Method and system for evaluating multiphase mixing effects based on image segmentation-reorganization

Abstract

The application discloses a kind of based on image segmentation-reorganization's multiphase mixing effect evaluation method and system, it is related to multiphase flow detection technical field, first with the color image obtained by shooting to multiphase mixing process as real-time image, real-time image is carried out block processing, obtain after block image, then after block image is iteratively calculated to reorganize, each iteration will be after block image in the each pixel point of image block to be handled the pixel value of image block to be handled the average pixel value of image block, obtain reorganization image, and based on reorganization image calculation obtains reorganization score, finally based on the reorganization score calculated each time iteration calculates integral, to evaluate the multiphase mixing effect of multiphase mixing process according to integral, the smaller integral, the better multiphase mixing effect, to be able to directly process color image to evaluate multiphase mixing effect, can receive all information of image, and evaluation precision is high.

Description

Technical Field

[0001] This invention relates to the field of multiphase flow detection technology, and in particular to an evaluation method and system for the multiphase mixing effect based on image segmentation-reforming. Background Technology

[0002] Multiphase mixing processes, such as gas-liquid, gas-solid, or solid-liquid mixing processes, are widely found in many fields, including metallurgy, chemical engineering, and power plants. The quality of multiphase mixing directly affects the operating efficiency and reaction state of the mixing system. Generally speaking, mixing uniformity refers to the degree of uniform distribution of each component in any volume when various materials are mixed under the action of external force. The mixing uniformity of a mixing system is the most important indicator for evaluating the multiphase mixing effect of a multiphase mixing process. In the field of multiphase flow detection, non-invasive detection is a simple and reliable method, and image processing is an important branch of non-invasive detection methods. However, most existing image processing-based methods for evaluating multiphase mixing effects first convert the image to grayscale and then evaluate the multiphase mixing effect based on the grayscale image. This means that not all information from the image can be received during the evaluation process, resulting in low evaluation accuracy.

[0003] Therefore, it is particularly important to propose an evaluation method for the multiphase mixing effect based on color images. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for evaluating the multiphase mixing effect based on image segmentation and remodeling. This method can directly process color images to evaluate the multiphase mixing effect with high accuracy.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] An evaluation method for the multiphase mixing effect based on image segmentation-renormalization, the evaluation method comprising:

[0007] Acquire real-time images; the real-time images are color images obtained by capturing images of a multiphase mixing process;

[0008] The real-time image is divided into blocks to obtain a block-based image consisting of multiple image blocks of the same size;

[0009] Randomly select one of the image blocks as the image block to be processed;

[0010] The pixel values ​​of each pixel in the image block to be processed in the segmented image are unified to the average pixel value of the image block to be processed, thus obtaining a reshaped image; a reshaping score is calculated based on the reshaped image.

[0011] Select the next unselected image block as the image block to be processed in the next iteration, use the reconstructed image as the block-after image in the next iteration, and return to the step of "unifying the pixel values ​​of each pixel point of the image block to be processed in the block-after image to the average pixel value of the image block to be processed", until all the image blocks have been selected;

[0012] The integral is calculated based on the renormalization score obtained in each iteration, and the multiphase mixing effect of the multiphase mixing process is evaluated based on the integral; the smaller the integral, the better the multiphase mixing effect.

[0013] In some embodiments, the real-time image is also a grayscale image obtained by capturing a multiphase mixing process; the multiphase mixing process is a gas-liquid mixing process, a gas-solid mixing process, a solid-liquid mixing process, or a gas-solid-liquid mixing process.

[0014] In some embodiments, before segmenting the real-time image, the evaluation method further includes:

[0015] When the real-time image is a color image, the pixel value of each pixel in each channel of the real-time image is normalized to obtain a first processed image, and the first processed image is used as a new real-time image; the pixel value of each pixel in each channel of the first processed image is within [-1, 1].

[0016] When the real-time image is a grayscale image, the pixel value of each pixel in the real-time image is normalized to obtain a second processed image, and the second processed image is used as a new real-time image; the pixel value of each pixel in the second processed image is within [0, 1].

[0017] In some embodiments, the real-time image is also a temperature field distribution cloud map or a concentration field distribution cloud map.

[0018] In some embodiments, the formula for calculating the average pixel value of the image block to be processed is:

[0019]

[0020] Where, p ij (r) represents the average pixel value of the image block to be processed in the r-th iteration; λ1 represents the number of horizontal pixels in the image block to be processed, l = 1, 2, ..., λ1; λ2 represents the number of vertical pixels in the image block to be processed, m = 1, 2, ..., λ2; p l，m (r-1) is the pixel value of the pixel in the l-th row and m-th column of the image block to be processed in the (r-1)-th iteration.

[0021] In some embodiments, the formula for calculating the renormalization score is:

[0022]

[0023] Among them, Q r L is the renormalization score at the r-th iteration; r The number of horizontal image patches in the reconstructed image, i = 1, 2, ..., L r H r The number of vertical image patches in the reconstructed image, j = 1, 2, ..., H r ;p ij (r) represents the average pixel value of the image block in the i-th row and j-th column of the reshaped image during the r-th iteration.

[0024] In some embodiments, the formula for calculating the integral is:

[0025]

[0026] Where C is the integral; N is the total number of iterations; Q r+1 Q is the renormalization score at the (r+1)th iteration; r Let be the renormalization score at the r-th iteration.

[0027] An evaluation system for the multiphase mixture effect based on image segmentation-renormalization, the evaluation system comprising:

[0028] An image acquisition module is used to acquire real-time images; the real-time images are color images obtained by capturing images of a multiphase mixing process.

[0029] The block processing module is used to perform block processing on the real-time image to obtain a block image including multiple image blocks of the same size;

[0030] The iterative calculation module is used to randomly select one of the image blocks as the image block to be processed; unify the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed, to obtain a reshaped image; calculate a reshaping score based on the reshaped image; select the next unselected image block as the image block to be processed in the next iteration, use the reshaped image as the segmented image in the next iteration, and return to the step of "unifying the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed", until all the image blocks have been selected;

[0031] The evaluation module is used to calculate an integral based on the renormalization score obtained in each iteration, and to evaluate the multiphase mixing effect of the multiphase mixing process based on the integral; the smaller the integral, the better the multiphase mixing effect.

[0032] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0033] This invention provides a method and system for evaluating the multiphase mixing effect based on image segmentation and reshaping. First, a color image obtained by capturing the multiphase mixing process is used as a real-time image. The real-time image is segmented to obtain segmented images. Then, iterative calculations are performed on the segmented images to reshape them. In each iteration, an image block is randomly selected as the image block to be processed. The pixel values ​​of each pixel in the image block to be processed are unified to the average pixel value of the image block to obtain a reshaped image. A reshaping score is calculated based on the reshaped image. Finally, an integral is calculated based on the reshaping score obtained in each iteration. The multiphase mixing effect of the multiphase mixing process is evaluated based on the integral. The smaller the integral, the better the multiphase mixing effect. This method allows direct processing of color images to evaluate the multiphase mixing effect, can receive all image information, and has high evaluation accuracy. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of the evaluation method provided in Embodiment 1 of the present invention;

[0036] Figure 2 This is a schematic diagram of the processing procedure of the evaluation method provided in Embodiment 1 of the present invention;

[0037] Figure 3 This is a schematic diagram of the gas-liquid mixing process provided in Embodiment 1 of the present invention;

[0038] Figure 4 This is a system block diagram of the evaluation system provided in Embodiment 2 of the present invention. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] The purpose of this invention is to provide a method and system for evaluating the multiphase mixing effect based on image segmentation and remodeling. This method can directly process color images to evaluate the multiphase mixing effect with high accuracy.

[0041] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] Example 1:

[0043] This embodiment provides a method for evaluating the multiphase mixing effect based on image segmentation-remodeling, such as... Figure 1 and Figure 2 As shown, the evaluation method includes:

[0044] S1: Acquire a real-time image; the real-time image is a color image obtained by capturing the multiphase mixing process;

[0045] The multiphase mixing process in this embodiment can be a gas-liquid mixing process, a gas-solid mixing process, a solid-liquid mixing process, or a gas-solid-liquid mixing process. For example... Figure 3 As shown, this is a schematic diagram of the gas-liquid mixing process. The gas from the gas supply device enters the gas-liquid mixing tank through the gas pipeline and spray gun to carry out gas-liquid mixing. The gas-liquid mixing process is captured by a camera to obtain a real-time image of the gas-liquid mixing process. Subsequently, the real-time image is processed by S2-S4 to determine the mixing complexity of the gas-liquid mixing process, so as to evaluate the mixing effect of the gas-liquid mixing process.

[0046] The real-time image in this embodiment can be a color image obtained by capturing the multiphase mixing process, or it can be a grayscale image obtained by capturing the multiphase mixing process. Therefore, the evaluation method of this embodiment can be applied to both color and grayscale images, greatly expanding the applicability of multiphase mixing effect evaluation. The method for obtaining the real-time image is as follows: The multiphase mixing process is captured using a camera, and real-time video frames are extracted from the captured video to obtain the real-time image. The camera can be a high-speed camera.

[0047] Preferably, in this embodiment, the acquired real-time image can be preprocessed with Gaussian filtering or other methods to remove image noise and enhance image quality.

[0048] To facilitate subsequent processing, this embodiment can also perform pixel normalization processing on the real-time image. That is, before S2, the evaluation method of this embodiment further includes: when the real-time image is a color image, the pixel value of each pixel is a three-dimensional vector, and the pixel values ​​of the RGB components of each pixel are encoded into the range [-1, 1]. When the real-time image is a grayscale image, the pixel value of each pixel is transformed into a number located in the range [0, 1]. The specific process is as follows:

[0049] (1) When the real-time image is a color image, the pixel value of each pixel in each channel of the real-time image is normalized to obtain the first processed image, and the first processed image is used as the new real-time image. S2 is executed, and the pixel value of each pixel in each channel of the first processed image is within [-1, 1].

[0050] In this embodiment, the channels refer to the RGB three channels. Taking the R channel as an example, the formula used for normalization of the R channel is:

[0051]

[0052] Where, p R R represents the pixel value of the first processed image in the R channel; R represents the pixel value of the real-time image in the R channel.

[0053] It should be noted that the formula used for normalizing the other channels is the same as the formula used for normalizing the R channel. Simply replace the pixel values ​​of the R channel with the pixel values ​​of the other channels. This will not be elaborated further here.

[0054] (2) When the real-time image is a grayscale image, the pixel value of each pixel in the real-time image is normalized to obtain the second processed image, and the second processed image is used as the new real-time image. S2 is executed, and the pixel value of each pixel in the second processed image is within [0, 1].

[0055] At this point, the formula used for normalization is:

[0056]

[0057] Where p1 is the pixel value of the pixel in the second processed image; p is the pixel value of the pixel in the real-time image.

[0058] Preferably, the real-time image in this embodiment can also be a temperature field distribution cloud map or a concentration field distribution cloud map. The temperature field reflects the spatial and temporal distribution of temperature, and the concentration field reflects the spatial and temporal distribution of concentration. In this embodiment, the temperature field distribution cloud map and the concentration field distribution cloud map can be generated by imaging the temperature and concentration in a multiphase mixing process, or by imaging the temperature and concentration of other objects in other processes. Other objects can be materials, etc. In this case, the real-time image acquisition method can be: obtaining the temperature field distribution cloud map using an infrared thermal imager, and obtaining the concentration field distribution cloud map using a concentration measurement device. When the temperature field and concentration field that change over time are obtained, real-time image extraction can be performed on the temperature field and concentration field to obtain a real-time temperature field distribution image (i.e., a temperature field distribution cloud map) and a real-time concentration field distribution image (i.e., a concentration field distribution cloud map). Subsequently, the temperature field distribution cloud map or the concentration field distribution cloud map is processed using S2-S4, and can then be used to evaluate the distribution characteristics of the temperature field or the concentration field, expanding the scope of application of the evaluation method in this embodiment.

[0059] The real-time image in this embodiment can be an image within the time range of 0 to t1, an image within the time range of t1 to t2, or an image at a certain moment. That is, the real-time image in this embodiment can be one image or multiple images. When there are multiple consecutive images, the subsequent S2-S4 are executed cyclically for each image.

[0060] S2: Perform block processing on the real-time image to obtain a block image comprising multiple image blocks of the same size;

[0061] Assuming the real-time image in this embodiment has L horizontal pixels and H vertical pixels, and each image block has λ1 horizontal pixels and λ2 vertical pixels, then segmenting the real-time image is equivalent to dividing the L×H pixel-sized real-time image into multiple λ1×λ2 pixel-sized image blocks. The segmentation result is as follows: Figure 2 As shown, Figure 2 In this process, the image segmented is the image after block division, where the number of image blocks is determined by the values ​​of L, H, λ1, and λ2.

[0062] S3: Randomly select one of the image blocks as the image block to be processed; unify the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed, and obtain a reshaped image; calculate the reshaping score based on the reshaped image; select the next unselected image block as the image block to be processed in the next iteration, use the reshaped image as the segmented image in the next iteration, and return to the step of "unifying the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed", until all the image blocks have been selected;

[0063] This embodiment performs iterative calculations on the segmented image to reshape it. In each iteration, one image block is selected for processing. The pixel values ​​of all pixels within that block are replaced with uniform pixel values ​​to reshape the segmented image. A reshaping score is calculated until all image blocks have been processed, at which point the reshaping process ends. In this embodiment, an unselected image block can be randomly selected as the image block to be processed, regardless of the selection order. Alternatively, image blocks can be selected sequentially from left to right and from top to bottom.

[0064] In S3, when the image block is λ1×λ2 pixels in size, the formula for calculating the average pixel value of the image block to be processed is:

[0065]

[0066] Where, p ij (r) represents the average pixel value of the image block to be processed in the r-th iteration; λ1 represents the number of horizontal pixels in the image block to be processed, l = 1, 2, ..., λ1; λ2 represents the number of vertical pixels in the image block to be processed, m = 1, 2, ..., λ2, and all pixels in the image block to be processed are traversed through l and m; p l，m (r-1) represents the pixel value of the pixel in the l-th row and m-th column of the image block to be processed during the (r-1)-th iteration. It should be noted that when the real-time image is a color image, the pixel value can be selected from the pixel value of any channel.

[0067] The formula for calculating the renormal fraction is:

[0068]

[0069] Among them, Q r L is the renormalization score at the r-th iteration; r The number of horizontal image patches in the reconstructed image, i = 1, 2, ..., L r H r The number of vertical image patches in the reconstructed image, j = 1, 2, ..., H r ;p ij (r) represents the average pixel value of the image patch in the i-th row and j-th column of the reshaped image at the r-th iteration. The above formula means that the reshaping score is obtained by averaging the squares of the average pixel values ​​of all image patches in the reshaped image and multiplying them by the area of ​​the image patch.

[0070] When λ1=λ2=λ, the formula for calculating the reorganization fraction is:

[0071]

[0072] like Figure 2 As shown, in this embodiment, during each iteration, the image block to be processed can be directly replaced with a single pixel. The value of a single pixel is also the average pixel value of the image block to be processed. However, this will cause the resolution (i.e. the number of pixels) of the reorganized image and the image after segmentation to be different. At this time, it is necessary to rescale the coarser image (i.e., the reorganized image) to a finer image (i.e., the image after segmentation) to ensure that the number of pixels contained in the two is the same, that is, the image size is the same. Then the reorganization score is calculated. By ensuring that the image size does not change during the operation, the accuracy and reliability of the reorganization score calculation result are avoided.

[0073] S4: Calculate the integral based on the renormalization score obtained in each iteration, and evaluate the multiphase mixing effect of the multiphase mixing process based on the integral; the smaller the integral, the better the multiphase mixing effect.

[0074] The formula for calculating integrals is:

[0075]

[0076] Where C is the integral, N is the total number of iterations; Q r+1 Q is the renormalization score at the (r+1)th iteration; r Let C be the renormalization score at the r-th iteration. The integral C represents the features that appear at the new scale of the renormalized image obtained in the last iteration. By analyzing the magnitude of C, we can see the changes in the multiphase mixing effect. The smaller C is, the lower the image complexity, that is, the more uniform the image, and the better the multiphase mixing effect.

[0077] This embodiment addresses the problems existing in existing methods for evaluating multiphase mixing effects by providing a method based on image segmentation and reshaping. This method is simple, reliable, and highly universal, applicable to both color and grayscale images. It can be used to evaluate multiphase mixing effects as well as the distribution characteristics of temperature and concentration fields. In other words, this method can evaluate multiphase mixing effects based on color or grayscale images generated during the multiphase mixing process, understanding the evolution of the uniformity and complexity of each phase distribution. It can also be used to detect the uniformity or complexity of temperature and concentration fields, achieving precise control over the distribution states of different process fields. Because this method can be applied not only to the detection of multiphase mixing effects but also to the detection of other temperature and concentration fields, its reliability and universality are significantly enhanced.

[0078] Compared with existing methods for evaluating the effects of multiphase mixing, the evaluation method of this embodiment has the following advantages:

[0079] (1) The evaluation method of this embodiment can perform imaging detection of multiphase mixing without damaging the mixing system, and can intuitively display the multiphase mixing state.

[0080] (2) The evaluation method in this embodiment is simple to operate and has a wide range of applications. It can be used not only for the detection of multiphase mixing effect, but also for the detection of other temperature fields and concentration fields.

[0081] (3) The evaluation method in this embodiment can be calculated based on the original color image, preserving image information to the greatest extent possible, and has high evaluation accuracy.

[0082] (4) The evaluation method in this embodiment can quickly obtain the uniformity or complexity of the field distribution and accurately grasp the field distribution state and characteristics. In guiding the production process, it can reduce the inefficiency caused by unreasonable design of multiphase mixing systems and further optimize the design or various parameters of multiphase mixing systems.

[0083] Example 2:

[0084] This embodiment provides an evaluation system for the multiphase mixing effect based on image segmentation-renormalization, such as... Figure 4 As shown, the evaluation system includes:

[0085] Image acquisition module M1 is used to acquire real-time images; the real-time images are color images obtained by capturing images of a multiphase mixing process.

[0086] The block processing module M2 is used to perform block processing on the real-time image to obtain a block image including multiple image blocks of the same size;

[0087] The iterative calculation module M3 is used to randomly select one of the image blocks as the image block to be processed; unify the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed, and obtain a reshaped image; calculate a reshaping score based on the reshaped image; select the next unselected image block as the image block to be processed in the next iteration, use the reshaped image as the segmented image in the next iteration, and return to the step of "unifying the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed", until all the image blocks have been selected;

[0088] Evaluation module M4 is used to calculate an integral based on the renormalization score obtained in each iteration, and to evaluate the multiphase mixing effect of the multiphase mixing process based on the integral; the smaller the integral, the better the multiphase mixing effect.

[0089] Each embodiment in this specification focuses on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be found in the method section.

[0090] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for evaluating the multiphase mixing effect based on image segmentation-renormalization, characterized in that, The evaluation methods include: Acquire real-time images; the real-time images are color images obtained by capturing images of a multiphase mixing process; The real-time image is divided into blocks to obtain a block-based image consisting of multiple image blocks of the same size; Randomly select one of the image blocks as the image block to be processed; The pixel values ​​of each pixel in the image block to be processed in the segmented image are unified to the average pixel value of the image block to be processed, thus obtaining a reshaped image; a reshaping score is calculated based on the reshaped image. Select the next unselected image block as the image block to be processed in the next iteration, use the reconstructed image as the block-after image in the next iteration, and return to the step of "unifying the pixel values ​​of each pixel point of the image block to be processed in the block-after image to the average pixel value of the image block to be processed", until all the image blocks have been selected; The integral is calculated based on the renormalization score obtained in each iteration, and the multiphase mixing effect of the multiphase mixing process is evaluated based on the integral; the smaller the integral, the better the multiphase mixing effect. The formula for calculating the renormalization fraction is as follows: ； in, For the first The renormalization score at the next iteration; This represents the number of pixels in the horizontal direction of the image block to be processed. This represents the number of vertical pixels in the image block to be processed. The number of horizontal image patches in the reconstructed image. =1, 2, ..., ; The number of vertical image patches in the reconstructed image. =1, 2, ..., ; For the first In the reshaped image during the nth iteration, the th Line number The average pixel value of the image patch in the column; The formula for calculating the integral is: ； in, For integration; This represents the total number of iterations. For the first Renormalization score at +1 iteration; For the first The renormalization score at the next iteration.

2. The evaluation method according to claim 1, characterized in that, The real-time image is also a grayscale image obtained by capturing the multiphase mixing process; the multiphase mixing process is a gas-liquid mixing process, a gas-solid mixing process, a solid-liquid mixing process, or a gas-solid-liquid mixing process.

3. The evaluation method according to claim 2, characterized in that, Before performing block processing on the real-time image, the evaluation method further includes: When the real-time image is a color image, the pixel value of each pixel in each channel of the real-time image is normalized to obtain a first processed image, and the first processed image is used as a new real-time image; the pixel value of each pixel in each channel of the first processed image is within [-1, 1]. When the real-time image is a grayscale image, the pixel value of each pixel in the real-time image is normalized to obtain a second processed image, and the second processed image is used as a new real-time image; the pixel value of each pixel in the second processed image is within [0, 1].

4. The evaluation method according to claim 1, characterized in that, The real-time image is also a temperature field distribution cloud map or a concentration field distribution cloud map.

5. The evaluation method according to claim 1, characterized in that, The formula for calculating the average pixel value of the image block to be processed is: ； in, For the first The average pixel value of the image block to be processed in the next iteration; This represents the number of pixels in the horizontal direction of the image block to be processed. =1, 2, ..., ; This represents the number of vertical pixels in the image block to be processed. =1, 2, ..., ; For the first In the -1st iteration, the image block to be processed is in the... Line number The pixel value of the pixel in the column.

6. An evaluation system for the multiphase mixing effect based on image segmentation-renormalization, characterized in that, The evaluation system includes: An image acquisition module is used to acquire real-time images; the real-time images are color images obtained by capturing images of a multiphase mixing process. The block processing module is used to perform block processing on the real-time image to obtain a block image including multiple image blocks of the same size; The iterative calculation module is used to randomly select one of the image blocks as the image block to be processed; unify the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed, to obtain a reshaped image; calculate a reshaping score based on the reshaped image; select the next unselected image block as the image block to be processed in the next iteration, use the reshaped image as the segmented image in the next iteration, and return to the step of "unifying the pixel values ​​of each pixel of the image block to be processed in the segmented image to the average pixel value of the image block to be processed", until all the image blocks have been selected; The evaluation module is used to calculate an integral based on the renormalization score obtained in each iteration, and to evaluate the multiphase mixing effect of the multiphase mixing process based on the integral; the smaller the integral, the better the multiphase mixing effect. The formula for calculating the renormalization fraction is as follows: ； in, For the first The renormalization score at the next iteration; This represents the number of pixels in the horizontal direction of the image block to be processed. This represents the number of vertical pixels in the image block to be processed. The number of horizontal image patches in the reconstructed image. =1, 2, ..., ; The number of vertical image patches in the reconstructed image. =1, 2, ..., ; For the first In the reshaped image during the nth iteration, the th Line number The average pixel value of the image patch in the column; The formula for calculating the integral is: ； in, For integration; This represents the total number of iterations. For the first Renormalization score at +1 iteration; For the first The renormalization score at the next iteration.

Images