A volume data acquisition method based on image scanning

Abstract

The present invention provides a volume data acquisition method based on image scanning, comprising: preprocessing the image data based on a filtering window to obtain contour information of the target object; inputting the contour information into a trained three-dimensional reconstruction network to obtain volume data; wherein, using the filtering window to detect noise points on the image data to obtain a noise mean; when the noise mean in the filtering window is greater than a preset threshold, denoising the image data in the corresponding filtering window; calculating the mean of the noise total value ZFj of all homogeneous noise point analysis lines obtained in the filtering window to obtain the noise mean; the present invention classifies the noise points in the filtering window, and based on the distribution of the noise points in the filtering window, performs interference analysis on the grayscale values ​​of the noise points through the influence between homogeneity and heterogeneity, thereby realizing a more accurate calculation of the noise points in the filtering window, and further making the calculation of the noise points more precise.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a volume data acquisition method based on image scanning. Background Art

[0002] In many fields, such as medicine, industry, and architecture, accurate acquisition of object volume information is crucial for design, production, and management.

[0003] In the calculation process, the traditional volume data acquisition method has difficulty in achieving a more accurate calculation of the noise points within the filter window, thereby ensuring a more precise calculation of the noise points. Summary of the invention

[0004] In order to overcome the deficiencies of the prior art, an object of the present invention is to provide a volume data acquisition method based on image scanning.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A volume data acquisition method based on image scanning, comprising:

[0007] Preprocessing the image data based on a filter window to obtain contour information of the target object;

[0008] Inputting the contour information into a trained three-dimensional reconstruction network to obtain volume data;

[0009] wherein, a filter window is used to detect noise points on the image data to obtain a noise mean;

[0010] When the noise mean in the filtering window is greater than a preset threshold, denoising the image data in the corresponding filtering window;

[0011] The total noise values ​​ZFj of all the obtained isotropic noise analysis lines in the filtering window are averaged to obtain the noise mean.

[0012] As a further technical solution of the present invention: it also includes:

[0013] The collected volume data is processed and analyzed to generate a volume data report.

[0014] As a further technical solution of the present invention: the total noise value ZFj includes:

[0015] The total noise value ZFj of the homogeneous noise analysis line is calculated by the formula, where i represents the number of all noise points on the homogeneous noise analysis line, Xgi represents the interference influence coefficient of the i-th noise point on the homogeneous noise analysis line, and ZFdi represents the grayscale value of the i-th noise point on the homogeneous noise analysis line.

[0016] As a further technical solution of the present invention: the gray value of the ith noise point on the homogeneous noise point analysis line is obtained:

[0017] When homogeneous noise points are obtained, the analysis points are connected with adjacent image points to obtain the middle line of homogeneous noise points;

[0018] The same-sex analysis of adjacent image points is continued until the endpoints of the middle line of the same-sex noise points and the adjacent image points are not all noise points, and the same-sex analysis of image points is stopped;

[0019] Get the same-sex noise starting line and the same-sex noise middle line, mark them as the same-sex noise analysis line, and get the grayscale values ​​of all noise points on the same-sex noise analysis line.

[0020] As a further technical solution of the present invention: the process of obtaining the same-sex noise points is as follows:

[0021] Conduct homogeneity analysis on adjacent image points;

[0022] If the adjacent image points are all noise points, the adjacent image points are marked as noise points of the same nature;

[0023] If the adjacent image points are not all noise points, the adjacent image points are marked as heterogeneous noise points;

[0024] When homogeneous noise points are obtained, adjacent homogeneous noise points are connected to obtain a homogeneous noise starting line, and the homogeneous noise points at both ends of the homogeneous noise starting line are used as analysis points, and homogeneous analysis is performed on the analysis points and adjacent image points;

[0025] If the analysis point and the adjacent image points are both noise points, the analysis point and the adjacent image points are marked as noise points of the same nature.

[0026] As a further technical solution of the present invention: the process of obtaining the interference influence coefficient Xgi is:

[0027] Obtain the same-sex interference influence coefficient and the opposite-sex interference influence coefficient, and mark them as Xgti and Xgyi respectively. Calculate the interference influence coefficient Xgi through the formula; where a1 and a2 are both proportional coefficients.

[0028] As a further technical solution of the present invention: the process of obtaining the same-sex interference influence coefficient Xgti is:

[0029] Taking the ith noise point of the homogeneous noise point analysis line as the analysis object, obtaining the grayscale value of the noise point adjacent to the ith noise point, and marking it as the ith adjacent noise point grayscale value;

[0030] The grayscale value of the ith adjacent noise point is calculated to be different from the grayscale value of the ith noise point to obtain the adjacent grayscale difference; the adjacent grayscale difference is calculated to be a ratio of the grayscale value of the ith noise point to obtain the first single point influence ratio BD1m, where m represents the number of noise points adjacent to the ith noise point;

[0031] The same-sex interference influence coefficient Xgti is calculated through the formula.

[0032] As a further technical solution of the present invention: the process of obtaining the heterosexual interference influence coefficient Xgyi is:

[0033] Obtain the non-noise point corresponding to the signal with large contrast between the noise point and the non-noise point, mark it as a non-noise point with large contrast, calculate the difference between the grayscale value of the non-noise point with large contrast and the grayscale value of the i-th noise point, and obtain the grayscale difference between adjacent non-noise points with large contrast;

[0034] Then, the ratio of the adjacent contrast noise non-grayscale difference to the grayscale value of the i-th noise point is calculated to obtain the second single point influence ratio BD2n, where n represents the number of non-noise points adjacent to the i-th noise point.

[0035] The heterosexual interference influence coefficient Xgyi is calculated through the formula.

[0036] As a further technical solution of the present invention: taking the ith noise point of the isotropic noise point analysis line as the analysis object, obtaining the grayscale value of the non-noise point adjacent to the ith noise point, and marking it as the grayscale value of the ith adjacent non-noise point;

[0037] The grayscale value of the ith adjacent non-noise point is calculated to be different from the grayscale value of the ith noise point to obtain the adjacent noise-non-grayscale difference; the adjacent noise-non-grayscale difference is compared with the adjacent noise-non-grayscale difference threshold;

[0038] If the adjacent noise non-grayscale difference is greater than or equal to the adjacent noise non-grayscale difference threshold, a signal with a large contrast between the noise point and the non-noise point is generated.

[0039] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0040] The present invention classifies the noise points in the filtering window, and based on the distribution of the noise points in the filtering window, performs interference analysis on the grayscale values ​​of the noise points through the influence between the same and different properties, thereby achieving a more accurate calculation of the noise points in the filtering window, thereby making the calculation of the noise points more precise. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 The flowchart is a method for determining the noise mean value in volume data acquisition according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The purpose of the present invention is to provide a volumetric data acquisition method based on image scanning. First, the image data is filtered using a filter window, so that the background area of ​​the image of the target object can be stripped out, making the contour and texture of the target area clearer, and the attention mechanism is introduced into the autoencoder network to achieve high-quality three-dimensional reconstruction of the target object, thereby improving the accuracy of volumetric data acquisition.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a volume data acquisition method based on image scanning, comprising:

[0047] Step 100: Scanning a target object based on an image scanning device to obtain image data of the surface of the target object; the number of the image scanning device is at least 1;

[0048] Step 200: pre-processing the image data based on a filtering window to obtain contour information of the target object;

[0049] Step 300: inputting the contour information into a trained 3D reconstruction network to obtain volume data;

[0050] The method for constructing the three-dimensional reconstruction network includes:

[0051] Step 301: Construct a neural network based on the autoencoder 2D-3D attention mechanism;

[0052] Step 302: Initialize the parameters of the neural network;

[0053] Step 302: feed-forward the input training image in the initialized neural network and calculate the training loss value of the input training image projection;

[0054] Step 304: using the error back propagation method to obtain the change value of the parameters of each layer in the neural network and update the parameters of the corresponding layer until the training loss value is lower than a preset threshold or the number of training times reaches a preset value, thereby obtaining a trained 3D reconstruction model.

[0055] Specifically, this embodiment can be applied to the medical field, the industrial field, and the construction field: the medical field is used to measure the volume of organs, tumors, etc., to assist medical diagnosis and treatment. The industrial field is used to measure the volume of parts, products, etc., to assist production and quality control. The construction field is used to measure the volume of buildings, land, etc., to assist design and planning.

[0056] Preferably, it also includes:

[0057] The collected volume data is processed and analyzed to generate a volume data report.

[0058] Specifically, this embodiment can perform data processing and analysis, and process and analyze the collected volume data, such as generating reports, comparing the volumes of different objects, and the like.

[0059] Furthermore, the 3D reconstruction model in this embodiment includes: a feature extraction network, a 3D model generation decoding network, and an attention mechanism network. The feature extraction network is mainly responsible for extracting features from the original data. The input image resolution is 64×64. In the embodiment of the present invention, a residual network is used to obtain low-dimensional rich features. Next, a convolutional long short-term memory network is used to make the features have both spatial and temporal connections. A 5×5 convolution kernel is used to obtain a 32×32×64 feature map. Finally, multi-layer convolution is used to control the dimension of the implicit vector. 3×3, 4×4, and 5×5 convolution kernels are used respectively, and finally a string of 1×1×512 implicit vectors is obtained. The implicit vector will be used as the input of the decoding part and the long short-term memory network.

[0060] In order to obtain the 3D model, the 3D model generation decoding network uses all 3D convolution kernels in a symmetrical form with the image encoding part. First, the convolutional long short-term memory network is used to obtain a 3×3×3, 512-channel feature map. Next, three layers of convolution are also used, and the convolution kernel sizes are 4×4×4, 5×5×5, and 6×6×6, respectively. Finally, a 32×32×32, 1-channel 3D model is obtained. The effective combination of convolution and convolutional long short-term memory network can ensure the accuracy of the 3D model.

[0061] The attention mechanism network part is designed to input the original data and the implicit vector obtained by the encoding part, so that the long short-term memory network is continuously updated and finally the angle of the image can be output. The image corresponding to the angle is input into the network again to facilitate faster completion of 3D reconstruction, and finally use as few images as possible to achieve high-precision 3D reconstruction results. In this network, the input image and the implicit vector are combined and input into the long short-term memory network. The long short-term memory network continuously updates the hidden layer, and then passes through the fully connected layer to obtain the image to be input for the next frame, and then inputs it into the image encoding network again to realize the cycle.

[0062] Preferably, preprocessing the image data based on a filtering window to obtain contour information of the target object includes:

[0063] Using a filter window to detect noise points on the image data, and obtaining a noise mean;

[0064] When the noise mean in the filtering window is greater than a preset threshold, denoising the image data in the corresponding filtering window;

[0065] Slide the filter window and return to the step of "when the noise mean in the filter window is greater than the preset threshold, denoising the image data in the corresponding filter window" until the entire image data is traversed to obtain denoised data;

[0066] Taking any point on the denoised data as the center, take a neighborhood window and calculate the grayscale average of all pixels in the neighborhood window;

[0067] The grayscale average value of the corresponding pixel points is taken as the output of the central pixel point to obtain the mean image data;

[0068] Calculating the correlation between the mean image data and the denoised data, and obtaining an optimal segmentation threshold according to the correlation;

[0069] The denoised data is segmented using the optimal segmentation threshold to obtain the segmented contour information.

[0070] Preferably, using a filter window to detect noise points on the image data to obtain a noise mean value comprises:

[0071] A noise point detection model is constructed according to the mean and median of each image point in the filtering window; the formula of the noise point detection model is: The noise point detection model is:

[0072] Among them, f(x) represents the similar noise value of pixel x, y(x) represents the gray value of pixel x, and u men (x) represents the grayscale mean of all pixels in the filter window centered on pixel x. represents the mean gradient of pixel x, is the grayscale median of all pixels in the filter window centered on pixel x, Represents the gradient value of pixel x in the horizontal direction;

[0073] Detecting each image point in the filtering window using the noise point detection model to obtain a similar noise value for each image point;

[0074] The corresponding image points with a value greater than the similar noise value are regarded as noise points;

[0075] The noise mean is determined according to the number of noise points and the number of each image point in the filter window.

[0076] The number of noise points is divided into regions. The specific division process is as follows:

[0077] According to the noise points determined in the image data within the filter window, classification analysis is performed to obtain noise point lines;

[0078] The classification analysis process includes the following steps:

[0079] Conduct homogeneity analysis on adjacent image points;

[0080] If the adjacent image points are all noise points, the adjacent image points are marked as noise points of the same nature;

[0081] If the adjacent image points are not all noise points, the adjacent image points are marked as heterogeneous noise points;

[0082] When homogeneous noise points are obtained, adjacent homogeneous noise points are connected to obtain a homogeneous noise starting line, and the homogeneous noise points at both ends of the homogeneous noise starting line are used as analysis points, and homogeneous analysis is performed on the analysis points and adjacent image points;

[0083] If the analysis point and the adjacent image points are both noise points, the analysis point and the adjacent image points are marked as noise points of the same nature;

[0084] If the analysis point and the adjacent image points are not all noise points, the analysis point and the adjacent image points are marked as heterogeneous noise points;

[0085] When homogeneous noise points are obtained, the analysis points are connected with adjacent image points to obtain the middle line of homogeneous noise points;

[0086] The same-sex analysis of adjacent image points is continued until the endpoints of the middle line of the same-sex noise points and the adjacent image points are not all noise points, and the same-sex analysis of image points is stopped;

[0087] Get the same-sex noise starting line and the same-sex noise middle line, mark them as the same-sex noise analysis line, and get the grayscale values ​​of all noise points on the same-sex noise analysis line;

[0088] The total noise value ZFj of the same-sex noise point analysis line is calculated by the formula ZFj=Xg1*ZFd1+Xg2*ZFd2+...+Xgi*ZFdi, where i represents the number of all noise points on the same-sex noise point analysis line, Xgi represents the interference influence coefficient of the i-th noise point on the same-sex noise point analysis line, and ZFdi represents the grayscale value of the i-th noise point on the same-sex noise point analysis line;

[0089] The noise total value ZFj of all the obtained isotropic noise analysis lines in the filter window is averaged to obtain the noise mean;

[0090] Among them, the process of obtaining the interference influence coefficient Xgi is:

[0091] Obtain the same-sex interference influence coefficient and the opposite-sex interference influence coefficient, and mark them as Xgti and Xgyi respectively. The interference influence coefficient Xgi is calculated; where a1 and a2 are both proportional coefficients, a1 is 0.65, and a2 is 0.68;

[0092] First, specifically, the process of obtaining the same-sex interference influence coefficient Xgti is:

[0093] Taking the ith noise point of the homogeneous noise point analysis line as the analysis object, obtaining the grayscale value of the noise point adjacent to the ith noise point, and marking it as the ith adjacent noise point grayscale value;

[0094] The grayscale value of the ith adjacent noise point is calculated to be different from the grayscale value of the ith noise point to obtain the adjacent grayscale difference; the adjacent grayscale difference is calculated to be a ratio of the grayscale value of the ith noise point to obtain the first single point influence ratio BD1m, where m represents the number of noise points adjacent to the ith noise point;

[0095] The same-sex interference influence coefficient Xgti is calculated by the formula Xgti=BD11+BD12+...+BD1m;

[0096] Second, specifically, the process of obtaining the heterosexual interference influence coefficient Xgyi is:

[0097] Take the ith noise point of the homogeneous noise point analysis line as the analysis object, obtain the grayscale value of the non-noise point adjacent to the ith noise point, and mark it as the ith adjacent non-noise point grayscale value;

[0098] The grayscale value of the ith adjacent non-noise point is calculated to be different from the grayscale value of the ith noise point to obtain the adjacent noise-non-grayscale difference; the adjacent noise-non-grayscale difference is compared with the adjacent noise-non-grayscale difference threshold;

[0099] If the adjacent noise non-grayscale difference is greater than or equal to the adjacent noise non-grayscale difference threshold, a signal with a large contrast between noise points and non-noise points is generated;

[0100] If the adjacent noise non-grayscale difference is less than the adjacent noise non-grayscale difference threshold, a small contrast signal between the noise point and the non-noise point is generated;

[0101] Obtain the non-noise point corresponding to the signal with large contrast between the noise point and the non-noise point, mark it as a non-noise point with large contrast, calculate the difference between the grayscale value of the non-noise point with large contrast and the grayscale value of the i-th noise point, and obtain the grayscale difference between adjacent non-noise points with large contrast;

[0102] Then, the ratio of the adjacent contrast noise non-grayscale difference to the grayscale value of the i-th noise point is calculated to obtain the second single point influence ratio BD2n, where n represents the number of non-noise points adjacent to the i-th noise point.

[0103] The heterosexual interference influence coefficient Xgyi is calculated by the formula Xgyi=BD21+BD22+...+BD2m;

[0104] Furthermore, the present invention classifies the noise points in the filtering window, and based on the distribution of the noise points in the filtering window, performs interference analysis on the grayscale values ​​of the noise points through the influence between the same and different properties, thereby achieving a more accurate calculation of the noise points in the filtering window, thereby making the calculation of the noise points more precise.

[0105] Furthermore, this embodiment constructs a noise point detection model based on the mean, median and gradient mean of each image point in the filtering window, which can detect the difference between noise points and original pixel points from multiple aspects, thereby making the detection of noise points more accurate.

[0106] Preferably, when the noise mean in the filtering window is greater than a preset threshold, denoising the image data in the corresponding filtering window comprises:

[0107] The pseudo pixel variance is calculated based on the grayscale median of all pixels in the filter window; wherein the pseudo pixel variance calculation formula is:

[0108] in,

[0109] represents the pseudo pixel variance of the pixel point (a, b) in the area of ​​the filter window with a size of (2n+1)×(2n+1), mean(a, b) represents the grayscale median of the pixel point (a, b) in the filter window, and x(k, l) represents the grayscale value of the pixel point at the position (k, l);

[0110] A window denoising model is constructed using the pseudo pixel variance; the formula of the window denoising model is:

[0111] Among them, f(a, b) represents the gray value of the pixel (a, b) after denoising, D is an adjustable coefficient, and x(a, b) represents the gray value of the pixel (a, b) in the filter window.

[0112] Optionally, the original filtering algorithm, such as the mean filtering algorithm, performs mean processing on the pixels in each neighborhood of the image data (regardless of whether there are noise points), so the processed image will become blurred, and the present invention can find the noise points on the image data by using the noise point detection model, and then filter the corresponding noise points to smooth out the noise points in the image while maintaining the original pixel information of the image data. In practical applications, the present invention can set the corresponding detection interval value according to the actual scenario.

[0113] Furthermore, the present invention is based on a window denoising model. By denoising the image data within the corresponding filter window, the problem of disappearance of certain feature gradients in the image data caused by existing denoising methods (such as median filtering denoising, mean filtering denoising and wavelet denoising, etc.) can be alleviated. The original pixel information of the image can be retained to the greatest extent, thereby improving the image interpretation effect.

[0114] Preferably, calculating the correlation between the mean image data and the denoised data, and obtaining an optimal segmentation threshold according to the correlation, comprises:

[0115] Extracting the grayscale values ​​at the same position on the denoised data and the mean image data to form a grayscale array;

[0116] constructing a segmentation function using the grayscale array;

[0117] Get the preset segmentation array, and continuously adjust the preset segmentation array until the value of the segmentation function is maximized;

[0118] The segmentation array corresponding to the maximum value of the segmentation function is used as the optimal segmentation threshold.

[0119] The present invention segments images based on the principle of histogram, and can obtain the optimal segmentation threshold value as a whole according to the probability of gray value distribution of the image. By using the segmentation threshold value to segment the image, the background area and the target area of ​​the image data can be separated, which is convenient for technicians to identify and extract the image data.

[0120] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0121] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A volume data acquisition method based on image scanning, characterized in that: include: Preprocessing the image data based on a filter window to obtain contour information of the target object; Inputting the contour information into a trained three-dimensional reconstruction network to obtain volume data; wherein, a filter window is used to detect noise points on the image data to obtain a noise mean; When the noise mean in the filtering window is greater than a preset threshold, denoising the image data in the corresponding filtering window; The total noise value ZFj of all the obtained isotropic noise analysis lines in the filter window is averaged to obtain the noise mean; The total noise value ZFj includes: The total noise value ZFj of the same-sex noise point analysis line is calculated by the formula ZFj=Xg1*ZFd1+Xg2*ZFd2+...+Xgi*ZFdi, where i represents the number of all noise points on the same-sex noise point analysis line, Xgi represents the interference influence coefficient of the i-th noise point on the same-sex noise point analysis line, and ZFdi represents the grayscale value of the i-th noise point on the same-sex noise point analysis line; Obtaining the grayscale value of the i-th noise point on the homogeneous noise point analysis line: When homogeneous noise points are obtained, the analysis points are connected with adjacent image points to obtain the middle line of homogeneous noise points; The same-sex analysis of adjacent image points is continued until the endpoints of the middle line of the same-sex noise points and the adjacent image points are not all noise points, and the same-sex analysis of image points is stopped; Get the same-sex noise starting line and the same-sex noise middle line, mark them as the same-sex noise analysis line, and get the grayscale values ​​of all noise points on the same-sex noise analysis line; The process of obtaining homogeneous noise is as follows: Conduct homogeneity analysis on adjacent image points; If the adjacent image points are all noise points, the adjacent image points are marked as noise points of the same nature; If the adjacent image points are not all noise points, the adjacent image points are marked as heterogeneous noise points; When homogeneous noise points are obtained, adjacent homogeneous noise points are connected to obtain a homogeneous noise starting line, and the homogeneous noise points at both ends of the homogeneous noise starting line are used as analysis points, and homogeneous analysis is performed on the analysis points and adjacent image points; If the analysis point and the adjacent image points are both noise points, the analysis point and the adjacent image points are marked as noise points of the same nature; The process of obtaining the interference influence coefficient Xgi is as follows: Obtain the same-sex interference influence coefficient and the opposite-sex interference influence coefficient, and mark them as Xgti and Xgyi respectively. The interference influence coefficient Xgi is calculated; where a1 and a2 are both proportional coefficients; The process of obtaining the same-sex interference influence coefficient Xgti is as follows: Taking the ith noise point of the homogeneous noise point analysis line as the analysis object, obtaining the grayscale value of the noise point adjacent to the ith noise point, and marking it as the ith adjacent noise point grayscale value; The grayscale value of the ith adjacent noise point is calculated to be different from the grayscale value of the ith noise point to obtain the adjacent grayscale difference; the adjacent grayscale difference is calculated to be a ratio of the grayscale value of the ith noise point to obtain the first single point influence ratio BD1m, where m represents the number of noise points adjacent to the ith noise point; The same-sex interference influence coefficient Xgt i is calculated by the formula Xgti=BD11+BD12+...+BD1m; The process of obtaining the heterogeneous interference influence coefficient Xgyi is as follows: Obtain the non-noise point corresponding to the signal with large contrast between the noise point and the non-noise point, mark it as a non-noise point with large contrast, calculate the difference between the grayscale value of the non-noise point with large contrast and the grayscale value of the i-th noise point, and obtain the grayscale difference between adjacent non-noise points with large contrast; Then, the ratio of the adjacent contrast noise non-grayscale difference to the grayscale value of the i-th noise point is calculated to obtain the second single point influence ratio BD2n, where n represents the number of non-noise points adjacent to the i-th noise point. The heterosexual interference influence coefficient Xgyi is calculated by the formula Xgyi=BD21+BD22+...+BD2m.

2. The volume data acquisition method based on image scanning according to claim 1, characterized in that: Also includes: The collected volume data is processed and analyzed to generate a volume data report.

3. The volume data acquisition method based on image scanning according to claim 2, characterized in that: Take the ith noise point of the homogeneous noise point analysis line as the analysis object, obtain the grayscale value of the non-noise point adjacent to the ith noise point, and mark it as the ith adjacent non-noise point grayscale value; The grayscale value of the ith adjacent non-noise point is calculated to be different from the grayscale value of the ith noise point to obtain the adjacent noise-non-grayscale difference; Compare the adjacent noise non-grayscale difference value with the adjacent noise non-grayscale difference threshold; If the adjacent noise non-grayscale difference is greater than or equal to the adjacent noise non-grayscale difference threshold, a signal with a large contrast between the noise point and the non-noise point is generated.

Images