A method and system for constructing a three-dimensional model of a Ferris wheel

Abstract

The present application relates to the field of image processing technology, and specifically to a method and system for constructing a three-dimensional model of a Ferris wheel, the method comprising: obtaining each rotating image of the Ferris wheel at each acquisition time in each acquisition cycle; obtaining the number of a type of pixel points in each area of ​​each rotating image of the Ferris wheel; obtaining the regional clarity of each area in each rotating image of the Ferris wheel; obtaining the detail significance coefficient of each pixel point in each area of ​​each rotating image of the Ferris wheel; obtaining the texture complex pixel points in each area of ​​each rotating image of the Ferris wheel; dividing the front and rear adjacent areas; obtaining the detail texture complexity of each area in each rotating image of the Ferris wheel; obtaining the defogging weight of each area in each rotating image of the Ferris wheel; obtaining the updated contrast of each area in each rotating image of the Ferris wheel; enhancing each rotating image of the Ferris wheel according to the updated contrast to construct a three-dimensional model of the Ferris wheel. The present application improves the accuracy of constructing a three-dimensional model of the Ferris wheel.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a method and system for constructing a three-dimensional model of a Ferris wheel. Background Art

[0002] Three-dimensional models play a vital role in modern design and engineering, especially in the structural and exterior design of Ferris wheels. Designers and engineers can use three-dimensional modeling technology to intuitively visualize the various parts of the Ferris wheel, making it easier to make design modifications and adjustments. However, foggy weather conditions will have a certain impact on the construction process of three-dimensional models. When using drones for aerial photography in foggy conditions, the image of the rotating Ferris wheel may become blurred due to the scattering of light by particles in the fog. In this case, the edge of the Ferris wheel may not be clear enough and detailed information may be missing, resulting in the inability to accurately restore certain key parts of the three-dimensional model. These blurred images will affect the designer's judgment and modification of the appearance of the Ferris wheel, making modeling in foggy conditions more challenging.

[0003] In a foggy environment, when using a drone for aerial photography, the image of the Ferris wheel is not clear due to the scattering of light by particles in the fog, the edge of the Ferris wheel is blurred, and detail information is missing. The existing defogging method based on image enhancement can improve the image quality to a certain extent, but due to factors such as information loss, noise interference and scene complexity, it is impossible to completely restore the clarity and details of the Ferris wheel image taken in foggy weather. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a method and system for constructing a three-dimensional model of a Ferris wheel. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for constructing a three-dimensional model of a Ferris wheel, the method comprising the following steps:

[0006] Acquire each Ferris wheel rotation image at each acquisition time in each acquisition cycle; divide each Ferris wheel rotation image into each area;

[0007] Based on the average grayscale value of the pixels in the area, the number of pixels of a type in each area of ​​each Ferris wheel rotation image is obtained;

[0008] Based on the number of pixels of a type in the region, the average grayscale value of the pixels, and the average discrete degree, the regional clarity of each region in each Ferris wheel rotation image is obtained;

[0009] Based on the consistency of the grayscale values ​​of the pixels in the region and the average of the gradient amplitudes, the detail significance coefficient of each pixel in each region of each Ferris wheel rotation image is obtained;

[0010] Acquire the texture complex pixel points of each area in each Ferris wheel rotation image based on the detail saliency coefficient;

[0011] Dividing each region in each Ferris wheel rotation image into front and rear adjacent regions;

[0012] Based on the average of the detail significance coefficient in the region and the number of pixels with complex texture, the detail texture complexity of each region in each Ferris wheel rotation image is obtained;

[0013] Obtain the defogging weight of each region in each Ferris wheel rotation image based on the regional clarity and detail texture complexity;

[0014] Based on the defogging weight and the contrast of the region, the updated contrast of each area in each Ferris wheel rotating image is obtained; each Ferris wheel rotating image is enhanced according to the updated contrast, and a binocular vision algorithm is used in combination with the enhanced Ferris wheel rotating image to construct a Ferris wheel three-dimensional model.

[0015] Furthermore, the method for obtaining the number of pixels of the first type is:

[0016] For each area in each Ferris wheel rotation image, the average grayscale value of all pixels in the area is calculated as the average grayscale mean of each area, and the number of pixels in each area whose grayscale value is greater than and equal to the average grayscale mean is counted as the number of pixels of a type in each area.

[0017] Furthermore, the method for obtaining the regional clarity is:

[0018] For each area in each Ferris wheel rotating image, calculate the ratio of the mean grayscale value of all pixels in the eight-neighborhood area of ​​each pixel in the area to the mean grayscale value of all pixels in the area, calculate the sum of all the ratios in the area, calculate the variance of the grayscale values ​​of all pixels in the area, calculate the product of the summation result, the variance and the number of pixels of a type, and use the sum of all the products in the area as the regional clarity of each area in each Ferris wheel rotating image.

[0019] Further, the method for obtaining the detail significant coefficient is:

[0020] Calculate the gradient amplitude of the grayscale value of each pixel in each Ferris wheel rotation image;

[0021] The calculation formula of the detail significance coefficient is: ; In the formula, Indicates the first In the region The detail significance coefficient of each pixel; Indicates the first within the region and The number of pixels with the same grayscale value; Indicates the first In the region The gradient amplitude of each pixel; Indicates the first The mean value of the gradient amplitude of pixels in the region; Indicates the first The number of pixels in the region; The function is a maximum value function; The function is a minimum value function; represents the absolute value function.

[0022] Furthermore, the method for obtaining the pixel points with complex texture is as follows:

[0023] For each region in each Ferris wheel rotation image, the mean of detail significance coefficients of all pixels in the region is calculated as a first mean, and the pixels in the region whose detail significance coefficients are greater than or equal to the first mean are taken as the texture complex pixels in each region.

[0024] Furthermore, the division of the front and rear adjacent areas of each area in each Ferris wheel rotation image includes:

[0025] For each area in each Ferris wheel rotating image at each acquisition time, if a Ferris wheel cabin appears in the area, obtain the area where the cabin is located in the Ferris wheel rotating images at the previous acquisition time and the next acquisition time of each acquisition time, as the front and back adjacent areas of the cabin area in the Ferris wheel rotating image at each acquisition time;

[0026] For the area where the Ferris wheel cabin does not appear in each Ferris wheel rotating image at each acquisition time, the area with the same position as the area in the Ferris wheel rotating images at the previous adjacent moment and the next adjacent moment of the acquisition time is used as the front and back adjacent areas of the area; the cabin is a manually marked area.

[0027] Furthermore, the calculation formula of the detail texture complexity is: ; In the formula, Indicates the first The detail texture complexity of each area; Indicates the first In the region The detail saliency coefficient of a pixel with complex texture; Indicates the rotation image of each Ferris wheel The number of pixels with complex textures in a region; Indicates the first The mean of detail saliency coefficients of all pixels with complex textures in a region and its adjacent regions before and after it; Indicates the first The mean value of the detail saliency coefficient of the pixels with complex textures in the adjacent regions before and after the region.

[0028] Furthermore, the defogging weight is a normalized value of the product of regional clarity and detail texture complexity.

[0029] Further, the method for obtaining the updated contrast is:

[0030] Obtaining the contrast of each region in each Ferris wheel rotation image;

[0031] The calculation formula for the updated contrast is: ; In the formula, Indicates the first Update contrast of each region; Indicates the first The dehazing weight of each region; Indicates the first The contrast of the area.

[0032] In a second aspect, an embodiment of the present application further provides a Ferris wheel three-dimensional model construction system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0033] This application has at least the following beneficial effects:

[0034] This application divides each Ferris wheel rotating image into regions, calculates the clarity and detail texture complexity of different regions, and then assigns different defogging weights to different regions. Finally, the processed image is enhanced to complete the construction of the Ferris wheel three-dimensional model, thereby improving the clarity and details of the Ferris wheel rotating image and the accuracy of the Ferris wheel three-dimensional model construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1A flowchart of a method for constructing a three-dimensional model of a Ferris wheel provided in one embodiment of the present application;

[0037] Figure 2 A flowchart for obtaining defogging weights is provided for one embodiment of the present application. DETAILED DESCRIPTION

[0038] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of a method and system for building a three-dimensional model of a Ferris wheel proposed in the present application, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0039] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0040] The specific scheme of a Ferris wheel three-dimensional model construction method and system provided by the present application is described in detail below with reference to the accompanying drawings.

[0041] See also Figure 1 , which shows a flowchart of a method for constructing a three-dimensional model of a Ferris wheel provided by an embodiment of the present application, the method comprising the following steps:

[0042] Step S1, acquiring the rotating images of each Ferris wheel at each acquisition time in each acquisition cycle.

[0043] Two drones are used to shoot the two rotating surfaces of the Ferris wheel at the same height as the center of the Ferris wheel's rotation, and the images obtained are denoised and grayed to obtain two Ferris wheel rotation images at each acquisition time. The interval between the acquisition times of the Ferris wheel rotation images is 1s, and the time for the Ferris wheel to rotate one circle is taken as an acquisition cycle, and each Ferris wheel rotation image at each acquisition time in each acquisition cycle is obtained. In this embodiment, Gaussian filtering is used for denoising, and the implementer can select other denoising methods according to actual conditions.

[0044] Step S2, divide each Ferris wheel rotating image into regions; based on the average grayscale value of the pixels in the region, obtain the number of a type of pixels in each region of each Ferris wheel rotating image; based on the number of a type of pixels in the region, the average grayscale value of the pixels and the average discrete degree, obtain the regional clarity of each region in each Ferris wheel rotating image; based on the consistency of the grayscale value of the pixels in the region and the average gradient amplitude, obtain the detail significance coefficient of each pixel in each region of each Ferris wheel rotating image.

[0045] Since there will be different thicknesses of fog in a foggy environment, different fog densities have different effects on the details of the image, so the image needs to be divided into regions according to different fog densities.

[0046] For the Ferris wheel rotation image at the acquisition time of each acquisition cycle, the Ferris wheel rotation image is evenly divided into two parts with a side length of Rectangles are used as the regions of the Ferris wheel rotation image.

[0047] In the process of building a 3D model using a binocular vision algorithm, the details of the Ferris wheel are obtained by taking images of it. However, the foggy environment has a significant impact on the image, causing the detailed features of the Ferris wheel to be blocked, thereby reducing the overall quality of the image. Directly defogging the image will cause the important features and details of the area with more details to be lost. Therefore, in order to effectively improve the image quality, it is necessary to perform targeted defogging according to the fog concentration in different areas to ensure that while removing the fog, the important features of the area with rich details are retained as much as possible, thereby improving the accuracy and completeness of the 3D model construction.

[0048] Due to the uneven distribution of fog, the density of fog in different areas of the image is different, so the degree of influence on the details of the Ferris wheel is different. For areas with thicker fog, it means that the fog concentration is higher and a greater degree of defogging is required; for areas with thinner fog, it means that the fog concentration is lower and a smaller degree of defogging is required. Therefore, the clarity of the area is analyzed according to the performance of different areas in the image.

[0049] When the fog density is high, the water droplets in the mist will scatter and absorb light, and the intensity of light reaching the camera will be weakened, causing the grayscale value in the Ferris wheel rotating image to generally decrease, which is manifested as a decrease in image brightness and contrast, resulting in loss of details and reduced clarity in the image. When the fog density is low, the grayscale value in the Ferris wheel rotating image is relatively high, the image brightness and contrast are better, the details are clearer, and the overall image quality is higher.

[0050] Furthermore, for each area in each Ferris wheel rotating image, the average grayscale value of all pixels in the area is calculated as the average grayscale mean of each area, and the number of pixels in each area whose grayscale value is greater than and equal to the average grayscale mean is counted as the number of pixels of a type in each area.

[0051] According to the above analysis, in order to reflect the clarity of each area in the Ferris wheel rotating image, based on the average gray value and average discreteness of the pixel points in the Ferris wheel rotating image, the regional clarity of each area in the Ferris wheel rotating image at each acquisition moment is obtained. The calculation formula is: ; In the formula, Indicates the first The regional clarity of the area; Indicates the first In the region The mean gray value of the pixel in the eight neighborhoods of the pixel; Indicates the first The mean gray value of pixels in the region; Indicates the first The number of pixels in the region; Indicates the first The variance of the grayscale values ​​of all pixels in the region; Indicates the first The number of pixels of one type in an area.

[0052] It should be noted that Indicates The ratio of the average grayscale of the neighborhood of a pixel to the average grayscale of the entire area. The larger the value, the brighter the neighborhood of the pixel is than the overall grayscale level of the area, which means that the clarity of the area is greater. The larger the value of The more pixels with larger gray values ​​there are in an area, the better the brightness is, that is, the clearer the area is; It reflects the degree of change of the gray value of the pixel in the i-th region. The larger the value, the smaller the gray value. The more obvious the difference in the grayscale of pixels in a region is, the richer the details of the image is. At this time, the larger the value of the obtained regional clarity is; conversely, the smaller the value of the obtained regional clarity is.

[0053] When shooting the Ferris wheel as it rotates, the passenger cabins hanging on the edge of the wheel will appear in different areas of the images as the Ferris wheel rotates. Due to the uneven thickness of the fog, the occlusion of the cabins in different areas is also different, so the detailed texture of the Ferris wheel in different areas is also inconsistent.

[0054] For each region in each Ferris wheel rotation image, the pixels belonging to the Ferris wheel region in different regions are more important than the other pixels, so it is necessary to determine the importance of each pixel in each region relative to other pixels.

[0055] Furthermore, when the fog concentration is higher, the edge of the cabin is blurred, the gradient of the corresponding complex texture pixels is smaller, the details in the area are fewer and the grayscale value of each pixel tends to be consistent; when the fog concentration is lower, the edge of the cabin is clear, the gradient of the corresponding complex texture pixels is larger, the details in the area are more and the grayscale of the pixels in the Ferris wheel area is larger than that of the pixels in the surrounding area, but the number is relatively small.

[0056] The Sobel operator is used to calculate the gradient amplitude of the grayscale value of each pixel in each Ferris wheel rotation image, and the correlation between the pixels is determined according to the gradient change between the pixels. The greater the correlation between the pixel and the surrounding pixels, and the fewer the number of adjacent pixels with the same grayscale value as the pixel, the more important the pixel is relative to other pixels in the cabin. The Sobel operator is a well-known technology and will not be described in detail in this embodiment.

[0057] According to the above analysis, in order to reflect whether the texture details in the Ferris wheel rotating image are obvious, the detail saliency coefficient of each pixel point in each area of ​​each Ferris wheel rotating image is calculated. The calculation formula is: ; In the formula, Indicates the first In the region The detail significance coefficient of each pixel; Indicates the first within the region and The number of pixels with the same grayscale value as the pixels; Indicates the first In the region The gradient amplitude of each pixel; Indicates the first The mean value of the gradient amplitude of pixels in the region; Indicates the first The number of pixels in the region; The function is a maximum value function; The function is a minimum value function; represents the absolute value function.

[0058] It should be noted that when The smaller the description in the region and The fewer the number of pixels with the same gray value, the The more likely the pixel is The details of each area. Indicates In the region The correlation between the pixel and the surrounding pixels. The larger the value, the In the region The more important a pixel is, the more likely it is to belong to the Ferris wheel area.

[0059] The above steps calculate the detail significance coefficient of each pixel in each area, that is, judge the importance of each pixel. The more pixels with complex texture, the more detailed features belonging to the Ferris wheel in the image. Then the fog concentration in this area is small, and a smaller defogging weight can be assigned.

[0060] Step S3, based on the detail significance coefficient, obtain the texture complex pixel points of each area in each Ferris wheel rotating image; divide the front and rear adjacent areas of each area in each Ferris wheel rotating image; based on the average of the detail significance coefficient in the area and the number of texture complex pixel points, obtain the detail texture complexity of each area in each Ferris wheel rotating image; based on the area clarity and detail texture complexity, obtain the defogging weight of each area in each Ferris wheel rotating image.

[0061] Furthermore, for each region in each Ferris wheel rotation image, the mean of detail significance coefficients of all pixels in the region is calculated as a first mean, and the pixels in the region whose detail significance coefficients are greater than or equal to the first mean are taken as the texture complex pixels in each region.

[0062] For each area in each Ferris wheel rotating image at each acquisition moment, if a Ferris wheel cabin appears in the area, the area where the cabin is located in the Ferris wheel rotating images at the previous acquisition moment and the next acquisition moment of each acquisition moment is obtained as the front and rear adjacent areas of the area of ​​the cabin in the Ferris wheel rotating image at each acquisition moment.

[0063] For an area where a Ferris wheel cabin does not appear in each Ferris wheel rotating image at each acquisition moment, an area in the Ferris wheel rotating image at the previous adjacent moment and the next adjacent moment of the acquisition moment with the same position as the area is taken as the front and rear adjacent areas of the area; the cabin is a manually marked area.

[0064] Specifically, the detail texture complexity of the region in the image is analyzed according to the relative size of the cockpit in the region in the image and the cockpit in the region in the adjacent image. The larger the cockpit in the region, the more corresponding texture complex pixels.

[0065] Further, according to the above analysis, in order to reflect the texture details of the cabin in the Ferris wheel rotation image, the detail texture complexity of each area in each Ferris wheel rotation image is calculated, and the calculation formula is: ; In the formula, Indicates the first The detail texture complexity of each area; Indicates the first In the region The detail saliency coefficient of a pixel with complex texture; Indicates the rotation image of each Ferris wheel The number of pixels with complex textures in an area; Indicates the first The mean of detail saliency coefficients of all pixels with complex textures in a region and its adjacent regions before and after it; Indicates the first The mean value of the detail saliency coefficient of the pixels with complex textures in the adjacent regions before and after the region.

[0066] It should be noted that What is calculated is the sum of detail saliency coefficients of all pixels with complex textures. The accumulation of the number of pixels with complex textures and their degree of expression directly reflects the detail complexity of the area. Reflects the The relative strength of the detail performance of the first area and the adjacent areas. The larger the value, the The more complex the detail texture of the area, the more complex the detail texture of the area. By multiplying the sum of the detail saliency coefficients by the ratio of the relative detail performance, a comprehensive detail texture complexity can be obtained. The detail complexity not only depends on the detail performance of the complex texture pixels inside the area, but also considers its comparison with the adjacent areas, thus obtaining a more comprehensive evaluation.

[0067] The detailed information of different areas in the rotating image of the Ferris wheel contains the image. Simply applying a unified defogging process to the entire image may cause the loss of important details of the Ferris wheel, thereby affecting the overall quality of the image and the construction of the Ferris wheel 3D model. While defogging, it is necessary to retain the detailed features of the Ferris wheel. A moderate defogging operation should be adopted, and different defogging weights should be assigned to each area according to the clarity of each area and the complexity of the detail texture. The greater the clarity and the complexity of the detail texture, the smaller the defogging weight should be assigned to the area.

[0068] Furthermore, in order to reflect the degree of defogging of the Ferris wheel rotating image, the defogging weight of each area in each Ferris wheel rotating image is calculated, and the calculation formula is: ; In the formula, Indicates the first The dehazing weight of each region; Indicates the first The regional clarity of the area; Indicates the first The detail texture complexity of each area; Represents the normalization function. The defogging weight acquisition flow chart is as follows: Figure 2 shown.

[0069] It should be noted that Description The higher the clarity of the area, the less it is affected by fog and the less fog is needed. The more detailed information of the Ferris wheel contained in an area, the lower the defogging degree and the defogging weight need to be, so as to avoid losing the details during defogging.

[0070] Step S4, obtaining an updated contrast of each area in each Ferris wheel rotating image based on the defogging weight and the contrast of the area; enhancing each Ferris wheel rotating image according to the updated contrast, and constructing a Ferris wheel three-dimensional model using a binocular vision algorithm combined with the enhanced Ferris wheel rotating image.

[0071] Apply the defogging weights that should be assigned to each region obtained above to each region of the image, and use the defogging weights to adjust the contrast of each region in the image to complete the image enhancement of the Ferris wheel rotating image. Obtain the contrast of each region in each Ferris wheel rotating image, and calculate the updated contrast of each region in each Ferris wheel rotating image based on the defogging weights of each region in each Ferris wheel rotating image. The calculation formula is: ; In the formula, Indicates the first Update contrast of each region; Indicates the first The dehazing weight of each region; Indicates the first The contrast of the area.

[0072] It should be noted that the updated contrast is the improved contrast. When the defogging weight of each area in each Ferris wheel rotating image is larger and the regional clarity is smaller, it means that the area needs to be enhanced to a greater extent. At this time, the updated contrast of the area is larger; conversely, the updated contrast of the area is smaller, thereby obtaining the defogging Ferris wheel rotating image.

[0073] At this point, the defogged images of the rotating Ferris wheel taken by the two drones at each acquisition moment are obtained, and the images with quality problems are processed to improve the clarity and details of the rotating Ferris wheel images. Based on the defogged images of the rotating Ferris wheel taken by the two drones at each acquisition moment, the three-dimensional model of the Ferris wheel is constructed using the binocular vision algorithm to improve the accuracy of the construction of the three-dimensional model of the Ferris wheel.

[0074] Based on the same inventive concept as the above method, an embodiment of the present application also provides a Ferris wheel three-dimensional model construction system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned Ferris wheel three-dimensional model construction methods are implemented.

[0075] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0076] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0077] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for constructing a three-dimensional model of a Ferris wheel, characterized in that: The method comprises the following steps: Acquire each Ferris wheel rotation image at each acquisition time in each acquisition cycle; divide each Ferris wheel rotation image into each area; Based on the average grayscale value of the pixels in the area, the number of pixels of a type in each area of ​​each Ferris wheel rotation image is obtained; Based on the number of pixels of a type in the region, the average grayscale value of the pixels, and the average discrete degree, the regional clarity of each region in each Ferris wheel rotation image is obtained; Based on the consistency of the grayscale values ​​of the pixels in the region and the average of the gradient amplitudes, the detail significance coefficient of each pixel in each region of each Ferris wheel rotation image is obtained; Acquire the texture complex pixel points of each area in each Ferris wheel rotation image based on the detail saliency coefficient; Dividing each region in each Ferris wheel rotation image into front and rear adjacent regions; Based on the average of the detail significance coefficient in the region and the number of pixels with complex texture, the detail texture complexity of each region in each Ferris wheel rotation image is obtained; Obtain the defogging weight of each region in each Ferris wheel rotation image based on the regional clarity and detail texture complexity; Based on the defogging weight and the contrast of the region, an updated contrast of each region in each Ferris wheel rotating image is obtained; each Ferris wheel rotating image is enhanced according to the updated contrast, and a three-dimensional model of the Ferris wheel is constructed by combining a binocular vision algorithm with the enhanced Ferris wheel rotating image; The method for obtaining the detail significant coefficient is: Calculate the gradient amplitude of the grayscale value of each pixel in each Ferris wheel rotation image; The calculation formula of the detail significance coefficient is: ; In the formula, Indicates the first In the region The detail significance coefficient of each pixel; Indicates the first within the region and The number of pixels with the same grayscale value; Indicates the first In the region The gradient amplitude of each pixel; Indicates the first The mean value of the gradient amplitude of pixels in the region; Indicates the first The number of pixels in the region; The function is a maximum value function; The function is a minimum value function; represents the absolute value function.

2. A method for constructing a three-dimensional model of a Ferris wheel as claimed in claim 1, characterized in that: The method for obtaining the number of the first type of pixels is: For each area in each Ferris wheel rotation image, the average grayscale value of all pixels in the area is calculated as the average grayscale mean of each area, and the number of pixels in each area whose grayscale value is greater than and equal to the average grayscale mean is counted as the number of pixels of a type in each area.

3. A method for constructing a three-dimensional model of a Ferris wheel as claimed in claim 1, characterized in that: The method for obtaining the regional clarity is: For each area in each Ferris wheel rotating image, calculate the ratio of the mean grayscale value of all pixels in the eight-neighborhood area of ​​each pixel in the area to the mean grayscale value of all pixels in the area, calculate the sum of all the ratios in the area, calculate the variance of the grayscale values ​​of all pixels in the area, calculate the product of the summation result, the variance and the number of pixels of a type, and use the sum of all the products in the area as the regional clarity of each area in each Ferris wheel rotating image.

4. A method for constructing a three-dimensional model of a Ferris wheel as claimed in claim 1, characterized in that: The method for obtaining the pixel points with complex texture is as follows: For each region in each Ferris wheel rotation image, the mean of detail significance coefficients of all pixels in the region is calculated as a first mean, and the pixels in the region whose detail significance coefficients are greater than or equal to the first mean are taken as the texture complex pixels in each region.

5. The method for constructing a three-dimensional model of a Ferris wheel according to claim 1, characterized in that: The dividing of the front and rear adjacent areas of each area in each Ferris wheel rotation image includes: For each area in each Ferris wheel rotating image at each acquisition time, if a Ferris wheel cabin appears in the area, obtain the area where the cabin is located in the Ferris wheel rotating images at the previous acquisition time and the next acquisition time of each acquisition time, as the front and back adjacent areas of the cabin area in the Ferris wheel rotating image at each acquisition time; For the area where the Ferris wheel cabin does not appear in each Ferris wheel rotating image at each acquisition time, the area with the same position as the area in the Ferris wheel rotating images at the previous adjacent moment and the next adjacent moment of the acquisition time is used as the front and back adjacent areas of the area; the cabin is a manually marked area.

6. A method for constructing a three-dimensional model of a Ferris wheel as claimed in claim 1, characterized in that: The calculation formula of the detail texture complexity is: ; In the formula, Indicates the first The detail texture complexity of each area; Indicates the first In the region The detail saliency coefficient of a pixel with complex texture; Indicates the rotation image of each Ferris wheel The number of pixels with complex textures in a region; Indicates the first The mean of detail saliency coefficients of all pixels with complex textures in a region and its adjacent regions before and after it; Indicates the first The mean value of the detail saliency coefficient of the pixels with complex textures in the adjacent regions before and after the region.

7. A method for constructing a three-dimensional model of a Ferris wheel as claimed in claim 1, characterized in that: The defogging weight is a normalized value of the product of the regional clarity and the detail texture complexity.

8. A method for constructing a three-dimensional model of a Ferris wheel as claimed in claim 1, characterized in that: The method for obtaining the updated contrast is: Obtaining the contrast of each region in each Ferris wheel rotation image; The calculation formula for the updated contrast is: ; In the formula, Indicates the first Update contrast of each region; Indicates the first The dehazing weight of each region; Indicates the first The contrast of the area.

9. A Ferris wheel three-dimensional model construction system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method for constructing a three-dimensional model of a Ferris wheel as described in any one of claims 1 to 8 are implemented.

Images