Image statistics based adaptive lut point cloud color mapping method and system

Abstract

The application discloses a kind of based on image statistics self-adapting LUT point cloud color mapping method and system, its method includes the following steps: reconstructing the three-dimensional point cloud of object and mapping to the coordinate system of corresponding texture image, obtain pixel coordinates;Texture image is converted to Lab color space, based on the color distribution characteristics of texture image, texture image is regionally segmented, and each area feature is extracted;Based on the color distribution characteristics of texture image, the extreme value of the pixel value of three color channels is calculated, and the extreme value of each color channel is adaptively adjusted using iterative histogram analysis method;Based on the extreme value of each color channel after adjustment and image segmentation area feature, lookup table is constructed;According to the pixel coordinates corresponding to three-dimensional point cloud, find the original color value in texture image, obtain the corresponding color output value of original color value through lookup table, output color point cloud.The application can efficiently and accurately generate color point cloud, effectively improve the texture restoration degree and visual expressiveness of three-dimensional point cloud model.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of computer vision, and in particular to an image statistics adaptive LUT point cloud color mapping method and system. BACKGROUND

[0002] Three-dimensional point cloud coloring technology has important application value in fields such as digital modeling, virtual reality, industrial detection, and cultural heritage protection. It significantly improves the realism and detail performance of the model by fusing two-dimensional texture information with three-dimensional geometric data. However, existing three-dimensional point cloud coloring methods still face many challenges in color information preservation, dynamic range compression, and chrominance correction.

[0003] Traditional three-dimensional reconstruction techniques can efficiently obtain geometric information of objects, but are easily affected by uneven lighting, wide dynamic range, and other factors during texture mapping, resulting in color distortion or loss of details. In addition, existing color correction methods rely heavily on manual parameter adjustment, making it difficult to adapt to brightness changes in different regions, resulting in a decline in overall color perception. In order to overcome the shortcomings of traditional three-dimensional point cloud coloring methods in color distortion, texture loss, and lighting robustness, three-dimensional reconstruction and texture mapping technology based on structured light projection has gradually become an important research direction to improve the realism of point clouds.

[0004] Texture mapping is a technique that maps two-dimensional images onto the surface of three-dimensional models. It establishes a correspondence between pixels on the image and vertices or faces of the three-dimensional model, thereby giving the model rich surface details and realism. In a digital raster projection system, texture mapping can combine three-dimensional reconstruction results to accurately map captured high-resolution image information onto reconstructed three-dimensional point clouds or mesh models, making the model not only have geometric shapes, but also present realistic colors and material characteristics. This process usually involves camera calibration, coordinate transformation, and texture coordinate calculation steps to ensure accurate alignment between two-dimensional images and three-dimensional models. Texture mapping technology is widely used in virtual reality, game development, industrial detection, and other fields, and can significantly improve the visual performance and application value of three-dimensional models.

[0005] Due to the diversity of color distribution and lighting conditions of texture images, as well as the differences in resolution and viewing angle between point cloud data and texture images, direct color mapping often cannot achieve ideal texture effects. SUMMARY

[0006] The main purpose of the present application is to provide an image statistics adaptive LUT point cloud color mapping method and system that can efficiently and accurately generate colored point clouds and improve the texture restoration degree of three-dimensional point clouds.

[0007] The technical solution adopted by the present application is:

[0008] Provided is an image-statistics-based adaptive LUT point cloud color mapping method, comprising the following steps:

[0009] S1, reconstruct a three-dimensional point cloud of an object by digital light projection and map to a coordinate system of a corresponding texture image to obtain pixel coordinates corresponding to all three-dimensional point clouds;

[0010] S2, convert the texture image to a Lab color space, perform region segmentation on the texture image based on color distribution characteristics of the texture image, and extract color, position and size characteristics of each region;

[0011] S3, based on the color distribution characteristics of the texture image, calculate extreme values of pixel values of three color channels, and adaptively adjust the extreme values of each color channel using an iterative histogram analysis method, stop iteration optimization when the extreme value of each channel of two adjacent iterations changes by less than a preset value, and obtain adjusted extreme values of each color channel;

[0012] S4, based on the adjusted extreme values of each color channel and the region segmentation characteristics of the image, construct a lookup table for color enhancement;

[0013] S5, find the original color value in the texture image according to the pixel coordinates corresponding to the three-dimensional point cloud, and then obtain the corresponding color output value of the original color value through the lookup table, and output a color point cloud.

[0014] In the above technical solution, during the reconstruction of the three-dimensional point cloud of the object, a camera is used to capture the object to obtain three-dimensional point cloud data, and a color picture is captured at the same position as a texture image; based on the internal and external parameters of the camera, each point cloud in the world coordinate system is projected to the coordinate system of the texture image through the camera matrix to obtain pixel coordinates.

[0015] In the above technical solution, the texture image is regionally segmented by a K-means clustering method in step S2.

[0016] In the above technical solution, after the texture image is converted to the Lab color space, the image complexity is calculated according to the standard deviation of the color channel, and the image is segmented into multiple regions according to the image complexity and the number of clusters.

[0017] In the above technical solution, the specific process of adaptively adjusting the extreme values of each color channel using an iterative histogram analysis method in step S3 is as follows:

[0018] In the initialization stage, the original extreme values of the pixel values of each color channel are obtained as the initial range;

[0019] During the iterative optimization phase, each iteration first calculates the histogram distribution of each channel within the current range, then redetermines the effective range based on the automatically changing proportion, adjusts the percentage of dark area ignoring and the percentage of highlight ignoring based on the exponential decay model, calculates the adjusted extreme values, and then introduces a convergence judgment mechanism. When the extreme value changes of each channel in two adjacent iterations are both less than the preset value, the iteration process is terminated, and the adjusted extreme values ​​are determined.

[0020] Following the above technical solution, step S4 specifically involves:

[0021] Based on the adjusted extreme values ​​of each color channel, a global mapping is first calculated. For each input pixel value of a color channel, when it is less than or equal to a first preset value... When the output global mapping value is 0, the output global mapping value is 0; when it is greater than or equal to the second preset value... When the value is 255, the output global mapping value is 255; when it is in When dealing with intervals, the effective range of the original pixel values ​​is determined. Linear stretching to [0, 255];

[0022] Calculate the brightness similarity weight and local adjustment amount for each segmented region, and then superimpose the global mapping value and the local adjustment amount to obtain the final mapping value of the enhanced pixel value.

[0023] A three-channel lookup table (LUT) is generated based on the final mapping value.

[0024] Following the above technical solution, in the three-channel lookup table (LUT), channel 0 stores the blue mapping value; channel 1 stores the green mapping value; and channel 2 stores the red mapping value.

[0025] The present invention also provides an image statistical adaptive LUT point cloud color mapping system, comprising:

[0026] The point cloud reconstruction and mapping module is used to reconstruct the 3D point cloud of an object through digital raster projection and map it to the coordinate system of the corresponding texture image to obtain the pixel coordinates of all 3D point clouds.

[0027] The region segmentation module is used to convert the texture image to the Lab color space, segment the texture image into regions based on the color distribution characteristics of the texture image, and extract the color, position and size features of each region;

[0028] The color optimization and adjustment module is used to calculate the extreme values ​​of pixel values ​​of the three color channels based on the color distribution characteristics of the texture image. It uses an iterative histogram analysis method to adaptively adjust the extreme values ​​of each color channel. The iterative optimization stops when the change of the extreme values ​​of each channel in two consecutive iterations is less than the preset value, and the adjusted extreme values ​​of each color channel are obtained.

[0029] The look-up table construction module is configured to construct a look-up table for color enhancement based on the adjusted color channel extreme values and the image segmentation region features.

[0030] The point cloud color mapping module is configured to find the original color value in the texture image according to the pixel coordinates corresponding to the three-dimensional point cloud, and then obtain the corresponding color output value of the original color value through the look-up table, and output the color point cloud.

[0031] According to the above technical solution, the point cloud reconstruction mapping module specifically generates the three-dimensional point cloud of the object by projecting a stripe pattern and collecting a deformed image, combining phase solving and triangulation.

[0032] The application further provides a computer storage medium, which stores a computer program executable by a processor, and the computer program executes the image-statistics-based adaptive LUT point cloud color mapping method according to the above technical solution.

[0033] The application has the following beneficial effects: the application adjusts the extreme values of the color channels adaptively through an iterative histogram analysis method, constructs a look-up table for color enhancement in combination with segmentation region feature analysis, effectively improves the texture mapping quality through the adaptive LUT color enhancement technology based on image statistics, and solves the color distortion problem caused by uneven illumination.

[0034] Further, a segmented linear color enhancement strategy combining global and local is adopted to adaptively optimize the color distribution of each region, enhance the detail performance of the low-contrast region, and make the three-dimensional point cloud coloring more natural and realistic.

[0035] Of course, implementing any product of the application does not necessarily need to achieve all the advantages mentioned above. DETAILED DESCRIPTION

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0037] Figure 1 is a flowchart of the image-statistics-based adaptive LUT point cloud color mapping method according to the embodiments of the application;

[0038] Figure 2 is a reconstructed point cloud image of the object of the point cloud color mapping system according to the application;

[0039] Fig. 3(a) is a texture picture collected by the point cloud color mapping system according to the application;

[0040] Figure 3(b) is the texture image processed by the LUT lookup table mapping algorithm of the present invention;

[0041] Figure 4 This is the result of K-means clustering segmentation of the original image of this invention;

[0042] Figure 5(a) is a histogram of the red channel of the original image of the present invention;

[0043] Figure 5(b) is a histogram of the green channel of the original image of the present invention;

[0044] Figure 5(c) is a histogram of the blue channel of the original image of the present invention;

[0045] Figure 6 This is the color mapping map of the point cloud for object reconstruction according to the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0047] It should be noted that the illustrations provided in the embodiments of the present invention are only schematic representations of the basic concept of the present invention. Therefore, the illustrations only show the components related to the present invention and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0048] In this invention, it should also be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first" and "second" are used only for descriptive and distinguishing purposes and should not be construed as indicating or implying relative importance.

[0049] Furthermore, it should be noted that the features of the various embodiments of the present invention can be combined or integrated in whole or in part, and as those skilled in the art will understand, they can interact and operate in different ways. Each embodiment can be implemented independently of each other or in association with one another.

[0050] like Figure 1 As shown, the image statistical adaptive LUT point cloud color mapping method of this invention includes the following steps:

[0051] S1, reconstruct a three-dimensional point cloud of an object by digital light projection and map to a coordinate system of a corresponding texture image, to obtain pixel coordinates corresponding to all three-dimensional point clouds;

[0052] S2, convert the texture image to a Lab color space, perform region segmentation on the texture image based on color distribution characteristics of the texture image, and extract color, position and size characteristics of each region;

[0053] S3, based on the color distribution characteristics of the texture image, calculate the extreme values of the pixel values of the three color channels, and adaptively adjust the extreme values of each color channel using an iterative histogram analysis method, stop iteration optimization when the extreme value changes of each channel of two adjacent iterations are less than a preset value, and obtain the adjusted extreme values of each color channel;

[0054] S4, based on the adjusted extreme values of each color channel and the image segmentation region characteristics, construct a lookup table for color enhancement;

[0055] S5, find the original color value in the texture image according to the pixel coordinates corresponding to the three-dimensional point cloud, and then obtain the corresponding color output value of the original color value through the lookup table, and output the color point cloud.

[0056] Further, the step S1 of the present application specifically adopts a digital light projection system, which includes a color camera, a digital light projector and a computer, the object to be reconstructed is placed below the digital projector and the color camera, a stripe pattern is projected and a deformed image is collected, combined with phase solving and triangulation, to generate a three-dimensional point cloud of the object, as shown in Figure 2 .

[0057] In the reconstruction process, the camera can be used to shoot the object to obtain three-dimensional point cloud data, and at the same time, a color picture is shot at the same position for texture mapping. Based on the internal and external parameters of the camera, for each point cloud point in the world coordinate system , through the camera matrix , according to the projection formula of the pinhole camera model:

[0058]

[0059] wherein is a scale factor, is an intrinsic matrix, is a rotation matrix, is a translation vector. Projected to the image coordinate system, the pixel coordinates .

[0060] Further, the step S2 of the present application is performed on the input texture image As shown in Fig. 3(a), the BGR image is converted to Lab color space, first by computing Lab image Standard deviation of the channels computes the image complexity :

[0061]

[0062] Here , are the standard deviations of the channel and channel in Lab space, respectively. The number of clusters is chosen according to If , ; , ; , ; , Here .

[0063] The image can be segmented into multiple regions by K-means clustering algorithm :

[0064]

[0065] where is the number of cluster centers Here , is the Lab color feature vector of a pixel, is the th cluster set, is the th cluster center, is the Euclidean norm (computes the vector distance). The result of K-means clustering segmentation is shown in Figure 4 .

[0066] Finally, the color feature of each region is computed: for each segmented region , its average color vector is computed; the position feature: the geometric barycenter of the region ; the size feature: the relative area of the region .

[0067]

[0068]

[0069] ​​​

[0070] wherein , is the total number of pixels in the region , is the channel of the image is the pixel value at position , is the barycentric coordinate of the direction, is the barycentric coordinate of the direction, is the image width, is the image height.

[0071] Further, the step S3 of the present application adjusts the input texture image As shown in Fig. 3(a), by traversing all the pixel points of the image , the extreme values of the three color channels , including the minimum pixel value and the maximum pixel value , are calculated respectively;

[0072] An iterative histogram analysis method can be used to adjust the color range of each channel: in the initialization stage, the original extreme values of each channel are obtained as the initial range; in the iterative optimization stage, the histogram distribution of each channel in the current range is calculated first in each iteration:

[0073]

[0074] Here is the pixel value, The red, green, and blue channel histograms of the original image are shown in Fig. 5(a), Fig. 5(b), and Fig. 5(c) respectively. Then the effective range is re-determined according to the automatically changing scale, and the dark area ignore percentage and the highlight ignore percentage are adjusted based on an exponential decay model:

[0075]

[0076]

[0077] Here is the number of current iteration. The adjusted minimum pixel value and the maximum pixel value are calculated:

[0078]

[0079]

[0080] represents the total number of pixels in the image, A convergence judgment mechanism is introduced. When the changes of the extreme values of each channel in two adjacent iterations are both less than a preset value (e.g., 2 pixel values), the iteration process is terminated, and the adjusted minimum pixel value is determined and the maximum pixel value Extreme noise or abnormal pixels are avoided from affecting the overall color distribution.

[0081] Further, the step S4 of the present application is based on the adjusted extreme values of each channel, and first calculates a global mapping. For each color channel, the input pixel value is mapped according to the following rules:

[0082] When ,

[0083] When , When ,

[0084] wherein is a mapping table of the pixel value in the color channel .

[0085] Then, for each segmented region , the brightness similarity weight and the adjustment amount are calculated:

[0086]

[0087]

[0088] wherein represents the normalized brightness of the current pixel value, represents the average brightness of the region . Finally, the global mapping value is superimposed with the local adjustment amount to obtain the final enhanced pixel value :

[0089]

[0090] A three-channel lookup table LUT with a size of 1x256 is generated, and the storage order is consistent with the BGR format of OpenCV: the 0th channel stores the blue (B) mapping value; the 1st channel stores the green (G) mapping value; and the 2nd channel stores the red (R) mapping value.

[0091] Further, the step S5 of the present application is performed for each point in the point cloud​​​ , according to the pre-calculated texture coordinates , i.e. the pixel coordinates above. If and (where and are the width and height of the texture image, respectively), the original color value is obtained from the texture image :

[0092]

[0093] The original color value is enhanced by a lookup table (LUT), and the adjusted value is obtained by looking up the table directly for each pixel channel value. Each color channel The input pixel value is used as an index to look up the corresponding output value from the LUT:

[0094]

[0095] where represents the mapping table in the lookup table for the corresponding channel , is the channel component of the original color value, and the target color value after color mapping processing is obtained . The texture picture after LUT processing (as shown in Fig. 3 (b)), it can be seen that the original texture picture (i.e. Fig. 3 (a)) is low in overall brightness, the details are blurred, and the details in the dark area are lost. The texture picture after LUT processing has significantly improved image quality.

[0096] The enhanced color value is converted from the BGR format of OpenCV to the RGB format supported by the PCL point cloud, the original geometric coordinates of the point are preserved , and the converted RGB color value is assigned to the point cloud, obtaining a three-dimensional point cloud with enhanced color effect, as shown in Figure 6 . The texture picture processed by the LUT lookup table is mapped to the point cloud, and the color brightness of the point cloud is significantly enhanced, the image quality is significantly improved, and the point cloud presents a clearer texture effect.

[0097] The image statistical adaptive LUT point cloud color mapping system based on the embodiments of the present application is mainly used to implement the above-mentioned method embodiments, and the system comprises:

[0098] A point cloud reconstruction mapping module is configured to reconstruct a three-dimensional point cloud of an object by digital raster projection and map the three-dimensional point cloud to a coordinate system of a corresponding texture image, to obtain pixel coordinates corresponding to all three-dimensional point clouds;

[0099] The regional division module is configured to convert the texture image to a Lab color space, divide the texture image into regions based on color distribution characteristics of the texture image, and extract color, position and size characteristics of each region;

[0100] The color optimization adjustment module is configured to calculate extreme values of pixel values of three color channels based on color distribution characteristics of the texture image, adaptively adjust the extreme values of the color channels using an iterative histogram analysis method, stop iteration optimization when the extreme values of each channel of two adjacent iterations change by less than a preset value, and obtain adjusted extreme values of each color channel.

[0101] The lookup table construction module is configured to construct a lookup table for color enhancement based on the adjusted extreme values of each color channel and the region characteristics of the image segmentation.

[0102] The point cloud color mapping module is configured to find original color values in the texture image according to pixel coordinates corresponding to the three-dimensional point cloud, and then obtain corresponding color output values of the original color values through the lookup table, and output the color point cloud.

[0103] The point cloud reconstruction mapping module specifically generates the three-dimensional point cloud of the object by projecting a stripe pattern and collecting a deformed image, combining phase solving and triangulation.

[0104] Each module is mainly used to implement each step of the method embodiment, and will not be described here.

[0105] The present application also provides a computer readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application store, etc., which stores a computer program, and the program is executed by a processor to realize corresponding functions. The computer readable storage medium of the present embodiment is executed by the processor to realize the image statistics based adaptive LUT point cloud color mapping method of the method embodiment.

[0106] In summary, the present application adopts a lookup table (LUT) based color mapping algorithm to optimize the quality and consistency of texture mapping. The LUT based color mapping of the present application is a process of analyzing color statistics characteristics of a texture image, constructing a piecewise linear mapping relationship, and redistributing an original color space to an enhanced color space, thereby effectively improving the contrast and color balance of texture mapping and improving the visual effect and realism of the point cloud model.

[0107] It should be noted that, according to the needs of implementation, each step / component described in the present application can be split into more steps / components, or two or more steps / components or part operations of the steps / components can be combined into a new step / component, to achieve the purpose of the present application.

[0108] The size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0109] It should be understood that, for those skilled in the art, improvements or changes can be made according to the above description, and all these improvements and changes shall fall within the protection scope of the appended claims of the present application.

Claims

1. A point cloud color mapping method based on image statistical adaptive LUT, characterized in that, Includes the following steps: S1. Reconstruct the 3D point cloud of the object through digital raster projection and map it to the coordinate system of the corresponding texture image to obtain the pixel coordinates corresponding to all 3D point clouds. S2. Convert the texture image to the Lab color space, segment the texture image into regions based on the color distribution characteristics of the texture image, and extract the color, position and size features of each region; S3. Based on the color distribution characteristics of the texture image, calculate the extreme values ​​of the pixel values ​​of the three color channels, and use an iterative histogram analysis method to adaptively adjust the extreme values ​​of each color channel. Stop the iterative optimization when the change of the extreme values ​​of each channel in two adjacent iterations is less than the preset value, and obtain the adjusted extreme values ​​of each color channel. S4. Based on the adjusted extreme values ​​of each color channel and the image segmentation region features, construct a lookup table for color enhancement; S5. Find the original color value in the texture image based on the pixel coordinates corresponding to the 3D point cloud, and then obtain the corresponding color output value of the original color value through a lookup table to output the colored point cloud. Specifically, step S4 involves: calculating the global mapping based on the extreme values ​​of each color channel after adjustment; calculating the brightness similarity weight and local adjustment amount of each segmented region; and superimposing the global mapping value and the local adjustment amount to obtain the final mapping value of the enhanced pixel value. A three-channel lookup table (LUT) is generated based on the final mapping value.

2. The image statistical adaptive LUT point cloud color mapping method according to claim 1, characterized in that, In the process of reconstructing the 3D point cloud of an object, a camera is used to photograph the object to obtain 3D point cloud data, and at the same location, a color image is taken as a texture image. Based on the camera's intrinsic and extrinsic parameters, each point cloud in the world coordinate system is projected onto the coordinate system of the texture image through the camera matrix to obtain pixel coordinates.

3. The image statistical adaptive LUT point cloud color mapping method according to claim 1, characterized in that, In step S2, the texture image is segmented using the K-means clustering method.

4. The image statistical adaptive LUT point cloud color mapping method according to claim 2, characterized in that, After converting the texture image to the Lab color space, the image complexity is calculated based on the standard deviation of the color channels, and the number of clusters is selected based on the image complexity to divide the image into multiple regions.

5. The image statistical adaptive LUT point cloud color mapping method according to claim 1, characterized in that, The specific process of adaptively adjusting the extreme values ​​of each color channel using the iterative histogram analysis method in step S3 is as follows: During the initialization phase, the original extreme values ​​of pixel values ​​for each color channel are obtained as the initial range; During the iterative optimization phase, each iteration first calculates the histogram distribution of each channel within the current range, then redetermines the effective range based on the automatically changing proportion, adjusts the percentage of dark area ignoring and the percentage of highlight ignoring based on the exponential decay model, calculates the adjusted extreme values, and then introduces a convergence judgment mechanism. When the extreme value changes of each channel in two adjacent iterations are both less than the preset value, the iteration process is terminated, and the adjusted extreme values ​​are determined.

6. The image statistical adaptive LUT point cloud color mapping method according to claim 1, characterized in that, The calculation of the global mapping in step S4 is specifically as follows: for each input pixel value of a color channel, when it is less than or equal to a first preset value... When the output global mapping value is 0, the output global mapping value is 0; when it is greater than or equal to the second preset value... When the global mapping value is 255, the output value is 255; when it is in When dealing with intervals, the effective range of the original pixel values ​​is determined. Linear stretch to [0, 255].

7. The image statistical adaptive LUT point cloud color mapping method according to claim 1, characterized in that, In the three-channel lookup table (LUT), channel 0 stores the blue mapping value; channel 1 stores the green mapping value; and channel 2 stores the red mapping value.

8. A point cloud color mapping system based on image statistical adaptive LUT, characterized in that, include: The point cloud reconstruction and mapping module is used to reconstruct the 3D point cloud of an object through digital raster projection and map it to the coordinate system of the corresponding texture image to obtain the pixel coordinates of all 3D point clouds. The region segmentation module is used to convert the texture image to the Lab color space, segment the texture image into regions based on the color distribution characteristics of the texture image, and extract the color, position and size features of each region; The color optimization and adjustment module is used to calculate the extreme values ​​of pixel values ​​of the three color channels based on the color distribution characteristics of the texture image. It uses an iterative histogram analysis method to adaptively adjust the extreme values ​​of each color channel. The iterative optimization stops when the change of the extreme values ​​of each channel in two consecutive iterations is less than the preset value, and the adjusted extreme values ​​of each color channel are obtained. The lookup table construction module is used to construct a lookup table for color enhancement based on the adjusted extreme values ​​of each color channel and the features of the image segmentation region. The point cloud color mapping module is used to find the original color value in the texture image based on the pixel coordinates of the 3D point cloud, and then obtain the corresponding color output value of the original color value through a lookup table to output the colored point cloud. Specifically, the lookup table construction module is used to calculate the global mapping based on the extreme values ​​of each color channel after adjustment; calculate the brightness similarity weight and local adjustment amount of each segmented region; and superimpose the global mapping value and the local adjustment amount to obtain the final mapping value of the enhanced pixel value. A three-channel lookup table (LUT) is generated based on the final mapping value.

9. The image statistical adaptive LUT point cloud color mapping system according to claim 8, characterized in that, The point cloud reconstruction mapping module specifically generates a 3D point cloud of an object by projecting a stripe pattern and acquiring deformed images, combined with phase calculation and triangulation.

10. A computer storage medium, characterized in that, It contains a computer program that can be executed by a processor, which performs the image statistical adaptive LUT point cloud color mapping method according to any one of claims 1-7.

Images