An image quality simulation test method and apparatus for imaging systems based on image quality analysis systems

Abstract

This invention relates to the field of image analysis and processing, and particularly to a method and apparatus for simulating and testing the image quality of an imaging system based on an image quality analysis system. This method acquires the design parameter set of the imaging system under test and constructs a simulation calculation model to describe the entire imaging physical process. It constructs a virtual test scene containing a digital test card image and reference illumination parameters, and generates environmental interference parameters independent of the reference illumination parameters based on preset environmental interference conditions. The digital test card image and reference illumination parameters are input into the simulation calculation model for simulation calculation. The simulation test results are then output and fused with the environmental interference parameters to obtain a virtual test image containing the effects of environmental interference. Finally, the virtual test image is input into an image quality analysis system for analysis, and the simulation test results of the imaging system under test under environmental interference conditions are output. This enables the simulation test to simulate non-ideal factors in the real test environment, improving the consistency between the simulation results and the actual test.

Description

Technical Field

[0001] This application relates to the field of image analysis and processing technology, and in particular to an imaging system image quality simulation test method and apparatus based on an image quality analysis system. Background Technology

[0002] Image quality testing of imaging systems primarily relies on physical testing methods. This involves using the imaging system under test to photograph a physical test card under standard lighting conditions, and then analyzing the image quality using an image quality analysis system. This testing method requires the completion of the hardware prototype, resulting in a long testing cycle, high cost, and difficulty in identifying problems in the early stages of development. To address these issues, the industry has begun exploring simulation testing methods for imaging systems. Existing simulation testing methods typically construct optical models, sensor models, and image signal processor models based on the imaging system's design parameters. They then generate simulated images by simulating ideal test card images and importing them into an image quality analysis system for analysis. However, these methods often assume ideal testing conditions, but to achieve realistic simulation testing, these non-ideal factors cannot be ignored, leading to discrepancies between simulation results and actual test results.

[0003] Secondly, some simulation methods that consider environmental interference often employ simple post-processing approaches, typically superimposing various interference factors as a single image post-processing step. This approach fails to reflect the different effects of various interference factors in the imaging chain and struggles to simulate the dynamic changes of interference factors over time, resulting in a discrepancy between simulation results and the actual performance of the imaging system in real-world environments. Therefore, existing technologies have shortcomings and require improvement. Summary of the Invention

[0004] To address one or more problems in the prior art, the main objective of this application is to provide an imaging system image quality simulation test method and apparatus based on an image quality analysis system.

[0005] To achieve the above-mentioned objectives, this application proposes an image quality simulation test method for an imaging system based on an image quality analysis system, the method comprising: In response to the image quality simulation test command of the imaging system, the design parameter set of the imaging system under test is obtained, the design parameter set including optical lens parameters, image sensor parameters and image signal processor parameters; Based on the design parameter set, a simulation calculation model is constructed to describe the physical process of the entire imaging process of the imaging system under test. Construct a virtual test scenario, which includes a digital test card image and reference lighting parameters; Based on the preset types of environmental interference conditions, environmental interference parameters are generated. These environmental interference parameters are used to simulate non-ideal factors in a real test environment. The digital test card image and reference illumination parameters are input into the simulation calculation model, which performs simulation calculations according to the imaging physical process and outputs the simulation test results. Based on the output simulation test results, the environmental interference parameters are fused to obtain a virtual test image that includes the effects of environmental interference; The virtual test image is input into the image quality analysis system for analysis, and the analysis results output by the image quality analysis system are received. The analysis results are used as the simulation test results of the imaging system under the environmental interference conditions.

[0006] This application also provides an imaging system image quality simulation testing device based on an image quality analysis system, including: The acquisition module is used to acquire the design parameter set of the imaging system under test in response to the imaging system image quality simulation test command. The design parameter set includes optical lens parameters, image sensor parameters and image signal processor parameters. The first construction module is used to construct a simulation calculation model based on the design parameter set, which describes the physical process of the full-link imaging of the imaging system under test. The second construction module is used to construct a virtual test scenario, which includes a digital test card image and reference lighting parameters. The generation module is used to generate environmental interference parameters according to the preset environmental interference condition types. The environmental interference parameters are used to simulate non-ideal factors in the real test environment. The input module is used to input the digital test card image and reference illumination parameters into the simulation calculation model, perform simulation calculations according to the imaging physical process through the simulation calculation model, and output the simulation test results. The fusion module is used to fuse the environmental interference parameters based on the output simulation test results to obtain a virtual test image that includes the effects of environmental interference. The receiving module is used to input the virtual test image into the image quality analysis system for analysis, and to receive the analysis results output by the image quality analysis system, and to use the analysis results as the simulation test results of the imaging system under test under the environmental interference conditions.

[0007] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods described above.

[0008] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0009] The imaging system image quality simulation testing method and apparatus based on an image quality analysis system in this application introduces environmental interference parameters independent of the reference illumination parameters, enabling the simulation process to realistically simulate various non-ideal factors objectively existing in the physical testing environment. By generating independent environmental interference parameters and fusing them with ideal simulation results, the final virtual test image simultaneously includes the inherent characteristics of the imaging system and the influence of environmental interference, improving the consistency between simulation results and physical test results. Secondly, this method allows for end-to-end simulation testing on a computer by obtaining the design parameters of the imaging system. Engineers can predict the final imaging effect during the optical design and sensor selection stages, and adjust the design scheme in a timely manner based on the simulation results, shortening the development cycle and reducing trial-and-error costs. Furthermore, this method utilizes image quality analysis system tools to ensure the consistency of simulation results with industry testing standards. By importing the virtual test image into the image quality analysis system for analysis, the image quality evaluation indicators output by this method are compatible with the evaluation system used in physical testing. Engineers can directly compare simulation results with historical measured data, lowering the technical promotion threshold. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating an embodiment of an imaging system image quality simulation test method based on an image quality analysis system according to this application. Figure 2 This is a flowchart illustrating an embodiment of an imaging system image quality simulation test method based on an image quality analysis system according to this application. Figure 3 This is a schematic block diagram of the structure of an imaging system image quality simulation test device based on an image quality analysis system according to an embodiment of this application; Figure 4 This is a schematic block diagram of the structure of a computer device according to an embodiment of this application; Figure 5 This is a comparative schematic diagram of an imaging system image quality simulation test method based on an image quality analysis system according to an embodiment of this application.

[0011] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0013] Reference Figure 1 This application provides a method for simulating and testing the image quality of an imaging system based on an image quality analysis system. The method includes: S1. In response to the imaging system image quality simulation test command, obtain the design parameter set of the imaging system under test, the design parameter set including optical lens parameters, image sensor parameters and image signal processor parameters; S2. Based on the design parameter set, construct a simulation calculation model to describe the physical process of the entire imaging process of the imaging system under test; S3. Construct a virtual test scenario, which includes a digital test card image and reference lighting parameters; S4. Generate environmental interference parameters according to the preset environmental interference condition type. The environmental interference parameters are used to simulate non-ideal factors in the real test environment. S5. Input the digital test card image and reference illumination parameters into the simulation calculation model, perform simulation calculations according to the imaging physical process through the simulation calculation model, and output the simulation test results; S6. Based on the output simulation test results, the environmental interference parameters are fused to obtain a virtual test image including the effects of environmental interference; S7. Input the virtual test image into the image quality analysis system for analysis, and receive the analysis results output by the image quality analysis system. Use the analysis results as the simulation test results of the imaging system under the environmental interference conditions.

[0014] As described in steps S1-S3 above, in S1, when an engineer needs to evaluate the image quality of an imaging system under development, they can send a simulation test command through the operating terminal. Upon receiving the command, the system first reads the complete set of design parameters for the imaging system from the design database. These parameters specifically include three aspects: optical lens parameters such as focal length, F-number, distortion coefficient, and modulation transfer function curves for each field of view; image sensor parameters such as pixel size, quantum efficiency, full-well capacity, dark current noise level, and analog-to-digital conversion bit depth; and image signal processor parameters such as demosaicing algorithm type, noise reduction intensity, sharpening coefficient, color correction matrix, and gamma correction curve. These parameters are the foundational data for subsequently building the simulation model and directly determine the accuracy of the simulation results. After obtaining the design parameters in S2, a simulation calculation model that can completely describe the imaging physical process is constructed based on these parameters. The construction logic of this model is established step-by-step according to the transmission path of light from the object being photographed to the final digital image. Specifically, the model first constructs an optical transfer function based on optical lens parameters to describe how the lens blurs an ideal image; secondly, it constructs a photoelectric conversion model based on image sensor parameters to describe how light is converted into electrical signals and the various noises introduced in this process; finally, it constructs a digital image processing model based on image signal processor parameters to describe how the electrical signals are processed by a series of algorithms to ultimately become a digital image visible to the human eye. These three sub-models are linked together in the order of the imaging physical process to form an end-to-end full-link simulation calculation model. This transforms abstract design parameters into executable mathematical calculation logic. Before performing simulation calculations, S3 needs to construct a standardized test environment. This step generates a virtual test scenario within the computer, mainly including two elements. The first element is a digital test card image. The system generates a high-precision ideal image based on the digital definition of international standard test cards, such as the wedge-shaped line pairs and concentric circle patterns in the ISO 12233 resolution test card, the 24 standard color blocks and their corresponding spectral reflectance data in the ColorChecker color card, and the grayscale gradient areas in the dynamic range test card. These digital test card images are free of noise and distortion, making them ideal inputs for subsequent simulation calculations. The second element is the baseline lighting parameters, which set parameters such as the spectral power distribution, color temperature, illuminance, and uniformity of the standard light source to simulate ideal lighting conditions in the test environment. The combination of these two elements constructs a repeatable and standardized virtual test environment, providing a unified benchmark for cross-sectional comparisons between different imaging systems.

[0015] In addition, the construction process of the optical transfer function (OPF) involves using acquired optical lens parameters, including measured modulation transfer function curves for each field of view, distortion coefficients, and relative illumination data, to construct the OTF through interpolation fitting. If only theoretical data output from optical design software is available during the design phase, this theoretical data is directly used for function modeling. If measured data from a physical lens is available, the measured data is used to calibrate and correct the theoretical model, making the OTF more closely approximate the actual performance of the lens. The construction process of the photoelectric conversion model involves establishing a mathematical expression for photoelectric conversion based on image sensor parameters, including quantum efficiency curves, full-well capacity, dark current noise level, readout noise variance, and analog-to-digital conversion bit depth. The core of this model is to construct the mapping relationship between input light intensity and output digital value, while simultaneously superimposing a noise model that conforms to the physical characteristics of the sensor. The parameters of the noise model are directly derived from the sensor specifications or measured data. For example, dark current noise is modeled using a Poisson distribution based on measured mean and variance, and readout noise is modeled using a Gaussian distribution. The construction process of the digital image processing model involves replicating the ISP processing flow in a simulation environment based on the image signal processor parameters, including the demosaic algorithm type, white balance gain coefficient, color correction matrix, gamma correction curve, and noise reduction intensity coefficient. This model is constructed according to the real ISP processing pipeline, implementing the algorithms of each stage in code and using the acquired parameters as input configurations. For example, the nine coefficients in the color correction matrix are directly substituted into the color correction algorithm, and the gamma correction curve is loaded in the form of a lookup table.

[0016] As described in steps S4-S7 above, in step S4, since the actual test environment is not ideal, various interference factors always exist. The core value of this step lies in introducing these interference factors into the simulation process. Multiple types of environmental interference conditions are pre-defined, such as spectral drift caused by light source aging, sensor thermal noise caused by changes in ambient temperature, and stray light interference caused by imperfections in the test darkroom. For each type of interference, the system generates specific environmental interference parameters based on statistical data or physical models accumulated in actual tests. Taking light source spectral drift as an example, the system determines the statistical distribution range of spectral drift based on measured spectral data of the same model of light source at different stages of its lifespan, and then generates a specific light source spectral drift amount through random sampling. These environmental interference parameters are independent of the reference illumination parameters and participate in the subsequent simulation process as additional parameters. This step breaks the limitation of traditional simulation methods that only focus on the physical characteristics of the imaging system itself, making the simulation results closer to the real situation of actual testing.

[0017] Step S5 performs a full-link simulation calculation under ideal conditions. The digital test card image and reference lighting parameters generated in S3 are input into the simulation calculation model constructed in S2. The model performs calculations sequentially according to the physical order of optical transmission, photoelectric conversion, and digital image processing. Specifically, the model first performs convolution operations on the digital test card image based on the optical lens parameters to simulate the blurring effect of the lens on image details; then, it samples and quantizes the optical image based on the image sensor parameters, adding inherent dark current noise and readout noise from the sensor during the quantization process; finally, it performs a series of processing steps on the raw data output by the sensor, including de-mosaicing, white balance, color correction, and gamma correction, based on the image signal processor parameters, ultimately outputting a digital image as the simulation test result. It is important to note that the simulation test result output at this point is obtained under ideal environmental conditions, without any added environmental interference factors; it represents the theoretical imaging effect of the imaging system under perfect testing conditions. Step S6 combines the ideal simulation results with actual environmental interference. The system acquires environmental interference parameters generated by S4, including stray light intensity coefficient, sensor temperature noise figure, and light source spectral shift, and fuses these parameters with the ideal simulation test results output by S5. In the specific fusion process, the system applies interference from different physical domains sequentially according to the order of the imaging physical processes. First, optical domain interference is applied, generating a stray light distribution map based on the stray light intensity coefficient and superimposing it onto the ideal simulation image to simulate the local spotting or overall fogging effect formed by stray light on the image plane. Then, electrical domain interference is applied, generating a noise image conforming to a specific statistical distribution based on the sensor temperature noise figure and adding it to the image already carrying stray light to simulate the increased noise caused by rising sensor temperature. Finally, digital domain interference is applied, adjusting the RGB channel gain of the image based on the light source spectral shift to simulate the impact of light source spectral changes on color reproduction. Through this series of fusion operations, a virtual test image is finally generated that simultaneously contains the inherent characteristics of the imaging system and the effects of environmental interference. S7 uses image quality analysis tools to quantitatively evaluate the simulation results. The virtual test image generated by S6 is imported into an image quality analysis system, such as the iQsTest software. iQsTest automatically identifies the test card area in the image and calls the corresponding algorithm to perform image quality analysis according to the preset analysis module. For example, for an image containing wedge-shaped line pairs, iQsTest calls the spatial frequency response module to calculate the modulation transfer function value; for an image containing standard color blocks, iQsTest calls the color analysis module to calculate the color difference value; for an image containing grayscale areas, iQsTest calls the noise analysis module to calculate the signal-to-noise ratio and dynamic range. The system receives these analysis results output by iQsTest and outputs them to the engineer as the simulation test results of the imaging system under test under specific environmental interference conditions.Engineers can use these results to determine whether the current design meets the requirements, or to compare the performance differences of different design schemes under the same environmental disturbances.

[0018] It should be added that, in specific implementation, the imaging system image quality simulation testing method described in this invention can be executed by a simulation test server deployed in the R&D environment. This server establishes a data communication connection with the engineer's design terminal, the design parameter database, and the image quality analysis system. There are two main modes for triggering the simulation test process. The first is a manual trigger mode, where, after completing the preliminary design of a module of the imaging system, the engineer can send an imaging system image quality simulation test command to the simulation test server through their design terminal, requesting a preliminary evaluation of the imaging performance of the current design scheme. The second is an automatic trigger mode, where the simulation test server monitors the design parameter database in real time. When it detects a new set of design parameters added to the database or a change in existing design parameters, it automatically generates the current simulation test command and initiates the image quality simulation test process for the updated design scheme, thereby achieving a closed-loop linkage between design iteration and performance verification.

[0019] As described above, by introducing environmental interference parameters independent of the reference illumination parameters, the simulation process can realistically simulate various non-ideal factors objectively existing in the physical testing environment. By generating independent environmental interference parameters and fusing them with ideal simulation results, the final virtual test image simultaneously includes the inherent characteristics of the imaging system and the influence of environmental interference, improving the consistency between simulation results and physical test results. Furthermore, this method allows for the completion of end-to-end simulation testing on a computer by obtaining the design parameters of the imaging system. Engineers can predict the final imaging effect during the optical design and sensor selection stages, and adjust the design scheme in a timely manner based on the simulation results, shortening the development cycle and reducing trial-and-error costs.

[0020] Reference Figure 2 In one embodiment, the step of generating environmental interference parameters based on a preset type of environmental interference condition includes: S41. Obtain the preset environmental interference condition type, wherein the environmental interference condition type includes at least one of light source spectral drift type, sensor thermal noise type and stray light interference type; S42. Based on each type of environmental interference condition, obtain the interference characteristic data corresponding to the type; S43. Based on the interference characteristic data, generate environmental interference parameters that are independent of the reference illumination parameters. The environmental interference parameters include at least one of the following: light source spectral shift, sensor temperature noise figure, and stray light intensity coefficient.

[0021] As described above, select one or more environmental interference condition types from the preset interference type library according to the needs of the actual test scenario. These preset types may include light source spectral drift type, sensor thermal noise type, and stray light interference type. The light source spectral drift type is used to simulate changes in the spectral distribution of lighting equipment caused by aging, temperature changes, or power supply fluctuations in the actual test environment. For example, the filament evaporation of incandescent lamps after prolonged use leads to an increase in color temperature, and the spectrum of LED light sources gradually stabilizes during the preheating stage. These changes cause variations in the spectral characteristics of the light source, thus affecting the color reproduction performance of the imaging system. The sensor thermal noise type is used to simulate the increase in noise generated by the image sensor due to temperature rise during operation. In actual testing, the sensor temperature gradually rises after the camera is turned on, especially during continuous shooting or in high-temperature environments, significantly increasing dark current noise and random noise, leading to a decrease in image signal-to-noise ratio. The introduction of this interference type allows the simulation to reflect the performance differences of the imaging system at different operating temperatures. The stray light interference type is used to simulate interference caused by non-target light sources entering the lens in the test environment. In the actual test darkroom, despite light-blocking measures, stray light may still exist due to factors such as wall reflections, direct light from the light source, or internal reflections from the lens. These stray lights can create localized light spots or overall fogging effects on the image surface, affecting image contrast and sharpness. Based on the test requirements specified by the engineer in the simulation test instructions, or according to the preset test scenario configuration, the type of environmental interference condition to be simulated in this simulation is selected from the above types. This selection process ensures the relevance of the simulation test, preventing the introduction of unnecessary interference factors that complicate the analysis of simulation results. After determining the type of environmental interference to be simulated, basic data describing its characteristics needs to be obtained for each selected interference type. The accuracy and authenticity of this interference characteristic data directly affect the credibility of the simulation results. For the light source spectral drift type, the characteristic data of the light source spectrum changing with time and usage conditions are obtained. This data can come from laboratory-accumulated measured data, such as multiple spectral power distribution curves obtained from spectral tests of the same model of light source at different aging stages; it can also come from the aging characteristic descriptions in the product specifications provided by the light source manufacturer; or it can come from continuous monitoring records of spectral changes of typical light sources during preheating. This data describes the possible range and trend of light source spectrum changes under normal usage conditions. For the sensor thermal noise type, the correlation data between the image sensor noise characteristics and temperature is obtained. Specifically, the system acquires measured dark current noise values ​​from dark-field imaging of the same sensor model at different operating temperatures. These data can be used to establish the correlation between temperature and noise figure. For example, a certain sensor has a dark current noise of 2e- at 25 degrees Celsius, which increases to 5e- at 45 degrees Celsius. These data points constitute the basic description of the temperature noise characteristics.For stray light interference, data on stray light distribution characteristics under specific test environments is obtained. This data can be obtained by calibrating an actual darkroom, for example, by placing a point light source in the darkroom and using imaging equipment to acquire stray light distribution images at different positions and angles, from which the intensity coefficient and spatial distribution characteristics of the stray light are extracted. Alternatively, stray light response data for a specific optical structure at different incident angles can be obtained based on the analysis results of lens stray light characteristics using optical simulation software. The authenticity and representativeness of the interference characteristic data are ensured. Data sources can be historical test records from the laboratory, publicly available industry databases, or data acquired through specially designed calibration experiments. This reflects the actual performance of interference factors in real test environments. Based on the obtained interference characteristic data, specific environmental interference parameters that can be used for subsequent fusion calculations are generated. For light source spectral drift, a specific light source spectral drift amount is generated based on the obtained spectral drift characteristic data. For example, the obtained data shows that after 1000 hours of aging, the spectrum of a certain type of light source generally shifts by 2% to 5% in the red band and by 1% to 3% in the blue band. Based on these statistical characteristics, a specific spectral shift of the light source is generated through random sampling, such as a 3.2% shift in the red band and a 1.8% shift in the blue band, which serves as the spectral shift parameter for this simulation. For sensor thermal noise, a specific sensor temperature noise figure is generated based on the acquired temperature noise correlation data. For example, system data shows that a certain sensor has a noise figure of 1.0 at 25 degrees Celsius, 1.8 at 35 degrees Celsius, and 3.2 at 45 degrees Celsius. A temperature value is selected based on a preset temperature fluctuation range, and the corresponding noise figure is determined. This noise figure will be used to control the intensity of noise addition in subsequent simulations. For stray light interference, a specific stray light intensity coefficient is generated based on the acquired stray light distribution characteristics data. For example, system data shows that in a standard darkroom, the stray light intensity at a specific location from the light source is approximately 0.5% to 2% of the main light source illuminance. Based on these measured statistical values, a specific stray light intensity coefficient is selected for subsequent generation of a stray light distribution map. The reason these generated environmental interference parameters are emphasized to be independent of the reference illumination parameters is that, in a real test environment, the presence and intensity of interference factors have no direct functional relationship with the reference illumination conditions set during the test. Whether the engineer sets the light box to 5500K or 6500K, spectral drift due to light source aging may still occur; whether the reference illuminance is 1000 lux or 500 lux, the increase in noise caused by the rise in sensor temperature exists independently. This independence ensures that the generation of environmental interference parameters is not affected by the value of the reference illumination parameters, and is more consistent with real physical processes.

[0022] In one embodiment, the step of generating environmental interference parameters independent of the reference illumination parameters based on the interference characteristic data includes: Analyze the interference characteristic data and extract the interference feature values ​​corresponding to each type of environmental interference condition; For the type of light source spectral drift, based on the statistical distribution range of the extracted light source spectral drift, a light source spectral drift is generated by random sampling and used as an environmental interference parameter for the type of light source spectral drift. For the sensor thermal noise type, based on the extracted correlation curve between sensor temperature and noise figure, a temperature value is selected within a preset temperature fluctuation range, and the noise figure corresponding to the temperature value is determined based on the correlation curve, which serves as the environmental interference parameter for the sensor thermal noise type. For stray light interference types, based on the measured statistical values ​​of the extracted stray light intensity coefficients, a stray light intensity coefficient is selected as the environmental interference parameter for the stray light interference type. At least one of the generated light source spectral shift, sensor temperature noise figure, and stray light intensity figure is used as an environmental interference parameter independent of the reference illumination parameters.

[0023] As described above, the acquired interference characteristic data is first analyzed to extract feature values ​​that can be directly used to generate environmental interference parameters. For light source spectral drift, the interference characteristic data is usually represented by a set of measured curves or statistical distribution data of the light source spectrum under different conditions. The system extracts the statistical distribution range of the spectral drift through analysis. For sensor thermal noise, the interference characteristic data is represented by the measured noise values ​​of the sensor at different temperatures. The system establishes a correlation curve between temperature and noise figure through analysis. For stray light interference, the interference characteristic data is represented by the stray light distribution information collected during the calibration of the test environment. The system extracts the measured statistical value of the stray light intensity coefficient through analysis. This analysis and extraction process transforms the original, potentially complex interference characteristic data into concise and clear feature values. For light source spectral drift, a specific light source spectral drift is generated through random sampling based on the extracted statistical distribution range of the light source spectral drift. This random sampling method simulates the randomness and uncertainty of light source spectral drift in actual testing, because in a real environment, even light sources from the same batch that have undergone the same aging time will have different values ​​of spectral drift. By using random sampling, the spectral shift of the light source generated in each simulation test may differ, thus simulating individual differences in the light source or changes in spectral characteristics at different aging stages, making the simulation results statistically significant. For sensor thermal noise, a specific temperature value is first selected within a preset temperature fluctuation range. The setting of this fluctuation range needs to consider the actual application scenario of the imaging system under test; for example, the normal operating temperature range of consumer electronics products and automotive cameras differs significantly. After determining the temperature value, the noise figure corresponding to that temperature value is determined based on the extracted temperature-noise correlation curve. The physical significance of this process is that the sensor's operating temperature directly affects its noise level; the higher the temperature, the greater the dark current noise and random noise. By establishing the correlation between temperature and noise, the system can generate a noise figure that conforms to physical laws based on the set operating temperature, allowing the simulation results to reflect performance differences under different operating temperatures. For stray light interference, a stray light intensity coefficient is selected based on the measured statistical value of the extracted stray light intensity coefficient. The selection method can be random sampling or specified according to specific testing needs. For example, when evaluating the impact of stray light in the worst case, the upper limit of the statistical value can be selected; when a large number of simulation tests are needed to obtain statistical results, random sampling can be used. This stray light intensity coefficient will be used to generate a stray light distribution map to simulate the impact of stray light on image quality. It is important to emphasize that these parameters are independent of each other and with respect to the reference illumination parameters. The magnitude of the spectral shift of the light source does not depend on the setting of the reference color temperature, the level of the sensor temperature noise figure does not depend on the intensity of the reference illuminance, and the value of the stray light intensity coefficient does not depend on changes in the reference illumination parameters.This independence ensures that the generation process of environmental interference parameters is not affected by the reference lighting conditions, which is more in line with the objective law that interference factors and main lighting conditions are independent of each other in the real physical world.

[0024] It is worth noting that existing technologies often adopt a one-size-fits-all approach to handle all environmental interference, such as simply adding fixed values ​​of noise or color shift to all types of interference. This approach ignores the inherent physical differences between different interference factors. Spectral drift of the light source affects color reproduction, sensor thermal noise affects the image signal-to-noise ratio, and stray light affects contrast and local details. The mechanisms by which these three types of interference affect image quality are completely different, and therefore, different parameter generation methods should be used. This solution designs differentiated parameter generation strategies for the three main types of interference. For spectral drift of the light source, a random sampling method based on the statistical distribution range is used to reflect the randomness of spectral drift; for sensor thermal noise, a method based on temperature selection and correlation curve lookup is used to reflect the deterministic effect of temperature on noise; for stray light interference, a selection method based on measured statistical values ​​is used to reflect the statistical characteristics of stray light intensity. This differentiated approach makes the simulation results closer to the real physical process.

[0025] In one embodiment, the step of fusing the environmental interference parameters with the output simulation test results to obtain a virtual test image including the effects of environmental interference includes: The simulation test results are obtained, and the simulation test results are digital images output after being processed by the simulation calculation model; The environmental interference parameters were analyzed to obtain the light source spectral shift, sensor temperature noise figure, and stray light intensity coefficient. Based on the stray light intensity coefficient, a stray light distribution map is generated, and the stray light distribution map is superimposed on the digital image to obtain an intermediate image carrying stray light interference; Based on the sensor temperature noise coefficient, a temperature-related noise image is generated, and the noise image is added to the intermediate image carrying stray light interference to obtain an intermediate image carrying noise and stray light interference. Based on the spectral shift of the light source, the RGB channel gain coefficients of the intermediate image carrying noise and stray light interference are adjusted to simulate the effect of light source spectral changes on color reproduction, thereby obtaining a virtual test image that includes the effects of environmental interference.

[0026] As described above, after generating simulation test results under ideal conditions and independent environmental interference parameters, this embodiment fuses these environmental interference parameters with the simulation test results to obtain a virtual test image that reflects the impact of the real test environment. First, the previously output simulation test results are obtained; this result is a digital image generated after processing by the simulation calculation model. This image is a simulated output under ideal environmental conditions, considering only the optical, sensor, and ISP characteristics of the imaging system itself, without any added environmental interference factors. It can be considered as the theoretical imaging effect of the imaging system under test in a perfect test environment. Simultaneously, the previously generated environmental interference parameters are analyzed to extract the specific interference values ​​required for this simulation, including the light source spectral shift, sensor temperature noise figure, and stray light intensity coefficient. These parameters have been generated independently according to the aforementioned method and have no dependency on each other or on the reference illumination parameters; they are the direct input for subsequent fusion operations. Based on the stray light intensity coefficient and stray light interference, the interference in the optical domain, i.e., stray light, is processed first. Based on the analyzed stray light intensity coefficient, a stray light distribution map matching this coefficient is generated. This distribution map is not a simple uniform brightness image, but rather constructed based on the spatial distribution characteristics of stray light in the actual test environment. For example, in a standard darkroom, stray light is often more pronounced at the image edges, or forms localized spots when incident at specific angles. An image reflecting the actual spatial distribution of stray light can be generated by scaling a pre-calibrated stray light distribution template with an intensity coefficient. After generating the stray light distribution map, it is superimposed on the digital image output by S5. This superposition is not a simple addition of pixel values, but a fusion based on the physical characteristics of stray light. For example, superposition may produce a saturation effect for localized spots, while linear superposition is used for uniform fogging. After this step, an intermediate image carrying stray light interference is obtained, which now includes the impact of stray light on contrast and local details. Next, interference in the electrical domain, i.e., sensor thermal noise, is processed. Based on the analyzed sensor temperature noise coefficient, a temperature-related noise image is generated. This noise image is not simply random noise, but is generated based on the statistical characteristics of noise from the sensor at its actual operating temperature. For example, under high-temperature conditions, dark current noise dominates, and its distribution may exhibit spatial correlation; under low-temperature conditions, readout noise dominates, and its distribution is closer to Gaussian white noise. The system generates a noise image that conforms to the noise model corresponding to the temperature noise coefficient. After generating the noise image, it is added to an intermediate image that already carries stray light interference. This addition is a pixel-level superposition, which can be done using additive operations, while considering the correlation between noise and signal. For example, for photon shot noise, its variance is related to signal strength, while dark current noise is independent of the signal. Therefore, an appropriate noise addition method is selected based on the sensor characteristics.This step yields an intermediate image that simultaneously carries stray light interference and noise interference. Finally, the digital domain interference, specifically the impact of light source spectral shift on color reproduction, is addressed. Based on the analyzed spectral shift, the system adjusts the RGB channel gain coefficients of the intermediate image, which already carries the first two types of interference. Spectral shift means that the spectral distribution of the actual illumination light has changed relative to the reference illumination, causing a change in the response ratio of each color channel of the image sensor, ultimately resulting in an overall color cast in the image. Specifically, the compensation coefficient for each color channel can be calculated based on the spectral shift. For example, if the light source spectrum shifts towards red light, it means the energy of the red band increases relatively, resulting in a relatively high pixel value for the red channel in the image. To simulate this effect, the system can appropriately increase the gain of the red channel or decrease the gain of the blue and green channels. The adjustment can be globally uniform or locally adjusted based on the spatial distribution characteristics of the spectral shift. This step ultimately yields a virtual test image that simultaneously contains optical domain stray light interference, electrical domain sensor thermal noise interference, and the effects of digital domain light source spectral shift.

[0027] In one embodiment, the step of fusing the environmental interference parameters with the output simulation test results to obtain a virtual test image including the effects of environmental interference further includes: The sequence of occurrence of imaging physical processes is analyzed, and the fusion order of environmental interference parameters is determined based on the analysis results. The method for determining the fusion order includes: Based on the imaging physical aspects affected by each parameter in the environmental interference parameters, the light source spectral drift, sensor temperature noise figure, and stray light intensity coefficient are respectively classified as digital domain interference parameters, electrical domain interference parameters, and optical domain interference parameters. Based on the physical transmission path of the imaging light rays from incident to imaging, the fusion order is determined as follows: first fuse the optical domain interference parameters, then fuse the electrical domain interference parameters, and finally fuse the digital domain interference parameters.

[0028] As described above, this scheme determines the fusion order of environmental interference parameters based on the physical transmission path of the imaging light from incident to imaging. This physical transmission path is unidirectional and irreversible; the light first undergoes optical imaging, then photoelectric conversion, and finally digital image processing. Therefore, this order cannot be reversed. Consequently, the fusion of environmental interference must also follow this inherent order: first, optical domain interference parameters are fused; then, electrical domain interference parameters are fused; and finally, digital domain interference parameters are fused. The technical significance of this order determination method lies in ensuring the temporal consistency between the application of environmental interference and the imaging physical process. Stray light interference in the optical domain exists before the light enters the sensor, so it must be applied first to ensure that subsequent photoelectric conversion and digital processing are based on the optical image that already contains stray light. Sensor thermal noise in the electrical domain is generated during photoelectric conversion and must be applied after the optical image is formed and before digital processing to simulate the superposition of noise in the electrical signal stage. Spectral drift of the light source in the digital domain affects the final color reproduction and must be applied during digital image processing to simulate the response of white balance and color correction algorithms to spectral changes. Applying digital domain interference before optical domain interference would lead to confusion in the physical logic. The image affected by digital domain interference does not yet include stray light from the optical domain and noise from the electrical domain, while the application of optical domain interference would further affect the already adjusted colors. This sequence would not conform to the actual imaging process.

[0029] Compared with existing technologies, this embodiment often uses a simple superposition approach or applies interference according to a certain empirical order when dealing with multiple interferences, but lacks in-depth explanation of why the order is chosen. This approach is prone to problems when facing complex interference combinations. For example, when a new type of interference needs to be added, the empirical order may not provide effective guidance, resulting in an unreasonable insertion position of the new interference. This solution establishes a sequence determination method based on physical process analysis. As long as the physical links of each interference's influence are clearly defined, its position in the fusion sequence can be determined according to a unified principle, ensuring the scalability and universality of the method.

[0030] In one embodiment, before the step of generating environmental interference parameters according to a preset type of environmental interference condition, the method further includes: When the light source spectral drift and sensor temperature noise figure are generated simultaneously, the spectral energy change ratio corresponding to each color band is determined based on the light source spectral drift. According to the spectral energy change ratio, the noise intensity of the noise image corresponding to the sensor temperature noise coefficient in each color channel is weighted and adjusted.

[0031] As described above, this embodiment introduces a pre-processing step to handle the coupling effect when multiple environmental interferences coexist. It addresses the mutual influence between two interferences: light source spectral drift and sensor thermal noise. Specifically, when both the light source spectral drift and sensor temperature noise figure are generated simultaneously, the spectral energy change ratio for each color band is first determined based on the light source spectral drift. The physical basis of this step is that spectral drift causes a change in the distribution of light energy reaching the sensor across different color bands. For example, when the light source spectrum drifts towards red, the light energy in the red band relatively increases, while the light energy in the blue and green bands relatively decreases; and vice versa. In practice, a standard light source spectral power distribution curve needs to be obtained as a reference. This curve describes the distribution of light source energy with wavelength under reference illumination parameters. Then, this reference curve is transformed based on the generated light source spectral drift to obtain the drifted spectral power distribution curve. Next, the system calculates the integrated energy values ​​of the reference curve and the drifted curve in the red, green, and blue bands, respectively. Dividing the integrated energy value after drift in each color band by the baseline integrated energy value yields the proportion of spectral energy change for each color band. For example, assuming that the spectral drift of the light source generated in a simulation causes a 10% increase in the red band energy, no change in the green band energy, and an 8% decrease in the blue band energy, the system calculates the change proportions as follows: red band 1.1, green band 1.0, and blue band 0.92. These three proportions accurately quantify the impact of spectral drift on the energy distribution of each color channel. After determining the proportions of spectral energy change for each color band, the noise intensity of the noise image corresponding to the sensor temperature noise figure in each color channel is weighted and adjusted according to these proportions. The physical significance of this step is that the noise generated by the sensor is correlated with the received signal strength, especially shot noise, whose variance is proportional to the signal strength. When the spectral drift of the light source causes an increase in the light energy received by a certain color channel, the noise level of that channel will also increase accordingly; conversely, when the light energy decreases, the noise level will decrease. In practice, a baseline noise image is first generated based on the sensor temperature noise figure. This noisy image typically contains three color channels: red, green, and blue. The noise intensity of each channel is initially set to the same level, or a base value based on the sensor's characteristics. The system then applies the previously calculated spectral energy change ratio to the noise intensity of the corresponding color channel. Using the previous example, the noise intensity of the red channel is multiplied by 1.1, the green channel remains unchanged, and the blue channel is multiplied by 0.92. After this weighted adjustment, the noise in the red channel is relatively enhanced, the noise in the blue channel is relatively reduced, and the green channel remains unchanged.After weighted adjustment, the noise intensity of each color channel in the noisy image matches the spectral shifted light energy distribution. When such a noisy image is added to the simulation image during subsequent fusion, it can realistically reflect the impact of changes in the light source spectrum on noise characteristics. For example, with enhanced red light, the red areas of the image are not only brighter, but the noise particles on them are also more noticeable, which is consistent with actual physical phenomena.

[0032] It is worth mentioning that in real physical environments, various interference factors do not exist in isolation; they often have complex interactions. Spectral shifts in the light source alter the distribution of light energy reaching the sensor, and this change in energy distribution directly affects the noise characteristics generated by the sensor. If this coupling relationship is ignored, and the two types of interference are simply generated and applied independently, the intrinsic connection between them will be lost, causing the simulation results to deviate from the actual physical process. Current technologies typically employ decoupling when dealing with multiple environmental interferences, assuming that various interferences are independent and do not affect each other. Spectral shifts in the light source only affect color, and sensor thermal noise only affects the image's roughness; both act independently on the image, and the final effect is a simple superposition of the two. While this approach is acceptable when the interference intensity is weak, it reveals significant problems when the interference intensity is high or when high-precision simulation is required. For example, in the case of strong red light shift, the noise in the red region of the actual captured image will be more pronounced than in other regions. This is due to the increased shot noise caused by the enhanced light signal. If decoupling is used, the noise intensity in the red region is the same as in other regions, making it impossible to reproduce this realistic effect.

[0033] This solution addresses this coupling problem by introducing a weighted adjustment of the spectral energy change ratio and noise intensity. First, the influence of the light source spectral drift on the energy distribution of each color channel is extracted. Then, this influence is transferred to the noise generation stage, ensuring that the noise intensity distribution matches the spectral change. This approach aligns with the physical laws governing the correlation between sensor noise and signal, resulting in more realistic simulation results at the level of detail.

[0034] In one embodiment, the step of generating environmental interference parameters based on a preset type of environmental interference condition further includes: When the environmental interference condition type includes time-varying interference type, obtain the correspondence data between the parameter values ​​and time corresponding to the time-varying interference type; Based on the preset simulation duration and time sampling interval, the parameter values ​​corresponding to each sampling time point are extracted from the corresponding relationship data to generate a sequence of environmental interference parameters that change over time.

[0035] As described above, this embodiment targets environmental interference factors that dynamically change over time, expanding simulation testing from static single-point evaluation to dynamic continuous process evaluation. Specifically, when the preset environmental interference condition types include time-varying interference types, the corresponding parameter values ​​and time-related data for that type of interference are obtained. The physical basis of this step is that many interference factors in real environments are not constant but exhibit regular or random changes over time. For example, during the continuous operation of an automotive camera after vehicle startup, the sensor temperature gradually rises from room temperature to thermal equilibrium, a process that typically lasts for several minutes or even longer. This temperature change leads to a gradual increase in dark current noise and random noise. Similarly, during the preheating phase after LED light sources are turned on, their spectral characteristics undergo a drift from a cold to a hot state, usually requiring several minutes to stabilize. During this process, the spectral distribution of the light source continuously changes, thus affecting color reproduction in imaging. There are various ways to obtain this correlation data. For sensor temperature changes, continuous temperature monitoring of the sensor can be performed in actual testing, recording the temperature change curve over time from startup to stable operation. To detect spectral drift of a light source, spectral data can be continuously collected from the source as it transitions from activation to stabilization using a spectrometer, yielding a curve showing the spectral drift over time. This data can be a sequence of discrete points obtained from actual measurements, a function expression derived through fitting, or a theoretical curve calculated based on a physical model. Regardless of the specific form, this data accurately describes the time-dependent variation of a particular disturbance parameter. After obtaining the correlation data between parameter values ​​and time, the parameter values ​​corresponding to each sampling time point are extracted from this correlation data according to the preset simulation duration and time sampling interval, ultimately generating a time-varying sequence of environmental disturbance parameters. The simulation duration is set based on specific testing requirements. For example, to simulate the performance changes of a vehicle-mounted camera during 30 minutes of continuous operation, the simulation duration can be set to 30 minutes. The time sampling interval is set based on the details of the time changes to be captured and the needs of subsequent analysis. If the focus is on second-level changes, the sampling interval can be set to 1 second; if the focus is on minute-level trends, the sampling interval can be set to 1 minute. A smaller sampling interval results in a more refined parameter sequence, capturing more subtle time changes, but also increases the computational load. After determining the simulation duration and sampling interval, the parameter values ​​corresponding to each time point are extracted one by one from the acquired corresponding data according to the set time nodes. Taking the sensor temperature change as an example, assume that the acquired temperature change data shows that the temperature is 25 degrees Celsius at 0 seconds after power-on, rises to 32 degrees Celsius at 30 seconds, rises to 38 degrees Celsius at 60 seconds, rises to 42 degrees Celsius at 90 seconds, and reaches a stable 45 degrees Celsius at 120 seconds.If the simulation duration is set to 120 seconds and the sampling interval to 30 seconds, the system will extract the temperature values ​​corresponding to five time points, and then determine the noise figure corresponding to each time point based on the temperature-noise correlation curve, ultimately generating a time-varying parameter sequence containing five noise figure values. For light source spectral drift, a time-varying spectral drift sequence can be generated in the same way. For example, a certain LED light source has a spectral drift of 0% at second 0, 1% at second 10, 2.5% at second 30, and reaches a stable 3% at second 60. The system extracts the spectral drift at each time point according to the set sampling interval, generating the corresponding time-varying parameter sequence. After generating the time-varying environmental interference parameter sequence, these parameters will be used sequentially according to the time nodes during simulation calculations. Specifically, during the simulation of each frame, the system extracts the environmental interference parameter value for that time point from the parameter sequence based on the corresponding time point of that frame, and uses it for interference fusion calculation of that frame. The result of this processing is that the final multi-frame virtual test images are no longer independent static images, but constitute an image sequence that reflects the dynamic changes in the imaging system's performance over time. This image sequence can be input into an image quality analysis system for frame-by-frame analysis, yielding curves showing the change of image quality evaluation indicators over time, thereby comprehensively evaluating the performance of the imaging system in dynamic environments. For example, when the temperature of a vehicle-mounted camera sensor gradually increases, analyzing the signal-to-noise ratio (SNR) of each frame in the image sequence yields a curve showing a gradual decrease in SNR over time, clearly demonstrating the degree and rate of change of the impact of temperature increase on image quality. Similarly, regarding spectral drift during the preheating process of an LED light source, analyzing the color difference index of each frame yields a curve showing a gradual decrease and stabilization of color difference over time, reflecting the automatic white balance algorithm's ability to track and correct spectral changes.

[0036] In one feasible embodiment, obtaining the correspondence data between the parameter values ​​and time corresponding to the time-varying interference type includes: When the time-varying interference type is sensor thermal noise type, the sensor temperature change curve over time is obtained, and the temperature change curve is converted into a sensor temperature noise figure change curve over time based on the correlation between sensor temperature and noise figure. When the time-varying interference type is the light source spectral drift type, obtain the curve of the change of the light source spectral drift amount over time.

[0037] In this embodiment, when the time-varying interference type to be simulated is sensor thermal noise, the sensor temperature change curve over time is obtained. This curve describes the evolution of the sensor temperature over time during actual operation. Taking an in-vehicle camera as an example, when the vehicle starts, the sensor temperature is the same as the ambient temperature, assumed to be 25°C. As the camera continues to operate, the internal circuit heats up, causing the sensor temperature to gradually rise. It reaches 32°C in the first minute after power-on, 38°C in the second minute, and 42°C in the third minute, reaching thermal equilibrium and stabilizing at around 45°C after about 5 minutes. A complete temperature-time curve can be obtained through actual measurement or thermal simulation analysis. After obtaining the temperature change curve, it cannot be directly used for noise simulation because the noise figure and temperature need to be correlated through sensor characteristic data. The system converts the temperature change curve into a sensor temperature-noise figure change curve over time based on the pre-obtained correlation between sensor temperature and noise figure. This conversion process is based on measured noise data from the sensor at different temperature points. For example, in a laboratory, temperature control tests are conducted on the same model of sensor, measuring its dark current noise and readout noise at temperatures of 20℃, 30℃, 40℃, and 50℃. A functional relationship between temperature and noise figure is fitted using this data. Substituting the temperature value at each time point on the temperature change curve into this functional relationship yields the noise figure at the corresponding time point, thus forming a curve showing the noise figure changing over time. When the time-varying interference to be simulated is light source spectral drift, the curve showing the change of the light source spectral drift over time is directly obtained. Unlike sensor thermal noise, which requires a two-step conversion, the light source spectral drift itself is a parameter that can be directly used for simulation, making the acquisition process relatively straightforward. Taking the preheating process of an LED light source as an example, when the light source is turned on, its spectral characteristics are significantly different from those during stable operation. As the internal temperature gradually increases, the spectrum slowly drifts until it stabilizes. Actual measurement data shows that, after a certain type of LED light source is turned on, the drift relative to the reference spectrum is 0% at 0 seconds; at 10 seconds, the red light band drifts by 1%; at 30 seconds, the red light band drifts by 2.5%; and at 60 seconds, it reaches a stable state with a 3% drift in the red light band, which remains unchanged thereafter. Plotting these time points with the corresponding drift amounts yields the curve of the light source's spectral drift over time.

[0038] In one feasible embodiment, an image quality simulation test of a mobile phone camera during continuous shooting is conducted. This embodiment aims to compare the differences in performance between existing simulation methods and the method of this invention in simulating continuous shooting by a mobile phone camera. The test object is a rear camera of a mobile phone under development, with design parameters including an optical lens F-number of 1.8, a sensor pixel size of 1.0 micrometer, and an image signal processor employing a standard demosaic algorithm and automatic white balance function. The simulation test simulates the following real-world scenario: a user continuously shoots video with the mobile phone for 120 seconds. During this process, the camera sensor temperature gradually rises from 25°C upon startup to 45°C. Simultaneously, the LED light source in the shooting environment exhibits spectral drift within the first 60 seconds after startup, with the red light band energy gradually increasing by approximately 3%.

[0039] refer to Figure 5 The simulation includes two sets of comparative experiments: The first set uses existing techniques to generate a simulated image under ideal conditions, then applies the same fixed noise and color shift correction to all frames of the entire video. This method does not consider temperature changes over time, nor the coupling relationship between light source spectral drift and noise. The second set uses the proposed method. First, it generates a time-varying sensor temperature noise figure sequence to simulate the dynamic change of noise gradually increasing as the temperature rises from 25°C to 45°C; simultaneously, it generates a time-varying sequence of light source spectral drift to simulate the gradual increase in red light energy within the first 60 seconds. In the noise generation stage, the noise intensity of the red channel is weighted and adjusted according to the light source spectral drift to match the noise change in the red channel with the spectral energy change.

[0040] Reference Figure 3 This application also provides an imaging system image quality simulation testing device based on an image quality analysis system, comprising: The acquisition module 1 is used to acquire the design parameter set of the imaging system under test in response to the imaging system image quality simulation test command. The design parameter set includes optical lens parameters, image sensor parameters and image signal processor parameters. The first construction module 2 is used to construct a simulation calculation model based on the design parameter set, which is used to describe the physical process of the full-link imaging of the imaging system under test; The second construction module 3 is used to construct a virtual test scenario, which includes a digital test card image and reference lighting parameters. The generation module 4 is used to generate environmental interference parameters according to the preset environmental interference condition type. The environmental interference parameters are used to simulate non-ideal factors in the real test environment. Input module 5 is used to input the digital test card image and reference illumination parameters into the simulation calculation model, perform simulation calculations according to the imaging physical process through the simulation calculation model, and output simulation test results; The fusion module 6 is used to fuse the environmental interference parameters based on the output simulation test results to obtain a virtual test image including the effects of environmental interference. The receiving module 7 is used to input the virtual test image into the image quality analysis system for analysis, and to receive the analysis results output by the image quality analysis system, and to use the analysis results as the simulation test results of the imaging system under the environmental interference conditions.

[0041] As described above, it is understood that each component of the imaging system image quality simulation test device based on the image quality analysis system proposed in this application can realize the function of any one of the imaging system image quality simulation test methods based on the image quality analysis system as described above, and the specific structure will not be described in detail.

[0042] Reference Figure 4 This application also provides a computer device, which may be a server, and its internal structure may be as follows: Figure 4 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores monitoring data and other data. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements an image quality simulation test method for an imaging system based on an image quality analysis system.

[0043] The processor described above executes the imaging system image quality simulation test method based on the image quality analysis system, comprising: responding to the imaging system image quality simulation test command, acquiring a set of design parameters for the imaging system under test, the set of design parameters including optical lens parameters, image sensor parameters, and image signal processor parameters; constructing a simulation calculation model based on the set of design parameters to describe the physical process of the entire imaging chain of the imaging system under test; constructing a virtual test scenario, the virtual test scenario including a digital test card image and reference illumination parameters; generating environmental interference parameters based on a preset type of environmental interference conditions, the environmental interference parameters being used to simulate non-ideal factors in a real test environment; inputting the digital test card image and reference illumination parameters into the simulation calculation model, performing simulation calculations according to the imaging physical process through the simulation calculation model, and outputting simulation test results; based on the output simulation test results, fusing the environmental interference parameters to obtain a virtual test image including the influence of environmental interference; inputting the virtual test image into an image quality analysis system for analysis, receiving the analysis results output by the image quality analysis system, and using the analysis results as the simulation test results of the imaging system under test under the environmental interference conditions.

[0044] One embodiment of this application also provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements an image quality simulation test method for an imaging system based on an image quality analysis system, comprising the steps of: responding to an image quality simulation test instruction for an imaging system, acquiring a set of design parameters for the imaging system under test, the set of design parameters including optical lens parameters, image sensor parameters, and image signal processor parameters; constructing a simulation calculation model based on the set of design parameters to describe the physical process of the entire imaging chain of the imaging system under test; constructing a virtual test scenario, the virtual test scenario including a digital test card image and reference illumination parameters; and, according to a preset environment... Interference condition type, generating environmental interference parameters, which are used to simulate non-ideal factors in a real test environment; inputting the digital test card image and reference illumination parameters into the simulation calculation model, performing simulation calculations according to the imaging physics process through the simulation calculation model, and outputting simulation test results; based on the output simulation test results, fusing the environmental interference parameters to obtain a virtual test image including the influence of environmental interference; inputting the virtual test image into an image quality analysis system for analysis, receiving the analysis results output by the image quality analysis system, and using the analysis results as the simulation test results of the imaging system under test under the environmental interference conditions.

[0045] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0046] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0047] The above description is only a preferred embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural changes made based on the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for simulating and testing the image quality of an imaging system based on an image quality analysis system, characterized in that, The method includes: In response to the image quality simulation test command of the imaging system, the design parameter set of the imaging system under test is obtained, the design parameter set including optical lens parameters, image sensor parameters and image signal processor parameters; Based on the design parameter set, a simulation calculation model is constructed to describe the physical process of the entire imaging process of the imaging system under test. Construct a virtual test scenario, which includes a digital test card image and reference lighting parameters; Based on preset environmental interference condition types, environmental interference parameters are generated. These environmental interference parameters are used to simulate non-ideal factors in a real test environment. The environmental interference condition types include at least one of light source spectral drift type, sensor thermal noise type, and stray light interference type. The digital test card image and reference illumination parameters are input into the simulation calculation model, which performs simulation calculations according to the imaging physical process and outputs the simulation test results. Based on the output simulation test results, the environmental interference parameters are fused to obtain a virtual test image that includes the effects of environmental interference; The virtual test image is input into the image quality analysis system for analysis, and the analysis results output by the image quality analysis system are received. The analysis results are used as the simulation test results of the imaging system under the environmental interference conditions. The step of generating environmental interference parameters according to the preset environmental interference condition type includes: when the environmental interference condition type includes a time-varying interference type, obtaining the correspondence data between the parameter values ​​and time corresponding to the time-varying interference type; according to the preset simulation duration and time sampling interval, extracting the parameter values ​​corresponding to each sampling time point from the correspondence data to generate a sequence of environmental interference parameters that change with time. The step of obtaining the correspondence data between the parameter values ​​and time corresponding to the time-varying interference type includes: when the time-varying interference type is sensor thermal noise, obtaining the sensor temperature change curve over time, and converting the temperature change curve into a sensor temperature noise figure change curve over time based on the correlation between sensor temperature and noise figure; when the time-varying interference type is light source spectral drift, obtaining the light source spectral drift amount change curve over time. The simulation test results based on the output are fused with the environmental interference parameters to obtain a virtual test image including the effects of environmental interference. The steps include: analyzing the occurrence sequence of the imaging physical process, and determining the fusion order of the environmental interference parameters based on the analysis results. The method for determining the fusion order includes: classifying the light source spectral drift, sensor temperature noise figure, and stray light intensity coefficient into digital domain interference parameters, electrical domain interference parameters, and optical domain interference parameters respectively, based on the imaging physical links affected by each parameter in the environmental interference parameters; and determining the fusion order based on the physical transmission path of the imaging light from incident to imaging as follows: first fuse the optical domain interference parameters, then fuse the electrical domain interference parameters, and finally fuse the digital domain interference parameters. Before the step of generating environmental interference parameters according to the preset type of environmental interference conditions, the method further includes: when simultaneously generating the light source spectral drift and the sensor temperature noise coefficient, determining the spectral energy change ratio corresponding to each color band according to the light source spectral drift; and weighting the noise intensity of the noise image corresponding to the sensor temperature noise coefficient in each color channel according to the spectral energy change ratio.

2. An imaging system image quality simulation testing device based on an image quality analysis system, used to perform the method of claim 1, characterized in that, include: The acquisition module is used to acquire the design parameter set of the imaging system under test in response to the imaging system image quality simulation test command. The design parameter set includes optical lens parameters, image sensor parameters and image signal processor parameters. The first construction module is used to construct a simulation calculation model based on the design parameter set, which describes the physical process of the full-link imaging of the imaging system under test. The second construction module is used to construct a virtual test scenario, which includes a digital test card image and reference lighting parameters. The generation module is used to generate environmental interference parameters according to the preset environmental interference condition types. The environmental interference parameters are used to simulate non-ideal factors in the real test environment. The input module is used to input the digital test card image and reference illumination parameters into the simulation calculation model, perform simulation calculations according to the imaging physical process through the simulation calculation model, and output the simulation test results. The fusion module is used to fuse the environmental interference parameters based on the output simulation test results to obtain a virtual test image that includes the effects of environmental interference. The receiving module is used to input the virtual test image into the image quality analysis system for analysis, and to receive the analysis results output by the image quality analysis system, and to use the analysis results as the simulation test results of the imaging system under test under the environmental interference conditions.

3. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method of claim 1.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method of claim 1.

Images