Color consistency adjusting method and device for aviation linear array image and medium

Abstract

The invention provides a color consistency adjusting method and device for an aviation linear array image and a medium, and the method comprises an in-line correction step which is used for processing each scanning line of the aviation linear array image, and the in-line correction step comprises the steps: constructing a physical correction model based on an imaging angle, the physical correction model is applied to correct pixels in a scanning line; and the inter-line correction step is used for processing the air strip image subjected to the intra-line correction, and the inter-line correction step comprises the steps of constructing a gray scale trend model based on a time sequence and correcting pixels in the air strip by applying the gray scale trend model. According to the invention, the problem of non-ideal color correction effect caused by lack of explicit expression of an imaging physical process can be effectively avoided.

Description

Technical Field

[0001] This invention belongs to the field of remote sensing image processing and relates to a prior art technique for color consistency adjustment of aerial linear array images. Background Technology

[0002] Linear scan cameras, as important imaging devices in aerial and aerospace remote sensing, typically employ a pushbroom imaging method, which involves the sensor scanning line by line along the flight direction to acquire continuous images of ground features. This method offers advantages such as high resolution, high imaging speed, and large coverage, and is widely used in the production of large-area orthophotos. However, when acquiring remote sensing data, line scan cameras suffer from significant brightness unevenness both within and between rows due to differences in the reflection angles of ground features and the varying angles at which light is received by different pixels on the sensor. Specifically, this manifests as follows: Intra-line unevenness: Within the same scan line, the observed zenith angle of pixels varies continuously from the center to the edge, resulting in different incident angles of light on the image plane. Combined with the vignetting effect of the optical lens, this leads to a gradual grayscale transition that is "bright in the middle and dark at both ends." Inter-line unevenness: During flight, changes in solar altitude angle, atmospheric conditions, differences in ground reflection, and inconsistencies in sensor response time can cause color differences between different scanning time periods. Most existing homogenization methods do not fully consider the regular influence of imaging geometry on grayscale distribution. Their correction models lack explicit expressions of the imaging physical processes, resulting in inherent limitations in the consistency of the corrected images.

[0003] To address the need for image color consistency adjustment, the following main methods currently exist: First, histogram-based correction methods adjust the image grayscale distribution through histogram matching or equalization operations. However, these methods are prone to introducing color distortion when there are significant differences between images and are difficult to maintain detail integrity. Second, linear transformation methods based on statistical characteristics, such as Wallis filtering, achieve color uniformity by adjusting the image mean and standard deviation. However, this method is prone to speckle noise in uniform areas and has limited adaptability to overall brightness differences. Third, adjustment methods based on global optimization rely on a large number of feature points and use the least squares principle to achieve color consistency among multiple images. Although this method can achieve good overall results, it has high computational complexity and high requirements for data distribution and quality, making it difficult to apply to efficient processing scenarios for aerial linear array images.

[0004] The core idea of ​​the above methods lies in using mathematical means to statistically force adjustments to the grayscale distribution of images, without deeply considering the physical mechanisms of imaging. Under complex imaging conditions, this makes it difficult to fundamentally and systematically solve the problem of uneven grayscale distribution. Summary of the Invention

[0005] The purpose of this invention is to provide a method for color consistency adjustment of aerial linear array images. By establishing a radiometric correction model in two dimensions, namely, within and between scan lines, the method achieves efficient and accurate color consistency processing of the images.

[0006] This invention provides a method for adjusting the color consistency of aerial linear array images, comprising: An in-row correction step is used to process each scan line of an airborne linear array image, wherein the in-row correction step includes constructing a physical correction model based on the imaging angle and applying the physical correction model to correct the pixels within the scan line; The interline correction step is used to process the flight strip image after inline correction, wherein the interline correction step includes constructing a grayscale trend model based on time series and applying the grayscale trend model to correct the pixels within the flight strip.

[0007] Furthermore, the inline correction step includes: Each scan line is divided into multiple statistical windows and the average gray value of each window is calculated; Construct a physical correction model related to the imaging angle, the model including a cosine term and a linear term of the imaging angle; The model parameters are solved using the average grayscale value of each window and the corresponding imaging angle data; The grayscale of each pixel within the scan line is corrected based on the model parameters. Moreover, the imaging angle is calculated based on the focal length of the line scan camera, the pixel size, and the position of the pixel within the row.

[0008] Moreover, the parameters of the physical correction model are solved using the least squares method.

[0009] Furthermore, the interline correction step includes: The flight strip images are connected and arranged in chronological order of imaging time and divided into multiple windows. The average gray level of each window and the global average gray level of the flight strip are calculated. Establish and solve a trend model of the change of flight strip grayscale with time window sequence; The correction factor for each time window is calculated based on the global average gray level of the flight strip and the trend model. The correction factor is used to correct the pixel grayscale within the corresponding time window.

[0010] Moreover, the grayscale trend model is a high-order polynomial model with respect to the time window sequence number.

[0011] Furthermore, the correction factor is defined as the ratio of the global average gray level of the flight strip to the corresponding window gray level trend value calculated by the gray level trend model.

[0012] On the other hand, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a color consistency adjustment method for aerial linear array images as described above.

[0013] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a color consistency adjustment method for aerial linear array images as described above.

[0014] On the other hand, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements a color consistency adjustment method for aerial linear array images as described above.

[0015] This invention provides a color consistency adjustment technique for aerial linear array images. The core of this technique lies in the color correction models related to imaging angle within scanning rows and to row sequence between scanning rows. Compared to existing technologies, the difference in this invention is: Unlike traditional methods that use mathematical techniques to statistically force adjustments to image grayscale distribution, this method starts with the physical process of imaging in aerial line-scan cameras. Taking into account the push-broom characteristics of line-scan cameras, it decouples the grayscale inconsistency problem into two orthogonal dimensions: in-row and inter-row, for modeling and correction. A physical correction model related to the imaging angle is constructed for in-row models, while an inter-row correction model includes logarithmic and linear terms related to row numbers. The model parameters are solved through image grayscale sampling statistics, thereby achieving color consistency adjustment of the image. This approach effectively avoids the problem of unsatisfactory color correction results caused by the lack of explicit expression of the imaging physical process. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method according to an embodiment of the present invention.

[0017] Figure 2 This is a row grayscale histogram of an embodiment of the present invention, showing a "bright center and dark sides".

[0018] Figure 3 This is a schematic diagram of the inline window division in an embodiment of the present invention.

[0019] Figure 4 This is a schematic diagram of the middle pixel in an embodiment of the present invention.

[0020] Figure 5 This is a schematic diagram of the time series flight strip arrangement and grayscale changes of the flight strip in an embodiment of the present invention.

[0021] Figure 6 This is a schematic diagram of the flight strip window division according to an embodiment of the present invention. Detailed Implementation

[0022] The specific embodiments of the present invention will be described in detail below with reference to the examples and accompanying drawings.

[0023] This invention abandons the purely mathematical and statistical approach to adjustment, and instead starts from the imaging physics process. Specifically targeting the pushbroom characteristics of airborne linear array cameras, it decouples the grayscale inconsistency problem into two orthogonal dimensions, "in-row" and "between-row," for modeling and correction. In-row correction: Based on the scanning line imaging mechanism of airborne linear array cameras, a physical correction model related to the imaging angle is constructed, which includes the following forms: The term and the linear term are used to compensate for the differences in intra-row radiative response caused by observation geometry.

[0024] Interline correction: Based on the strip imaging process of an airborne linear array camera, a correction model related to the imaging process (scanning row number) is proposed. Its form includes logarithmic and linear terms of row number, which are used to correct strip noise and color deviation caused by inconsistency in scanning interline imaging.

[0025] This invention systematically eliminates the influence of imaging geometry and radiation conditions on image color by combining the above two-dimensional models, thereby achieving high-precision and high-efficiency color consistency processing of aerial linear array images.

[0026] Example 1 This embodiment provides a method for adjusting the color consistency of aerial linear array images, including: An in-row correction step is used to process each scan line of an airborne linear array image, wherein the in-row correction step includes constructing a physical correction model based on the imaging angle and applying the physical correction model to correct the pixels within the scan line; The interline correction step is used to process the flight strip image after inline correction, wherein the interline correction step includes constructing a grayscale trend model based on time series and applying the grayscale trend model to correct the pixels within the flight strip.

[0027] Example 2 like Figure 1 As shown, the color consistency adjustment method for aerial linear array images provided in this embodiment includes the following steps: Step 1: Divide the linear array image into segments and perform grayscale statistics on each segment.

[0028] In linear pushbroom imaging, the sensor generates tens of thousands of pixels per scan line. Influenced by factors such as the reflection angle of ground objects, vignetting of the optical lens, and inconsistent CCD pixel response, each line of image often exhibits a gradual change in grayscale distribution, with the center brighter and the ends darker. Figure 2The grayscale histogram of the rows is shown. To accurately quantify this radiative response, segmented statistical modeling of each row of pixels is required. The total number of segments... The selection of segments needs to strike a balance between model fitting accuracy and computational efficiency: too few segments will fail to accurately capture grayscale change trends, leading to correction residue; too many segments, while improving the finesse of trend description, will significantly increase computational load. This invention further proposes performing only odd-numbered segments, because when calculating the center index of the calculation window, for odd-numbered segments… , No. The center column index of each window can be directly calculated as an integer; while for even numbers... The center position is between two pixels, which requires additional processing.

[0029] To determine the optimal number of segments In this embodiment, an experiment was conducted using 1000 rows of actual linear array image data. For each row, the following operations were performed sequentially: according to different... The values ​​were divided into windows, the standard deviation of the grayscale within each row after correction was calculated, and finally the average of the standard deviations of all 1000 rows was taken. The results are shown in Table 1.

[0030] Table 1. Comparison of in-line correction performance and computational efficiency for different number of segments (N)

[0031] Using a standard deviation of 50 as the threshold for effective color uniformity, the results show that the number of segments... Effective inline correction can be achieved in the odd number range of 11 to 27, where when When the value is 21, the model achieves optimal overall performance in terms of fitting accuracy, computational efficiency, and anti-interference stability. The subsequent examples will use 21 windows for segmentation, as follows: Figure 3 The inline window division diagram is shown below.

[0032] For the Okay, number k Each window has an average grayscale value. The calculation formula is:

[0033] formula middle, Indicates the first line, number The original grayscale value of the column pixel. Indicates the first A collection of all pixel column indices contained in a window. Indicates the first The total number of pixels contained within a window. Average grayscale value. This data will be used as parameters for subsequent model solving.

[0034] Step 2: Establish an inline grayscale correction model and solve for the parameters.

[0035] This invention proposes that the grayscale value observed and imaged by the sensor... It is a geophysical concept grayscale Subject to imaging angle The resulting impact. To improve correction accuracy, this invention provides a physical correction model, establishing a function related to the imaging angle, which includes a cosine term and a first-order term of the imaging angle. Its form is:

[0036] The above model introduces This is mainly based on the geometric principle of Lambertian reflection: as the imaging angle increases, the projected area of ​​the ground object element in the sensor's line of sight increases according to... The decrease in regularity leads to a reduction in the received radiant energy. Introducing a first-order term... This can provide the model with additional flexibility, enabling it to adaptively fit and compensate for linear deviations not fully described by the cosine term in a data-driven manner, thereby achieving higher overall correction accuracy than a pure cosine model. Coefficients The value of is determined by the least squares method, and its positive or negative sign represents the direction of the deviation.

[0037] formula In the middle, coefficient For the model parameters to be determined, For the first The imaging angle of a pixel, which is the angle between its observation direction and the camera's optical axis, ranges from 0° to 90°. This represents the corresponding grayscale value.

[0038] Among them, imaging angle The calculation formula is as follows, precisely determined by the internal geometric relationships of the line scan camera:

[0039] In the formula, For the camera's focal length, The pixel size of the detector. To scan the column number of the center cell of the row, such as Figure 4 As shown in the figure. According to the above formula, the imaging angle can be calculated based on the focal length of the line scan camera, the pixel size, and the position of the pixel within the row.

[0040] Column number The calculation formula is:

[0041] in, The total number of pixels in a line of image, when If the result is not an integer, round down to the nearest integer.

[0042] Subsequently, the model parameters are solved using the average grayscale value of each window and the corresponding imaging angle data. This invention further proposes that the parameters of the physical correction model be solved using the least squares method. The average grayscale value of each window is substituted into... Substitute the average grayscale value of the window into The imaging angle of the center pixel of the window Substitution Multiple equations were constructed, and the model parameters were solved using least squares. .

[0043] The example lists 21 equations based on the 21 divided windows. Based on the statistical results of step 1, it is assumed that the reflectivity of objects within the same row is uniform, i.e., the ideal grayscale value of all pixels. Same, denoted as The core idea of ​​this hypothesis is to attribute the variation in grayscale across the entire row primarily to the influence of the imaging angle, rather than variations in the ground features themselves. Therefore, averaging the grayscale across the entire row can minimize the random fluctuations caused by the imaging angle, thus obtaining the best estimate of the ideal grayscale value for that row without angular influence. Based on this, the ideal grayscale... The value is set to be the average grayscale value of all windows in this row, and its calculation formula is as follows:

[0044] Here, the average grayscale value of the window is used. To represent the overall radiometric response of the ground features, and using the imaging angle at the center of the window. To approximate the angular conditions of this segment, observation equations are established for each window, resulting in a system of equations:

[0045] Among them, the imaging angle of the middle pixel of the window The calculation is as follows: Calculate the starting column index of the window (Inline starting index starts from 1):

[0046] Calculate the index of the end column of the window :

[0047] Calculate the middle column index of the window :

[0048] Will Substitute into the image angle calculation formula You can get .

[0049] The parameter solution process is as follows: model Rewrite in linear form:

[0050] By combining the 21 equations, a least-squares matrix equation is constructed. :

[0051] Solving the above overdetermined system of equations using the least squares method yields the optimal model parameter vector for this row of images. This set of parameters accurately describes the variation of grayscale response within the scanning line with the imaging angle.

[0052] Step 3: Perform grayscale correction on each pixel within the row.

[0053] After obtaining the parameters and ideal grayscale value After that, for the first in the industry For any pixel in the column, its original observed grayscale value is (Right now The corresponding imaging angle is The corrected grayscale value of this pixel. Calculated using the following formula:

[0054] This correction formula is essentially the inverse process of the model established in step 2 (formula (2)), aiming to remove the values ​​affected by imaging angle modulation (i.e., The distortion (represented by the image) affects the relative radiance of ground features under ideal conditions. By applying this calculation to each pixel within a row, a new linear array image with uniform grayscale within the row and effectively eliminating the "bright in the middle and dark at the edges" radiative distortion can be generated.

[0055] Step 4: Flight strip window division and grayscale statistics.

[0056] To analyze and correct the macroscopic fluctuations in image grayscale over flight time series, individual images need to be organized into continuous flight strips in chronological order and segmented for statistical analysis. Figure 5 The physical process of airborne linear array pushbroom imaging is shown in the image. A detailed explanation follows: The actual aerial imaging method involves flight paths that are connected end-to-end, with these flight paths arranged in parallel and overlapping in space to collectively cover the entire area. During flight, different flight paths are set according to the flight direction of the sensor-mounted platform (flight path 1, flight path 2, flight path 3... flight path n, where n represents the total number of flight paths obtained during flight); however, the flight path image arrangement method in this invention is different from... Figure 5 The physical spatial arrangement in the above figure is different. It arranges all the image data of the flight strips (flight strip 1, flight strip 2, flight strip 3... flight strip n) in the order of their imaging time, connecting them end to end to form an ultra-long, continuous image sequence.

[0057] To analyze and correct macroscopic fluctuations in image grayscale over flight time series, the aforementioned image sequence needs to be divided into multiple time windows. The core principle of window division is to ensure that each window contains a sufficient number of scan rows, thereby ensuring that the average grayscale value calculated based on a large number of pixels within the window has statistical robustness and can reliably reflect the macroscopic radiance level within that time period, avoiding excessive interference from local ground anomalies or noise. Window length ( The method for determining ) is as follows:

[0058] in, Line scan camera's line frequency (unit: lines / second), which is the number of image lines acquired per second; The preset number of scan lines expected to be included in each window. The value can usually be set to a large fixed number of rows based on experience (e.g., corresponding to 1 to 10 minutes of flight data, with 5 minutes of flight data being the preferred option), or dynamically determined based on the total number of flight strip rows (e.g., 1% to 5% of the total number of rows).

[0059] The embodiment divides the entire flight strip imagery into preset flight time intervals. Calculate the average gray level of each time window and the global average gray level of the flight strip: Based on a determined window length Total flight time The flight strip (5 minutes in the example) is divided into... A time window, such as Figure 6 As shown, the calculation formula is:

[0060] in To round up to ensure all data is covered.

[0061] Subsequently, statistics were performed on the window and global grayscale values. For the first... For each window, calculate its average gray level. The formula is as follows:

[0062] in, Indicates the first The set of all row numbers contained in a time window. For the first The total number of image rows contained within a window. For the first line, number The grayscale value after column pixel correction. This represents the total number of pixels in each row.

[0063] At the same time, calculate the global average gray level of the entire flight strip. This serves as the benchmark for the overall grayscale level of the flight strip. The formula is as follows:

[0064] in, This represents the total number of time windows divided within the flight strip.

[0065] Step 5: Modeling and solving the grayscale trend of the flight strip.

[0066] To accurately characterize and fit the complex nonlinear fluctuations of the average gray level of the window within the flight strip over time, this step establishes and solves a high-order polynomial trend model.

[0067] Observations show that the average gray level of the window within the flight strip is... With window serial number It exhibits complex nonlinear fluctuations. To accurately fit the macroscopic variation trend of the flight strip grayscale, this embodiment uses a window sequence number. The 9th degree polynomial is used as a trend model:

[0068] in, Let be the model coefficients to be determined. Window serial number of This model will be used in subsequent steps to calculate the radiation correction factor at the flight strip scale.

[0069] Next, the coefficients are solved based on the least squares principle.

[0070] Obtained by dividing the flight strip Data points Construct the following matrix equation:

[0071] in, , … Window serial number , to The second power term.

[0072] The equation can be simplified as follows: :

[0073]

[0074]

[0075] in, To observe the grayscale vector, For polynomial basis matrices, This is the coefficient vector.

[0076] According to the least squares principle, the optimal coefficient vector minimizes the sum of squared residuals between the predicted and observed values. The solution is given by the following normal equation:

[0077] Solving this equation yields the optimal coefficients of the polynomial model. .

[0078] Thus far, the grayscale trend model of the flight strip has been established. It has been fully identified and can be used for subsequent radiation correction.

[0079] Step 6: Construct the flight strip correction equation and pixel-level correction.

[0080] Step 5 determines the grayscale trend model of the flight strip. Next, this step aims to establish an inter-row pixel grayscale correction model and perform a sequential traversal and correction of each pixel in the aerial image.

[0081] Interline correction is achieved through a window sequence number. The relevant multiplicative factors are realized. Window correction factor The value is defined as the ratio of the global average gray level of the flight strip to the gray level trend value of that window.

[0082] in, The global average grayscale value of the flight strip calculated in step 4 is used as the target benchmark for correcting the grayscale consistency of the entire flight strip. The first one predicted by the polynomial model in step 5 The grayscale trend value of each window.

[0083] The correction factor automatically determines the correction direction for each window: when When the average gray level of the window is higher than the global baseline, it will be attenuated during correction. when When the average gray level of the window is lower than the global baseline, it indicates that the window's average gray level will be enhanced during correction.

[0084] For those located at the The final grayscale value of any pixel within a window after interline correction. Calculated by the following formula:

[0085] By applying the above formula to each pixel in the entire aerial image, a corrected image with good color consistency both within and between rows can be obtained.

[0086] To visually verify the effectiveness of the proposed intra-line and inter-line collaborative correction model, a comparative experiment was designed and conducted. The parameters for each process in the experiment were: total pixels of one line of image. center pixel Camera focal length Pixel size =5 Based on this, the maximum imaging angle at the edge is calculated. To further simulate radiation changes during real flight, the experiment constructed a continuous flight path image of 20 minutes in length and divided it into 10 time windows at 2-minute intervals.

[0087] The experiment quantitatively compares the proposed method with two mainstream homogenization methods: Wallis filtering and histogram matching. The method strictly follows the specific implementation steps, dividing a single-line image into 21 statistical windows to solve the physical correction model and performing pixel-by-pixel correction. Subsequently, the correction factor for each time window is calculated at the flight strip scale to eliminate inter-line differences. Finally, the standard deviation of gray levels within each line, the average standard deviation of gray levels between lines, and the overall standard deviation of the entire image are used as evaluation indicators. The results are shown in Table 2.

[0088] Table 2 Comparison of image grayscale standard deviations after processing by different methods

[0089] The experimental data comparison and analysis in Table 2 show that after processing by the method of this invention, the standard deviations of the inline, interline, and overall gray levels of the image are reduced to 41.63, 55.12, and 49.27, respectively, with a better improvement in uniformity than the two traditional methods. Therefore, the collaborative correction model proposed in this invention can solve the problem of inline and interline radiometric inhomogeneity in aerial linear array images, effectively ensuring color consistency while possessing both the accuracy of correction based on a physical model and good computational feasibility.

[0090] Example 3 This invention provides a color consistency adjustment system for aerial linear array images, comprising the following modules: The in-row correction module is used to process each scan line of the airborne linear array image, wherein the in-row correction step includes constructing a physical correction model based on the imaging angle and applying the physical correction model to correct the pixels within the scan line; The interline correction module is used to process the flight strip image after inline correction. The interline correction step includes constructing a grayscale trend model based on time series and applying the grayscale trend model to correct the pixels within the flight strip.

[0091] In specific implementation, the method proposed in the technical solution of this invention can be automatically executed by those skilled in the art using computer software technology. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of this invention and computer equipment including the computer program running the corresponding computer program, should also be within the protection scope of this invention.

[0092] The following embodiments describe the electronic device provided by the present invention. The electronic device described below can be referred to in correspondence with the color consistency adjustment method for aerial linear array images described above.

[0093] The electronic device may include a processor, a communications interface, memory, and a communication bus, wherein the processor, communications interface, and memory communicate with each other via the communication bus. The processor can call logical instructions in the memory to execute a color consistency adjustment method for aerial linear array images, mainly including the software processing part mentioned above.

[0094] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0095] On the other hand, embodiments of the present invention also provide a computer program product, the computer program product including a computer program, the computer program being stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute the software processing portion of the color consistency adjustment method for aerial linear array images provided by the above methods.

[0096] In another aspect, embodiments of the present invention also provide a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the software processing portion of the color consistency adjustment method for aerial linear array images provided by the above methods.

[0097] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0098] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for adjusting color consistency in aerial linear array images, characterized in that, include: An in-row correction step is used to process each scan line of an airborne linear array image, wherein the in-row correction step includes constructing a physical correction model based on the imaging angle and applying the physical correction model to correct the pixels within the scan line; The interline correction step is used to process the flight strip image after inline correction, wherein the interline correction step includes constructing a grayscale trend model based on time series and applying the grayscale trend model to correct the pixels within the flight strip.

2. The method for adjusting color consistency of aerial linear array images according to claim 1, characterized in that, The inline correction steps include: Each scan line is divided into multiple statistical windows and the average gray value of each window is calculated; Construct a physical correction model related to the imaging angle, the model including a cosine term and a linear term of the imaging angle; The model parameters are solved using the average grayscale value of each window and the corresponding imaging angle data; The grayscale of each pixel within the scan line is corrected based on the model parameters.

3. The method for adjusting color consistency of aerial linear array images according to claim 1, characterized in that: The imaging angle is calculated based on the focal length of the line scan camera, the pixel size, and the position of the pixel within the row.

4. The method for adjusting color consistency of aerial linear array images according to claim 1, characterized in that: The parameters of the physical correction model are solved using the least squares method.

5. The method for adjusting color consistency of aerial linear array images according to claim 1, characterized in that: The interline correction step includes: The flight strip images are connected and arranged in chronological order of imaging time and divided into multiple windows. The average gray level of each window and the global average gray level of the flight strip are calculated. Establish and solve a trend model of the change of flight strip grayscale with time window sequence; The correction factor for each time window is calculated based on the global average gray level of the flight strip and the trend model. The correction factor is used to correct the pixel grayscale within the corresponding time window.

6. The method for adjusting color consistency of aerial linear array images according to claim 1, characterized in that: The gray-scale trend model is a high-order polynomial model about the time window sequence number.

7. The method for adjusting color consistency of aerial linear array images according to claim 1, characterized in that: The correction factor is defined as the ratio of the global average gray level of the flight strip to the corresponding window gray level trend value calculated by the gray level trend model.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, it implements a color consistency adjustment method for aerial linear array images as described in any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements a color consistency adjustment method for aerial linear array images as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that: When the computer program is executed by the processor, it implements a color consistency adjustment method for aerial linear array images as described in any one of claims 1 to 7.

Images