhanced image. The process of dynamically adjusting the contrast and dynamic range of an image using an embedded processor and enhancing the detail features of the target region to output an optimized enhanced image includes: The contrast adjustment parameters are dynamically generated by the embedded processor based on the noise distribution characteristic values ​​in the feature data. The embedded processor dynamically generates dynamic range extension parameters based on the color offset feature values ​​in the feature data. The embedded processor identifies the target region and calculates detail enhancement parameters based on the pixel detail feature values ​​of the target region. The original image signal is processed synchronously in the embedded processor according to the contrast adjustment parameters, dynamic range expansion parameters, and detail enhancement parameters, and an optimized enhanced image is output.

2. The method according to claim 1, characterized in that, The process involves constructing an energy conservation constraint function using thermodynamic equations, verifying the matching degree between the noise distribution characteristics, the color shift characteristics, and the sensor's physical response characteristics, and filtering out feature data that conforms to physical laws, including: An energy conservation constraint function is constructed based on a preset thermodynamic equation, which includes noise distribution characteristics, color shift characteristics, and sensor physical response parameters. The feature energy value of the interaction between the noise distribution feature and the color shift feature is calculated pixel by pixel based on the energy conservation constraint function. The feature energy value of each pixel is matched with the pre-stored threshold range allowed by the sensor's physical response characteristics. Pixel feature data whose feature energy values ​​are within the allowable threshold range are selected as feature data that conform to physical laws.

3. The method according to claim 1, characterized in that, The process of processing the incident light signal received by the large target surface image sensor through an integrated liquid crystal modulator array, adjusting the polarization direction of each modulation unit of the liquid crystal modulator array according to the temperature gradient distribution, eliminating optical distortion, and outputting the distortion-corrected original image signal includes: A liquid crystal modulator array is integrated on the surface of the optical components of the large target image sensor, and the positional correspondence between each modulation unit in the liquid crystal modulator array and the pixel of the image sensor is established to generate a modulation unit coordinate mapping table. Based on the temperature data of each pixel in the temperature gradient distribution, the polarization adjustment parameters of the corresponding modulation unit are determined through the modulation unit coordinate mapping table. The polarization adjustment parameters drive each modulation unit in the liquid crystal modulator array to perform polarization direction adjustment, so that the incident light signal is projected onto the optical component through the adjusted liquid crystal modulator array, forming a compensation optical path to eliminate optical distortion. The original image signal, after distortion correction, is output through the compensation optical path.

4. The method according to claim 3, characterized in that, The step of driving each modulation unit in the liquid crystal modulator array to perform polarization direction adjustment according to the polarization adjustment parameters, so that the incident light signal is projected onto the optical component through the adjusted liquid crystal modulator array to form a compensation optical path to eliminate optical distortion, includes: The polarization adjustment parameters are input into the driving circuit of the liquid crystal modulator array to generate a voltage control signal corresponding to each modulation unit; The polarization direction is adjusted by changing the molecular arrangement direction of each modulation unit in the liquid crystal modulator array according to the voltage control signal. The incident light signal is made to generate a corresponding polarization deflection by passing through the liquid crystal modulator array after the polarization direction is adjusted, thereby generating a polarization-corrected light signal. The polarization-corrected light signal is projected onto the surface of the optical component to form a compensation optical path that eliminates optical distortion.

5. The method according to claim 1, characterized in that, The process involves inputting the original image signal into a dual-path feature extraction module. In this module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, while the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features. The original image signal is input into the dual-path feature extraction module; In the main path of the dual-path feature extraction module, the temperature noise mapping model is called to process the original image signal and extract the temperature-related noise distribution features of each pixel. In the auxiliary path of the dual-path feature extraction module, the original image signal is received synchronously and combined with the environmental humidity data to separate the humidity-induced color shift features of each pixel. The noise distribution features and color offset features maintain the same pixel coordinate mapping relationship.

6. The method according to claim 1, characterized in that, The temperature gradient distribution of each pixel in the large target surface image sensor is acquired, and environmental humidity data is simultaneously obtained. The temperature gradient distribution and environmental humidity data are correlated by spatial coordinates to form an environmental parameter matrix. Based on the environmental parameter matrix, a temperature noise mapping model is constructed, including: Temperature gradient distribution of each pixel of a large target image sensor is collected by a temperature sensor array, and ambient humidity data is acquired synchronously by a humidity sensor. The temperature gradient distribution data corresponding to each pixel is associated and bound with the ambient humidity data according to the two-dimensional spatial coordinates of the pixel. An environmental parameter matrix containing coordinate positions and their corresponding data is generated based on the associated data of all pixels. A temperature noise mapping model is directly constructed based on the mapping relationship between the temperature gradient values ​​in the environmental parameter matrix and the image noise.

7. An embedded image enhancement system for a large target area image sensor, characterized in that, include: The acquisition module is used to acquire the temperature gradient distribution of each pixel of the large target surface image sensor, simultaneously acquire environmental humidity data, associate the temperature gradient distribution and environmental humidity data according to spatial coordinates to form an environmental parameter matrix, and construct a temperature noise mapping model based on the environmental parameter matrix. The processing module is used to process the incident light signal received by the large target image sensor through an integrated liquid crystal modulator array, adjust the polarization direction of each modulation unit of the liquid crystal modulator array according to the temperature gradient distribution, eliminate optical distortion, and output the distortion-corrected original image signal. The extraction module is used to input the original image signal into the dual-path feature extraction module. In the dual-path feature extraction module, the main path extracts temperature-related noise distribution features based on the temperature noise mapping model, and the auxiliary path combines the environmental humidity data to separate humidity-induced color shift features. The verification module is used to verify the physical consistency between the noise distribution characteristics and the color shift characteristics. It constructs an energy conservation constraint function through thermodynamic equations, verifies the matching degree between the noise distribution characteristics and the color shift characteristics and the physical response characteristics of the sensor, and filters out feature data that conforms to physical laws. The output module is used to perform image enhancement operations based on the feature data, dynamically adjust the contrast and dynamic range of the image through the embedded processor and enhance the detailed features of the target area, and output the optimized enhanced image. The process of dynamically adjusting the contrast and dynamic range of an image using an embedded processor and enhancing the detail features of the target region to output an optimized enhanced image includes: The contrast adjustment parameters are dynamically generated by the embedded processor based on the noise distribution characteristic values ​​in the feature data. The embedded processor dynamically generates dynamic range extension parameters based on the color offset feature values ​​in the feature data. The embedded processor identifies the target region and calculates detail enhancement parameters based on the pixel detail feature values ​​of the target region. The original image signal is processed synchronously in the embedded processor according to the contrast adjustment parameters, dynamic range expansion parameters, and detail enhancement parameters, and an optimized enhanced image is output.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement an embedded image enhancement method for a large target surface image sensor as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements an embedded image enhancement method for a large target surface image sensor as described in any one of claims 1 to 6.

Images