Image correction optimization method, system and device and storage medium

Abstract

The invention discloses an image correction optimization method, system and device, and a storage medium, and the key points of the technical scheme are that the method comprises the steps: determining a first distortion parameter and a first image attribute of a current character image in an image obtaining device; screening out secondary processing records from the historical image processing records according to the first image attribute; generating a first correction coefficient according to a correction rule that historical correction values of automatic white balance in all the secondary processing records change along with time; determining an actual correction value of automatic white balance of the current character image, determining a predicted correction value of the current character image according to a correction rule, and generating a second correction coefficient according to a deviation between the actual correction value and the predicted correction value; and correcting the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain a second distortion parameter. According to the method, the actual equipment state and the real-time deviation are accurately reflected through the automatic white balance correction value, the distortion parameters are adjusted according to the automatic white balance correction value, and the accuracy and robustness of image preprocessing are improved.

Description

Technical Field

[0001] The present invention belongs to the field of image processing technology, and in particular relates to an image correction optimization method, system, device and storage medium. Background Art

[0002] Image correction technology is currently widely used in fields such as document digitization, automated surveillance, and intelligent transportation. In particular, image preprocessing is crucial for ensuring the accuracy of subsequent text recognition in machine vision systems. Existing technologies primarily rely on fixed algorithms or simple adaptive methods to correct geometric distortion in images. However, these methods often process single image data and fail to fully account for the impact of image quality degradation caused by acquisition equipment performance degradation and environmental changes.

[0003] However, because image analysis devices and image acquisition devices typically belong to different systems, it's difficult for the image analysis device to directly obtain the device's status information. Consequently, existing image correction methods are unable to dynamically compensate for local distortion caused by device performance degradation during the preprocessing phase. Existing applications of automatic white balance correction often rely on static parameter settings or simple real-time adjustments, failing to leverage time-series data from historical image processing records to analyze and predict long-term device performance changes. Consequently, the processing process cannot accurately reflect actual device status and immediate deviations. Summary of the Invention

[0004] The purpose of the present invention is to provide an image correction optimization method, system, device and storage medium, which accurately reflect the actual device status and real-time deviation through automatic white balance correction values, and adjust distortion parameters accordingly to improve the accuracy and robustness of image preprocessing.

[0005] A first aspect of the present invention provides an image correction optimization method, comprising:

[0006] determining a first distortion parameter and a first image attribute of a current text image, wherein the current text image originates from an image acquisition device;

[0007] Acquire historical image processing records of an image analysis device, and filter out secondary processing records from the historical image processing records according to the first image attribute;

[0008] generating a first correction coefficient according to a correction rule of changes in historical correction values of automatic white balance in all secondary processing records over time;

[0009] determining an actual correction value of the automatic white balance of the current text image, determining a predicted correction value of the current text image according to the correction rule, and generating a second correction coefficient according to a deviation between the actual correction value and the predicted correction value;

[0010] The first distortion parameter is corrected according to the first correction coefficient and the second correction coefficient to obtain a second distortion parameter.

[0011] In some embodiments, the historical image processing record includes: a second image attribute of the historical text image, and filtering out the secondary processing record from the historical image processing record according to the first image attribute includes:

[0012] performing vectorization processing on the first image attribute to obtain a first attribute vector, and performing vectorization processing on the second image attribute to obtain a second attribute vector;

[0013] Calculating the similarity between the first attribute vector and the second attribute vector to obtain attribute similarity;

[0014] The historical image processing records whose attribute similarity reaches the similarity threshold are used as secondary processing records.

[0015] In some embodiments, the image attributes include brightness, contrast, color distribution, noise level, local spectral dispersion, and sharpness; and calculating the similarity between the first attribute vector and the second attribute vector to obtain the attribute similarity includes:

[0016] Normalizing the first attribute vector to obtain a first brightness value, a first contrast value, a first color distribution value, a first noise level value, a first local spectral dispersion value, and a first sharpening intensity value;

[0017] Normalizing the second attribute vector to obtain a second brightness value, a second contrast value, a second color distribution value, a second noise level value, a second local spectral dispersion value, and a second sharpening intensity value;

[0018] Calculate the Euclidean distance between the first brightness value, the first contrast value, the first color distribution value, the first noise level value, the first local spectral dispersion value and the first sharpening intensity value, and the second brightness value, the second contrast value, the second color distribution value, the second noise level value, the second local spectral dispersion value and the second sharpening intensity value to obtain attribute similarity.

[0019] In some embodiments, generating the first correction coefficient according to a correction rule of the historical correction values of the automatic white balance in all secondary processing records changing over time includes:

[0020] Arrange the historical correction values of the automatic white balance in all secondary processing records in the order of timestamps to obtain a secondary correction value sequence;

[0021] For the secondary correction value sequence, statistical fitting is used to draw a correction curve of the automatic white balance correction value changing over time to obtain a correction rule;

[0022] The average slope of the calibration curve is calculated and used as a first calibration coefficient.

[0023] In some embodiments, determining the predicted correction value of the current text image according to the correction rule includes:

[0024] Determining a time point for collecting the current text image;

[0025] The correction rule is retrieved, and the acquisition time point is substituted into the correction rule to obtain a predicted correction value corresponding to the current text image.

[0026] In some embodiments, the secondary correction value sequence is subjected to statistical fitting to draw a correction curve showing the automatic white balance correction value changing over time to obtain a correction rule, including:

[0027] Drawing a scatter plot of automatic white balance correction values changing over time according to the secondary correction value sequence;

[0028] Calculating the autocorrelation of the scatter plot, and determining data characteristics based on the autocorrelation of the scatter plot;

[0029] selecting a corresponding statistical fitting model for the automatic white balance correction value according to the data characteristics;

[0030] Solving the parameters of the statistical fitting model to obtain model parameters;

[0031] A calibration curve is drawn according to the statistical fitting model and the model parameters to obtain a calibration rule.

[0032] In some embodiments, correcting the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain the second distortion parameter includes:

[0033] Performing a weighted sum calculation on the first correction coefficient and the second correction coefficient to obtain a weighted value;

[0034] Adding a preset adjustment factor to the weighted value to obtain a comprehensive correction factor;

[0035] The first distortion parameter is multiplied by a comprehensive correction factor to obtain a second distortion parameter.

[0036] A second aspect of the present invention provides an image correction optimization system, comprising:

[0037] a parameter attribute determination module, configured to determine a first distortion parameter and a first image attribute of a current text image, wherein the current text image originates from an image acquisition device;

[0038] an acquisition and screening module, configured to acquire historical image processing records of an image analysis device, and screen out secondary processing records from the historical image processing records according to the first image attribute;

[0039] A first coefficient generating module, configured to generate a first correction coefficient according to a correction rule of the historical correction values of the automatic white balance in all secondary processing records changing over time;

[0040] a second coefficient generating module, configured to determine an actual correction value of the automatic white balance of the current text image, determine a predicted correction value of the current text image according to the correction rule, and generate a second correction coefficient according to a deviation between the actual correction value and the predicted correction value;

[0041] The parameter correction module is used to correct the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain the second distortion parameter.

[0042] A third aspect of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0043] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0044] The technical solution provided by the present invention has the following advantages and effects: By integrating historical image processing records and real-time acquisition data, first and second correction coefficients are dynamically generated, thereby accurately correcting the first distortion parameter. This solution utilizes time-series analysis and prediction of automatic white balance correction amplitude values to effectively reflect the long-term performance degradation and immediate deviations of the target image acquisition device, compensating for the inability of image analysis devices to directly obtain source device status information. This greatly improves the accuracy and robustness of image preprocessing, thereby enhancing the accuracy and stability of text recognition. The overall solution is simple to implement, effective, and highly adaptable, effectively promoting the application of image processing technology in areas such as document digitization and automated monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 1 is a flow chart of the image correction optimization method provided by the present invention;

[0046] Figure 2 It is a structural block diagram of the image correction optimization system provided by the present invention;

[0047] Figure 3 It is a diagram of the internal structure of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] To facilitate understanding of the present invention, specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.

[0049] Unless otherwise specified or defined, the "first, second..." used in this article is only used to distinguish names and does not represent a specific quantity or order.

[0050] Unless stated otherwise or defined otherwise, the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0051] It should be noted that, in this document, “fixed to” or “connected to” may mean directly fixing or connecting to an element, or indirectly fixing or connecting to an element.

[0052] like Figure 1 As shown, this embodiment provides an image correction optimization method, including the following steps S1 to S5:

[0053] Step S1: Determine a first distortion parameter and a first image attribute of a current text image, wherein the current text image originates from an image acquisition device.

[0054] In practical applications, the target image acquisition device is responsible for acquiring the original image. For example, the target image acquisition device can be a digital camera, a smartphone camera, or a scanner.

[0055] Specifically, a current text image is obtained from a target image acquisition device. The current text image refers to the image currently being acquired and used for text recognition processing. The corresponding first distortion parameters refer to a set of quantitative parameters generated after a preliminary assessment of the geometric distortion present in the text image. These parameters are used to describe the uncertainty of the distortion mapping in the image. The set of quantitative parameters can be calculated using existing image processing algorithms. By analyzing the current text image, first image attributes are extracted from it. The image attributes are a set of quantitative feature parameters that can objectively reflect the image acquisition conditions and the state of the target image acquisition device, including at least brightness, contrast, color distribution, noise level, local spectral dispersion, and edge sharpness. The selected image attributes can largely reflect changes in the acquisition environment and device state, thereby ensuring accuracy in subsequent comparisons with historical data.

[0056] Step S2: Obtain historical image processing records of the image analysis device, and filter out secondary processing records from the historical image processing records according to the first image attribute.

[0057] In practical applications, image analysis devices primarily recognize text within text images. For example, these devices can be dedicated OCR servers, embedded text recognition modules, or fully-fledged intelligent visual processing robots. Image analysis devices and image acquisition devices typically exist as independent modules. Because they belong to different systems, the image analysis device cannot directly or conveniently obtain detailed status information from the image acquisition device.

[0058] Specifically, historical image processing records are obtained through the storage system inside the image analysis device. The historical image processing records include at least the automatic white balance correction value, timestamp, second image attributes and other processing log information related to image correction recorded during each image processing process. The historical image processing records provide data support for the subsequent extraction of device performance change trends based on time series analysis.

[0059] By screening out a predetermined number of secondary processing records that are consistent with the attributes of the current text image, it is possible to ensure that the acquisition conditions of the selected historical text image are highly matched with the acquisition conditions of the current image, thereby improving the accuracy of the correction coefficient.

[0060] Specifically, the filtering out of secondary processing records from historical image processing records according to the first image attribute includes:

[0061] performing vectorization processing on the first image attribute to obtain a first attribute vector, and performing vectorization processing on the second image attribute to obtain a second attribute vector;

[0062] Calculating the similarity between the first attribute vector and the second attribute vector to obtain attribute similarity;

[0063] The historical image processing records whose attribute similarity reaches the similarity threshold are used as secondary processing records.

[0064] In practical applications, the first image attributes (first brightness, first contrast, first color distribution, first noise level, first local spectral dispersion, and first sharpening intensity) are vectorized to produce the first brightness vector, first contrast vector, first color distribution vector, first noise level vector, first local spectral dispersion vector, and first sharpening intensity vector. The second image attributes (second brightness, second contrast, second color distribution, second noise level, second local spectral dispersion, and second sharpening intensity) are vectorized to produce the second brightness vector, second contrast vector, second color distribution vector, second noise level vector, second local spectral dispersion vector, and second sharpening intensity vector, facilitating subsequent calculations. Data comparison and similarity analysis ensure that the selected secondary processing records closely match the current text image in terms of image acquisition conditions and device status, providing a reliable data foundation for subsequent time series-based trend analysis.

[0065] Specifically, calculating the similarity between the first attribute vector and the second attribute vector to obtain attribute similarity includes:

[0066] Normalizing the first attribute vector to obtain a first brightness value, a first contrast value, a first color distribution value, a first noise level value, a first local spectral dispersion value, and a first sharpening intensity value;

[0067] Normalizing the second attribute vector to obtain a second brightness value, a second contrast value, a second color distribution value, a second noise level value, a second local spectral dispersion value, and a second sharpening intensity value;

[0068] Calculate the Euclidean distance between the first brightness value, the first contrast value, the first color distribution value, the first noise level value, the first local spectral dispersion value and the first sharpening intensity value, and the second brightness value, the second contrast value, the second color distribution value, the second noise level value, the second local spectral dispersion value and the second sharpening intensity value to obtain attribute similarity.

[0069] In practical applications, the first brightness vector, first contrast vector, first color distribution vector, first noise level vector, first local spectral dispersion vector, and first sharpening intensity vector are normalized to obtain a first brightness value, a first contrast value, a first color distribution value, a first noise level value, a first local spectral dispersion value, and a first sharpening intensity value. The second brightness vector, second contrast vector, second color distribution vector, second noise level vector, second local spectral dispersion vector, and second sharpening intensity vector are normalized to obtain a second brightness value, a second contrast value, a second color distribution value, a second noise level value, a second local spectral dispersion value, and a second sharpening intensity value. Normalizing the first attribute vector and the second attribute vector eliminates scale differences between the attributes and prevents similarity calculations from being affected by these scale differences. Attribute similarity between the first and second attribute vectors is calculated using Euclidean distance. The selected brightness, contrast, color distribution, noise level, local spectral dispersion, and sharpening intensity values can largely reflect changes in the acquisition environment and device status, thereby ensuring the accuracy of the attribute similarity calculation between the current text image and historical text images.

[0070] Step S3: Generate a first correction coefficient according to the correction rule of the historical correction values of the automatic white balance in all secondary processing records changing over time.

[0071] In practical applications, after long-term use, image acquisition devices can experience aging issues with their sensor and lens components. The automatic white balance correction value, or white balance offset, can directly reflect the overall performance of the device's sensor and lens, including optical component aging and sensor response variations. When device performance degrades, both automatic color temperature compensation and image geometric distortion correction require more significant adjustments. Therefore, the white balance offset can serve as a comprehensive indicator of device health, enabling dynamic adjustment of the first distortion parameter. In this method, the first distortion parameter can be calculated using existing spatial coordinate transformation algorithms (such as perspective correction and lens distortion correction models). The correction coefficients generated based on historical automatic white balance correction values are only used to fine-tune the confidence interval or step size estimated by the distortion correction model. In other words, rather than directly using the color correction values for geometric transformation, they serve as a reference for device performance changes, guiding the geometric distortion correction algorithm to appropriately increase (or decrease) the correction of potential distortion deviations during parameter optimization.

[0072] Specifically, generating the first correction coefficient according to the correction rule of the historical correction values of the automatic white balance in all the secondary processing records changing over time includes:

[0073] Arrange the historical correction values of the automatic white balance in all secondary processing records in the order of timestamps to obtain a secondary correction value sequence;

[0074] For the secondary correction value sequence, statistical fitting is used to draw a correction curve of the automatic white balance correction value changing over time to obtain a correction rule;

[0075] The average slope of the calibration curve is calculated and used as a first calibration coefficient.

[0076] In practical applications, the automatic white balance correction values in all secondary processing records are sorted in chronological order. This means that each automatic white balance correction value extracted from the secondary processing record is associated with its corresponding timestamp, resulting in a secondary correction value sequence to ensure data sorting accuracy and temporal consistency. A time series analysis is performed on the secondary correction value sequence. A statistical fitting method is used to fit these correction values to the time data, creating a trend curve reflecting the change in automatic white balance correction values over time. This curve, a calibration curve, is then used to visually demonstrate the pattern of automatic compensation changes during long-term device operation. The average slope of this calibration curve is then calculated, representing the average change in automatic white balance correction values per unit time, reflecting the degradation or drift trend of device performance.

[0077] The average slope of the correction curve is used as the first correction coefficient. This is because changes in device performance often lead to changes in the automatic white balance compensation requirements, and the average slope of the correction curve quantifies the rate of change of device performance. Using it as the first correction coefficient helps to dynamically compensate for the impact of device performance degradation when subsequently correcting the uncertainty parameters of the initial distortion mapping, thereby improving the accuracy and robustness of the entire image correction process; in addition, since the image analysis device cannot directly obtain the performance change information of the image acquisition device of the image it processes, the average slope of the correction curve obtained through historical data analysis can indirectly reflect the performance degradation of the image acquisition device, providing an effective reference for subsequent dynamic correction, thereby further improving the stability and reliability of the image preprocessing parameters.

[0078] Step S4: determining an actual correction value of the automatic white balance of the current text image, determining a predicted correction value of the current text image according to the correction rule, and generating a second correction coefficient according to a deviation between the actual correction value and the predicted correction value.

[0079] Specifically, determining the predicted correction value of the current text image according to the correction rule includes:

[0080] Determining a time point for collecting the current text image;

[0081] The correction rule is retrieved, and the acquisition time point is substituted into the correction rule to obtain a predicted correction value corresponding to the current text image.

[0082] In actual applications, the current text image is analyzed through an image processing algorithm to determine the amplitude value of the actual application of automatic white balance correction in the current text image, that is, the actual correction value. The actual correction value can be obtained by reading the parameters output by the image preprocessing module. At the same time, the image metadata or system timestamp is used to determine the acquisition time point of the image, thereby ensuring that the time information required for subsequent processing is accurate.

[0083] By calling the correction rule, taking the acquisition time point of the current text image as input, substituting it into the correction rule, i.e. the correction curve, the automatic white balance correction value corresponding to the acquisition time point is predicted, i.e. the predicted correction value.

[0084] Specifically, the secondary correction value sequence is subjected to statistical fitting to draw a correction curve showing the automatic white balance correction value changing over time, thereby obtaining a correction rule, including:

[0085] Drawing a scatter plot of automatic white balance correction values changing over time according to the secondary correction value sequence;

[0086] Calculating the autocorrelation of the scatter plot, and determining data characteristics based on the autocorrelation of the scatter plot;

[0087] selecting a corresponding statistical fitting model for the automatic white balance correction value according to the data characteristics;

[0088] Solving the parameters of the statistical fitting model to obtain model parameters;

[0089] A calibration curve is drawn according to the statistical fitting model and the model parameters to obtain a calibration rule.

[0090] In practical applications, a scatter plot of the automatic white balance correction value over time is drawn based on the secondary correction value sequence, with time as the horizontal axis and the automatic white balance correction value as the vertical axis. Autocorrelation is used to measure the correlation between data at different time points. If the autocorrelation coefficient shows periodic changes, that is, it has significant positive or negative values at certain fixed lag periods, the data characteristics are judged to be periodic. If the autocorrelation coefficient is close to 0 in most lag periods, the data characteristics are judged to be random. If the autocorrelation coefficient slowly decays with the increase of the lag period, the data is judged to have a trend. Then, a corresponding statistical fitting model is selected based on the data characteristics. For example, if the data has a trend, a linear regression model can be selected, and the secondary correction value sequence is used to solve the parameters of the linear regression model to obtain the corresponding model parameters. A correction curve is drawn based on the statistical fitting model and model parameters to obtain the correction law.

[0091] In this application, the actual correction value is compared with the predicted correction value to calculate the deviation between the two. This deviation reflects the difference between the actual compensation of the device and the historical trend expectation, and thus reflects the possible performance anomaly or degradation of the image acquisition device at the current moment. Using this deviation as a second correction coefficient to correct the uncertainty parameters of the initial distortion mapping can further improve the accuracy and robustness of image correction based on dynamic compensation for changes in device performance. This has the advantage of being able to reflect instantaneous fluctuations in device status in real time, thereby providing more stable and reliable preprocessing parameters for image analysis equipment.

[0092] Step S5: Correct the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain a second distortion parameter.

[0093] Specifically, the correcting the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain the second distortion parameter includes:

[0094] Performing a weighted sum calculation on the first correction coefficient and the second correction coefficient to obtain a weighted value;

[0095] Adding a preset adjustment factor to the weighted value to obtain a comprehensive correction factor;

[0096] The first distortion parameter is multiplied by a comprehensive correction factor to obtain a second distortion parameter.

[0097] In practical applications, the step of correcting the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain the second distortion parameter can be expressed by a correction formula, which is:

[0098]

[0099] Among them, U corrected Refers to the second distortion parameter, U inital Refers to the first distortion parameter, S refers to the first correction coefficient, that is, the average slope of the automatic white balance correction value over time, K1 refers to the adjustment weight corresponding to the first correction coefficient, A refers to the actual correction value, and P refers to the predicted correction value. Refers to the second correction coefficient, that is, the deviation of the actual correction value compared to the predicted correction value, K2 refers to the adjustment weight corresponding to the second correction coefficient, and 1 refers to the preset adjustment factor.

[0100] In an embodiment of the present invention, by combining the first correction coefficient and the second correction coefficient to correct the first distortion parameter, the long-term degradation trend and real-time deviation information of the image acquisition device performance can be captured simultaneously, so that the final correction factor is closer to the actual acquisition conditions and a better correction effect is achieved.

[0101] The first correction coefficient is obtained by statistically fitting the time evolution trend of the automatic white balance correction value in the secondary image processing record and calculating the average slope. It reflects the overall performance change trend of the image acquisition device over a long period of time. The second correction coefficient is generated by comparing the deviation between the actual correction value of the current text image and the correction value obtained based on historical trend prediction. It directly reveals the immediate difference between the current state of the image acquisition device and the expected state.

[0102] The first correction coefficient and the second correction coefficient are closely linked during the generation process: on the one hand, the first correction coefficient provides background information on the long-term performance evolution and provides a stable benchmark for the overall correction; on the other hand, the second correction coefficient supplements the real-time deviation. The combination of the two can more comprehensively correct the distortion mapping uncertainty caused by equipment degradation or environmental changes, and improve the text recognition accuracy and stability of image analysis equipment. This combination avoids the limitation that a single parameter cannot fully reflect the equipment status, and realizes dynamic real-time compensation.

[0103] For example, suppose that the current text image captured by a certain image acquisition device has a first distortion parameter U corrected 0.8, while the average slope S of the automatic white balance correction value obtained through historical data analysis is 0.05; at the same time, the actual correction value A of the current text image is 1.2, and the predicted correction value P obtained based on trend prediction is 1.0. The adjustment weight coefficients K1 and K2 are set to 0.8 and 0.6 respectively. According to the formula It can be calculated that: K1×S is 0.04, is 0.2, is 0.12, so the comprehensive correction factor is 1.16, and the final first distortion parameter is 0.8 multiplied by 1.16, which is 0.928.

[0104] During this implementation, the correction formula is retrieved and the first distortion parameter is modified based on the two correction coefficients to obtain the second distortion parameter. This second distortion parameter is then applied to the text recognition module of the image analysis device. As can be seen from the above example, the first correction coefficient captures device degradation through long-term trends, while the second correction coefficient reflects the immediate deviation between the actual and predicted values. The combination of the two not only improves the accuracy of the correction parameters but also enables the image analysis device to obtain more stable preprocessing parameters under different acquisition conditions, ultimately improving text recognition performance.

[0105] The image correction optimization method of the present invention dynamically generates first and second correction coefficients by integrating historical image processing records and real-time data acquisition, thereby accurately correcting the first distortion parameter. This solution utilizes time-series analysis and prediction of automatic white balance correction amplitude values to effectively reflect the long-term performance degradation and immediate deviations of image acquisition devices, compensating for the inability of image analysis devices to directly obtain device status information. This significantly improves the accuracy and robustness of image preprocessing, thereby enhancing the accuracy and stability of text recognition. The overall solution is simple to implement, effective, and highly adaptable, effectively promoting the application of image processing technology in areas such as document digitization and automated monitoring.

[0106] like Figure 2 As shown, an embodiment of the present invention further provides an image correction optimization system, comprising:

[0107] a parameter attribute determination module 10, configured to determine a first distortion parameter and a first image attribute of a current text image, wherein the current text image originates from an image acquisition device;

[0108] an acquisition and screening module 20 for acquiring historical image processing records of an image analysis device and screening out secondary processing records from the historical image processing records according to the first image attribute;

[0109] A first coefficient generating module 30 is configured to generate a first correction coefficient according to a correction rule of the historical correction values of the automatic white balance in all secondary processing records changing over time;

[0110] a second coefficient generating module 40 for determining an actual correction value of the automatic white balance of the current text image, determining a predicted correction value of the current text image according to the correction rule, and generating a second correction coefficient according to a deviation between the actual correction value and the predicted correction value;

[0111] The parameter correction module 50 is configured to correct the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain a second distortion parameter.

[0112] Each module of the aforementioned image correction and optimization system may be implemented in whole or in part through software, hardware, or a combination thereof. Each of the aforementioned modules and units may be embedded in or independent of a processor within a computer device in hardware form, or may be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0113] like Figure 3 As shown, an embodiment of the present invention discloses a computer device, including a memory and a processor, wherein the memory stores a computer program;

[0114] The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements the image correction optimization method described in the above embodiments.

[0115] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0116] An embodiment of the present invention further discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the image correction optimization method described in the above embodiments.

[0117] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application 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 various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0118] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. An image correction optimization method, characterized in that: include: determining a first distortion parameter and a first image attribute of a current text image, wherein the current text image originates from an image acquisition device; Acquire historical image processing records of an image analysis device, and filter out secondary processing records from the historical image processing records according to the first image attribute; generating a first correction coefficient according to a correction rule of changes in historical correction values of automatic white balance in all secondary processing records over time; determining an actual correction value of the automatic white balance of the current text image, determining a predicted correction value of the current text image according to the correction rule, and generating a second correction coefficient according to a deviation between the actual correction value and the predicted correction value; The first distortion parameter is corrected according to the first correction coefficient and the second correction coefficient to obtain a second distortion parameter.

2. The image correction optimization method according to claim 1, wherein: The historical image processing record includes: a second image attribute of the historical text image; and filtering out the secondary processing record from the historical image processing record according to the first image attribute includes: performing vectorization processing on the first image attribute to obtain a first attribute vector, and performing vectorization processing on the second image attribute to obtain a second attribute vector; Calculating the similarity between the first attribute vector and the second attribute vector to obtain attribute similarity; The historical image processing records whose attribute similarity reaches the similarity threshold are used as secondary processing records.

3. The image correction optimization method according to claim 2, wherein: Image attributes include: brightness, contrast, color distribution, noise level, local spectral dispersion, and sharpness; calculating the similarity between the first attribute vector and the second attribute vector to obtain attribute similarity includes: Normalizing the first attribute vector to obtain a first brightness value, a first contrast value, a first color distribution value, a first noise level value, a first local spectral dispersion value, and a first sharpening intensity value; Normalizing the second attribute vector to obtain a second brightness value, a second contrast value, a second color distribution value, a second noise level value, a second local spectral dispersion value, and a second sharpening intensity value; Calculate the Euclidean distance between the first brightness value, the first contrast value, the first color distribution value, the first noise level value, the first local spectral dispersion value and the first sharpening intensity value, and the second brightness value, the second contrast value, the second color distribution value, the second noise level value, the second local spectral dispersion value and the second sharpening intensity value to obtain attribute similarity.

4. The image correction optimization method according to claim 1, wherein: The generating of the first correction coefficient according to the correction rule of the historical correction values of the automatic white balance in all the secondary processing records changing over time includes: Arrange the historical correction values of the automatic white balance in all secondary processing records in the order of timestamps to obtain a secondary correction value sequence; For the secondary correction value sequence, statistical fitting is used to draw a correction curve of the automatic white balance correction value changing over time to obtain a correction rule; The average slope of the calibration curve is calculated and used as a first calibration coefficient.

5. The image correction optimization method according to claim 4, wherein: The determining the predicted correction value of the current text image according to the correction rule includes: Determining a time point for collecting the current text image; The correction rule is retrieved, and the acquisition time point is substituted into the correction rule to obtain a predicted correction value corresponding to the current text image.

6. The image correction optimization method according to claim 1, wherein: The method of using statistical fitting to draw a correction curve of the automatic white balance correction value changing over time for the secondary correction value sequence to obtain a correction rule includes: Drawing a scatter plot of automatic white balance correction values changing over time according to the secondary correction value sequence; Calculating the autocorrelation of the scatter plot, and determining data characteristics based on the autocorrelation of the scatter plot; selecting a corresponding statistical fitting model for the automatic white balance correction value according to the data characteristics; Solving the parameters of the statistical fitting model to obtain model parameters; A calibration curve is drawn according to the statistical fitting model and the model parameters to obtain a calibration rule.

7. The image correction optimization method according to claim 1, wherein: Correcting the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain the second distortion parameter includes: Performing a weighted sum calculation on the first correction coefficient and the second correction coefficient to obtain a weighted value; Adding a preset adjustment factor to the weighted value to obtain a comprehensive correction factor; The first distortion parameter is multiplied by a comprehensive correction factor to obtain a second distortion parameter.

8. Image correction and optimization system, characterized in that: include: a parameter attribute determination module, configured to determine a first distortion parameter and a first image attribute of a current text image, wherein the current text image originates from an image acquisition device; an acquisition and screening module, configured to acquire historical image processing records of an image analysis device, and screen out secondary processing records from the historical image processing records according to the first image attribute; A first coefficient generating module, configured to generate a first correction coefficient according to a correction rule of the historical correction values of the automatic white balance in all secondary processing records changing over time; a second coefficient generating module, configured to determine an actual correction value of the automatic white balance of the current text image, determine a predicted correction value of the current text image according to the correction rule, and generate a second correction coefficient according to a deviation between the actual correction value and the predicted correction value; The parameter correction module is used to correct the first distortion parameter according to the first correction coefficient and the second correction coefficient to obtain the second distortion parameter.

9. 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, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images