Method for optimizing image information of patient with acute abdominal disease

Abstract

The invention relates to the field of image processing, and provides an acute abdominal disease patient image information optimization method, which comprises the following steps: acquiring an X-ray abdominal plain film image of an acute abdominal disease patient, and taking the acquired image as a to-be-detected image; analyzing the gray features corresponding to the to-be-detected image to obtain a highlight performance feature image corresponding to the to-be-detected image; determining a denoising parameter corresponding to the highlight expression feature image, and performing denoising processing on the highlight expression feature image by using the denoising parameter to obtain an optimal highlight expression feature image; and carrying out optimization processing on the to-be-detected image by utilizing the optimal highlight expression feature image to obtain a patient information acquisition image. According to the method, the quality of the acute abdominal X-ray abdominal plain film image can be improved, and a reliable basis is provided for subsequent further analysis.

Description

Technical Field

[0001] The present application relates to the field of image processing, and in particular to a method for optimizing image information of patients with acute abdomen. Background Art

[0002] Acute abdomen is a common acute disease with acute abdominal pain as the prominent manifestation, characterized by rapid onset and rapid progression. Under normal circumstances, the contrast changes between abdominal organ tissues do not have obvious changes. When patients have typical symptoms of acute abdomen, the density between abdominal tissues changes, and abnormal X-ray characteristics appear. Therefore, X-ray abdominal plain films can be performed on patients, and further analysis can be performed based on the characteristics of the patient's X-ray abdominal plain films.

[0003] However, due to the complex structure of the human abdomen, the X-ray scattering and environmental noise during the imaging process lead to poor contrast and unclear contour details in the X-ray abdominal plain film images of patients obtained after acquisition. This increases the difficulty for doctors to analyze X-ray abdominal plain films in the subsequent process, which is not conducive to further judgment of the symptoms of patients with acute abdomen by subsequent doctors. Summary of the invention

[0004] The present invention provides an image information optimization method for acute abdomen patients, which can improve the quality of acute abdomen X-ray abdominal plain film images and provide a reliable basis for subsequent further analysis.

[0005] In a first aspect, the present application provides a method for optimizing image information of an acute abdomen patient, comprising: acquiring an X-ray abdominal plain film image of the acute abdomen patient, and using the acquired image as an image to be detected; Analyze the grayscale features corresponding to the image to be detected, and then obtain a prominent feature image corresponding to the image to be detected; Determining denoising parameters corresponding to the highlighted feature image, and performing denoising processing on the highlighted feature image using the denoising parameters to obtain an optimal highlighted feature image; The image to be tested is optimized using the optimally highlighted characteristic image to obtain a patient information acquisition image.

[0006] Furthermore, the analyzing the grayscale features corresponding to the image to be detected, and then obtaining the highlighted feature image corresponding to the image to be detected, includes: Determine the difference index of each pixel relative to the neighboring pixels based on the grayscale value of the pixel in the image to be detected; The highlighting coefficient of each pixel is determined based on the difference index of each pixel relative to the neighboring pixels, thereby obtaining a highlighting feature image corresponding to the image to be detected.

[0007] Furthermore, the highlighting coefficient of each pixel is determined based on the difference index of each pixel relative to the neighboring pixels, and then the highlighting feature image corresponding to the image to be detected is obtained, including: Determine a difference index distribution histogram corresponding to the image to be detected based on the difference index of each pixel point relative to the neighboring pixel points; The highlighting coefficient of each pixel is determined based on the difference index distribution histogram, and then the highlighting feature image corresponding to the image to be detected is obtained.

[0008] Furthermore, the determining of denoising parameters corresponding to the highlighted feature image and performing denoising processing on the highlighted feature image using the denoising parameters to obtain an optimal highlighted feature image includes: The non-subsampled contourlet algorithm is used to process the prominent feature image to obtain multiple different frequency sub-band images; Calculate the adaptive threshold of each frequency sub-band image, and use the adaptive threshold to perform denoising on the corresponding frequency sub-band image; The denoised frequency sub-band image is subjected to inverse contourlet transform to obtain the optimal highlighting feature image.

[0009] Further, the calculating of the adaptive threshold of each frequency sub-band image includes: Calculate the adaptive threshold adjustment coefficient of each frequency sub-band image based on the median and mean of the contourlet coefficients after pixel point decomposition in each frequency sub-band image; The adaptive threshold of each frequency sub-band image is calculated based on the adaptive threshold adjustment coefficient of each frequency sub-band image and the number of row pixels and the number of column pixels of the frequency sub-band image.

[0010] Furthermore, the denoising process of the corresponding frequency sub-band image using the adaptive threshold comprises: The grayscale values ​​of the pixels in the frequency sub-band image whose grayscale values ​​are less than the adaptive threshold of the frequency sub-band image are assigned to 0, and then the corresponding frequency sub-band image is denoised.

[0011] Furthermore, the optimizing process of the image to be tested by using the optimal highlighted characteristic image to obtain the patient information acquisition image includes: Constructing a value change distribution histogram of the optimal highlighting feature image, and obtaining a value distribution probability curve based on the value change distribution histogram, and using the value distribution probability curve as a change function of the histogram equalization algorithm; The histogram equalization algorithm is used to optimize the image to be detected based on the variation function, thereby obtaining the patient information acquisition image.

[0012] Furthermore, the method of optimizing the image to be detected based on the variation function by using a histogram equalization algorithm to obtain a patient information acquisition image includes: Construct a gray value histogram of the image to be detected; The histogram equalization algorithm is used to optimize the image to be detected based on the variation function and the gray value histogram, thereby obtaining the patient information acquisition image.

[0013] Furthermore, the adaptive threshold adjustment coefficient of each frequency sub-band image is calculated using the following formula: in, , are the median and mean of the contourlet coefficients after pixel decomposition in the frequency subband image, is the empirical value of small error probability, is the adaptive threshold adjustment coefficient of the Lth frequency sub-band image; The adaptive threshold of each frequency subband image is calculated using the following formula: in, is the adaptive threshold of the Lth frequency subband image, They respectively represent the number of row pixels and column pixels of the Lth frequency subband image as a percentage of the total number of pixels. It indicates that The logarithmic function with base .

[0014] Furthermore, the highlighting coefficient of each pixel is calculated using the following formula: in, represents the highlight coefficient of pixel i, Represents the difference index distribution histogram of the image to be detected. The numerical value of the frequency of the difference index distribution of the horizontal axis interval is calculated as follows: , Indicates that the difference index value in the difference distribution histogram falls into the The number of pixels in the horizontal axis interval, j represents the jth horizontal axis interval, It represents the maximum grayscale value in the image to be detected, and Z represents the size of the neighborhood area of ​​the pixel point. Represents the pixel point in the image to be detected The gray value of the pixel at the position, m represents the number of horizontal axis intervals in the difference index distribution histogram of the image to be detected, Represents the pixel point in the image to be detected The difference index of the position relative to the neighboring pixels, Represents the difference index distribution histogram of the image to be detected. The mean of the difference index corresponding to the horizontal axis interval is Express request The absolute value of .

[0015] The beneficial effects of the present application are different from the prior art. The method for optimizing image information of acute abdomen patients of the present application includes: collecting X-ray abdominal plain film images of patients with acute abdomen, and using the collected images as images to be detected; analyzing the grayscale features corresponding to the images to be detected, and then obtaining the highlighted performance feature images corresponding to the images to be detected; determining the denoising parameters corresponding to the highlighted performance feature images, and using the denoising parameters to denoise the highlighted performance feature images, and obtaining the optimal highlighted performance feature images; using the optimal highlighted performance feature images to optimize the images to be detected, and then obtaining patient information acquisition images. This method can improve the quality of acute abdomen X-ray abdominal plain film images and provide a reliable basis for subsequent further analysis. In addition, after adaptive denoising and optimization according to the numerical variation characteristics of the original image, the embodiment of the present invention obtains the optimal highlighted feature image. The numerical distribution curve of the optimal highlighted feature image is more consistent with the characteristics of the X-ray abdominal plain film image of the patient with acute abdomen. Therefore, the numerical distribution curve is used to optimize the numerical variation characteristics of the histogram of the original grayscale image of the patient with acute abdomen, avoiding the shortcoming of the difficulty in lacking a suitable transformation function in the traditional histogram equalization calculation process, resulting in poor quality of the final X-ray abdominal plain film image of the patient with acute abdomen. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic structural diagram of an embodiment of a method for optimizing image information of an acute abdomen patient according to the present invention; Figure 2 for Figure 1 A schematic diagram of the structure of an embodiment of step S2; Figure 3 for Figure 1 A structural diagram of an embodiment of step S3 in FIG. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] Due to the complex structure of the human abdomen, the X-ray scattering, environmental noise and other characteristics during the imaging process lead to the poor contrast and unclear contour details in the X-ray abdominal plain film images of patients obtained after collection. This increases the difficulty of doctors analyzing X-ray abdominal plain films in the subsequent process, which is not conducive to further judgment. Therefore, it is necessary to optimize and improve the X-ray abdominal plain film images of patients with acute abdomen, enrich the image features of the patient area, improve the quality of abdominal plain film images of patients with acute abdomen, and provide a more reliable and accurate basis for judgment for subsequent processing. The following is a detailed description of this application in conjunction with the accompanying drawings and embodiments.

[0019] See also Figure 1 , Figure 1 This is a structural schematic diagram of an embodiment of a method for optimizing image information of acute abdomen patients of the present invention, which specifically includes: S1, collecting X-ray abdominal plain film images of patients with acute abdomen, and using the collected images as images to be detected.

[0020] Specifically, based on the principle that X-ray energy has different penetration strengths on different tissue structures of the human body, X-ray abdominal plain film images of patients with acute abdomen can be obtained through X-ray equipment, and the obtained X-ray abdominal plain film images of patients with acute abdomen can be used as images to be detected.

[0021] S2, analyzing the grayscale features corresponding to the image to be detected, and then obtaining a highlighted feature image corresponding to the image to be detected.

[0022] In one embodiment, please combine Figure 2 , step S2 includes steps S21 to S22.

[0023] Since different tissue structures in the human abdomen have different penetration of X-ray energy, the corresponding grayscale change numerical characteristics will appear on the X-ray abdominal plain film images of patients with acute abdomen. However, due to the influence of X-ray scattering and environmental noise, the quality of X-ray abdominal plain film images is poor and the details are not clear. Therefore, the change of pixel value difference in the X-ray abdominal plain film images of acute abdomen can be calculated to highlight the change of numerical difference in the X-ray abdominal plain film images.

[0024] Specifically, S21, based on the grayscale value of the pixel in the image to be detected, determine the difference index of each pixel relative to the neighboring pixel. In one embodiment, the difference index of each pixel relative to the neighboring pixel is calculated using the following formula: In the above formula, It represents the pixel points in the X-ray abdominal plain film image of acute abdomen, i.e. the image to be detected. The gray value of the pixel at the position, Represents the pixel The central local area The pixel point at the position It indicates the The gray value at the pixel position is Represents the pixel The total number of pixels in the local area centered at , where the size of the local area is , Take the experience value as 5, Represents the pixel point in the image to be detected The difference index of the position relative to the neighboring pixels.

[0025] The above formula can be used to calculate the difference index in the X-ray abdominal plain film image of acute abdomen. The value of the pixel When the pixel is located in the abdominal tissue area of ​​the acute abdomen X-ray abdominal plain film, the grayscale value of the pixel will be larger, while the value of the background pixel in the image will be relatively small. At this time, the calculated difference index of the acute abdomen X-ray abdominal plain film image is The value of will be correspondingly larger. By calculating the difference index of different pixel positions in the X-ray abdominal plain film of acute abdomen, the pixel points at different positions in the X-ray abdominal plain film of acute abdomen are highlighted.

[0026] S22, determining a highlighting coefficient of each pixel point based on a difference index of each pixel point relative to neighboring pixels, and then obtaining a highlighting feature image corresponding to the image to be detected.

[0027] Specifically, the difference index distribution histogram corresponding to the image to be detected is determined based on the difference index of each pixel point relative to the neighboring pixel points. Further calculation is performed based on the difference index value in the local area of ​​the acute abdomen X-ray abdominal plain film image, i.e., the image to be detected, to obtain the difference distribution histogram of the local area of ​​the acute abdomen X-ray abdominal plain film. In this histogram, the horizontal axis represents the difference index distribution interval of the pixel points in the local area image of the acute abdomen X-ray abdominal plain film, and the empirical interval is taken. The difference index of the pixels in the region is divided into different value intervals; the vertical axis represents the number of pixels in the region with the difference index in the interval. The difference distribution histogram of the acute abdomen X-ray abdominal plain film image obtained by calculation can better characterize the numerical distribution characteristics in the local image of the acute abdomen X-ray abdominal plain film.

[0028] The highlighting coefficient of each pixel is determined based on the difference index distribution histogram, and then the highlighting characteristic image corresponding to the image to be detected is obtained. Specifically, the highlighting coefficient of each pixel is calculated using the following formula: in, represents the highlight coefficient of pixel i, Represents the difference index distribution histogram of the image to be detected. The numerical value of the frequency of the difference index distribution of the horizontal axis interval is calculated as follows: , Indicates that the difference index value in the difference distribution histogram falls into the The number of pixels in the horizontal axis interval, j represents the jth horizontal axis interval, Indicates the maximum grayscale value in the image to be detected. Usually, the empirical value is , Z represents the size of the neighborhood area of ​​the pixel, Represents the pixel point in the image to be detected The gray value of the pixel at the position, m represents the number of horizontal axis intervals in the difference index distribution histogram of the image to be detected, Represents the pixel point in the image to be detected The difference index of the position relative to the neighboring pixels, Represents the difference index distribution histogram of the image to be detected. The mean of the difference index corresponding to the horizontal axis interval is Express request The absolute value of .

[0029] The above formula can be used to calculate the prominent performance coefficient of acute abdomen X-ray abdominal plain film. The value of the pixel When the pixel is located in the abdominal tissue pixel area in the acute abdomen X-ray abdominal plain film image, the pixel Grayscale at position The value is relatively large, and the pixel Gray level variation coefficient at position At the same time, the more the difference indexes of different pixels in the area are in the local area, and the greater the difference in the values ​​of the difference indexes at different pixel positions, the greater the difference in the values ​​of the pixel points obtained by calculation. The prominence coefficient at the position The value of will increase accordingly. By calculating the highlighting coefficients at different pixel positions in the X-ray abdominal plain film image of acute abdomen, the pixel points of abdominal tissue in the X-ray abdominal plain film image of acute abdomen can be effectively highlighted and represented.

[0030] Through the above steps, the value of the highlight performance coefficient in the acquired acute abdomen X-ray abdominal plain film image can be calculated, and the acute abdomen X-ray abdominal plain film abdominal tissue highlight performance characteristic image can be constructed and calculated according to the highlight performance coefficient. In the acquired acute abdomen X-ray abdominal plain film highlight performance characteristic image, the pixel value in the abdominal tissue structure is greater than the background pixel value, and the more complex the tissue changes, the larger the corresponding highlight performance characteristic image value.

[0031] S3, determining denoising parameters corresponding to the highlighted feature image, and performing denoising processing on the highlighted feature image using the denoising parameters to obtain an optimal highlighted feature image.

[0032] In one embodiment, please combine Figure 3 , step S3 includes steps S31 to S33.

[0033] S31, using a non-subsampled contourlet algorithm to process the salient feature image to obtain a plurality of different frequency sub-band images.

[0034] However, due to the strong randomness of the values ​​of the abnormal noise in the original image, the difference index with the surrounding normal pixels will also be relatively large, which will cause great interference to the prominent representation of the abdominal tissue features in the acute abdomen plain film. This noise value usually belongs to the high-frequency part, so the prominent feature image is used as input and processed using the non-subsampled contourlet (NSCT) algorithm. The original acute abdomen plain film prominent feature image is decomposed into multiple different high-frequency sub-bands.

[0035] S32, calculating an adaptive threshold of each frequency sub-band image, and performing denoising processing on the corresponding frequency sub-band image using the adaptive threshold.

[0036] In one embodiment, the adaptive threshold adjustment coefficient of each frequency sub-band image is calculated based on the median and mean of the contourlet coefficients after pixel point decomposition in each frequency sub-band image. Specifically, the adaptive threshold adjustment coefficient of each frequency sub-band image is calculated using the following formula: in, , are the median and mean of the contourlet coefficients after pixel decomposition in the frequency subband image, is the empirical value of small error probability, and its specific value is , is the adaptive threshold adjustment coefficient of the Lth frequency subband image. The median and mean values ​​of the contourlet coefficients of the subband image , The calculation is performed in two aspects to avoid the influence of large errors when a single value is used to represent the frequency sub-band image coefficients.

[0037] The adaptive threshold of each frequency sub-band image is calculated based on the adaptive threshold adjustment coefficient of each frequency sub-band image and the number of row pixels and column pixels of the frequency sub-band image. In one embodiment, the adaptive threshold of each frequency sub-band image is calculated using the following formula: in, is the adaptive threshold of the Lth frequency subband image, They respectively represent the number of row pixels and column pixels of the Lth frequency subband image as a percentage of the total number of pixels. Expressed The logarithmic function with base .

[0038] Furthermore, the grayscale values ​​of the pixels in the frequency sub-band image whose grayscale values ​​are less than the adaptive threshold of the frequency sub-band image are assigned to 0, and then the corresponding frequency sub-band image is denoised. Specifically, the frequency is retained as The sub-band image of The pixel value that is less than the threshold is set to 0, and the corresponding frequency sub-band image is denoised.

[0039] S33 is used to perform inverse contourlet transform on the frequency sub-band image after denoising to obtain an optimal highlighting feature image.

[0040] Specifically, the processed multiple images with different frequencies are reconstructed by inverse contourlet transform to obtain corresponding highlighted optimized performance feature images.

[0041] S4, optimizing the image to be tested by using the optimal highlighted feature image, and obtaining a patient information collection image.

[0042] In a specific embodiment, a numerical change distribution histogram of the optimal highlighted feature image is constructed, and a numerical distribution probability curve is obtained based on the numerical change distribution histogram, and the numerical distribution probability curve is used as a change function of a histogram equalization algorithm; the histogram equalization algorithm is used to optimize the image to be detected based on the change function, thereby obtaining a patient information acquisition image.

[0043] Specifically, through the above analysis and calculation, we can obtain the optimal prominent characteristic image of acute abdomen, in which the noise factor has little influence, and the characteristic of abdominal tissue change is more obvious. With the interval step size of 1, we can construct the value change distribution histogram of the acute abdomen prominent optimization characteristic image, and according to the histogram, we can get the value distribution probability curve, which is recorded as The numerical distribution probability curve As a variation function of the histogram equalization algorithm, the histogram equalization algorithm is used to optimize the image to be detected based on the variation function, thereby obtaining a patient information acquisition image.

[0044] In another embodiment, a grayscale value histogram of the image to be detected is constructed. Specifically, the maximum grayscale value G in the original acute abdominal X-ray image is , a grayscale value histogram of acute abdomen X-ray abdominal plain film can be constructed. In this histogram, the horizontal axis changes with grayscale level 1 to obtain the grayscale value change interval, and the vertical axis is the number of pixel values ​​falling into the grayscale interval. The histogram equalization algorithm is used to optimize the image to be detected based on the change function and the grayscale value histogram, and then the patient information acquisition image is obtained. For the grayscale value histogram of acute abdomen X-ray abdominal plain film, the histogram equalization algorithm is used to optimize the original acute abdomen X-ray abdominal plain film grayscale image, where the change function in the histogram equalization algorithm is the value change distribution histogram distribution probability curve of the acute abdomen highlighting the optimized performance feature image .

[0045] According to the distribution probability curve F of the acute abdomen X-ray abdomen plain film highlighting and optimizing the characteristic image, the original acute abdomen X-ray abdomen plain film grayscale image is optimized to enrich the detail change characteristics of the abdominal tissue in the original image, improve the contrast between the abdominal tissue and the background, and make the abdominal tissue in the acute abdomen X-ray abdomen plain film image clearer.

[0046] The embodiment of the invention combines the numerical difference change characteristics in the acute abdomen X-ray abdominal plain film image to calculate and construct the numerical value of the difference index, and calculates the numerical value of the highlight performance coefficient based on the numerical value. The effective characteristic information of the abdominal tissue in the acute abdomen X-ray abdominal plain film image is highlighted and calculated through the numerical value, enriching the abdominal detail characteristics in the acute abdomen X-ray abdominal plain film image. Furthermore, the embodiment of the invention constructs an acute abdomen X-ray abdominal plain film abdominal tissue highlight performance feature image based on the highlight performance coefficient, and calculates the corresponding adaptive dynamic change threshold value of the noise in the image, effectively weakening the impact of the noise in the image on the image clarity quality. At the same time, the original acute abdomen X-ray abdominal plain film image is equalized according to the distribution of the optimized image histogram, which effectively avoids the shortcomings of the traditional histogram equalization algorithm that the quality of the transformed image details is poor due to improper setting of the change function, improves the quality of the acute abdomen X-ray abdominal plain film image, and provides a reliable basis for subsequent further analysis.

[0047] The above are only implementation methods of the present application, and the protection scope of the present application is not limited thereto. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the protection scope of the present application.

Claims

1. A method for optimizing image information of patients with acute abdomen, characterized in that: include: Acquire X-ray abdominal plain film images of patients with acute abdomen, and use the acquired images as images to be detected; Analyze the grayscale features corresponding to the image to be detected, and then obtain a prominent feature image corresponding to the image to be detected; Determining denoising parameters corresponding to the highlighted feature image, and performing denoising processing on the highlighted feature image using the denoising parameters to obtain an optimal highlighted feature image; Utilizing the optimal highlighted characteristic image to optimize the image to be tested, thereby obtaining a patient information acquisition image; The step of determining denoising parameters corresponding to the highlighted feature image and performing denoising processing on the highlighted feature image using the denoising parameters to obtain an optimal highlighted feature image includes: The non-subsampled contourlet algorithm is used to process the prominent feature image to obtain multiple different frequency sub-band images; Calculate the adaptive threshold of each frequency sub-band image, and use the adaptive threshold to perform denoising on the corresponding frequency sub-band image; Perform inverse contourlet transform on the denoised frequency sub-band image to obtain the optimal highlighting feature image; The step of optimizing the image to be tested by using the optimal highlighted feature image to obtain a patient information acquisition image includes: Constructing a value change distribution histogram of the optimal highlighting feature image, and obtaining a value distribution probability curve based on the value change distribution histogram, and using the value distribution probability curve as a change function of the histogram equalization algorithm; The histogram equalization algorithm is used to optimize the image to be detected based on the variation function, thereby obtaining the patient information acquisition image.

2. The method for optimizing image information of acute abdomen patients according to claim 1, characterized in that: The step of analyzing the grayscale features corresponding to the image to be detected, and then obtaining a highlighted feature image corresponding to the image to be detected, includes: Determine the difference index of each pixel relative to the neighboring pixels based on the grayscale value of the pixel in the image to be detected; The highlighting coefficient of each pixel is determined based on the difference index of each pixel relative to the neighboring pixels, thereby obtaining a highlighting feature image corresponding to the image to be detected.

3. The method for optimizing image information of acute abdomen patients according to claim 2, characterized in that: The step of determining the highlighting coefficient of each pixel point based on the difference index of each pixel point relative to neighboring pixels, and then obtaining the highlighting feature image corresponding to the image to be detected, includes: Determine a difference index distribution histogram corresponding to the image to be detected based on the difference index of each pixel point relative to the neighboring pixel points; The highlighting coefficient of each pixel is determined based on the difference index distribution histogram, and then the highlighting feature image corresponding to the image to be detected is obtained.

4. The method for optimizing image information of patients with acute abdomen according to claim 1, characterized in that: The step of calculating the adaptive threshold of each frequency sub-band image comprises: Calculate the adaptive threshold adjustment coefficient of each frequency sub-band image based on the median and mean of the contourlet coefficients after pixel point decomposition in each frequency sub-band image; The adaptive threshold of each frequency sub-band image is calculated based on the adaptive threshold adjustment coefficient of each frequency sub-band image and the number of row pixels and the number of column pixels of the frequency sub-band image.

5. The method for optimizing image information of acute abdomen patients according to claim 1, characterized in that: The denoising process of the corresponding frequency sub-band image using the adaptive threshold comprises: The grayscale values ​​of the pixels in the frequency sub-band image whose grayscale values ​​are less than the adaptive threshold of the frequency sub-band image are assigned to 0, and then the corresponding frequency sub-band image is denoised.

6. The method for optimizing image information of acute abdomen patients according to claim 1, characterized in that: The method of optimizing the image to be detected based on the variation function by using the histogram equalization algorithm to obtain the patient information acquisition image includes: Construct a gray value histogram of the image to be detected; The histogram equalization algorithm is used to optimize the image to be detected based on the variation function and the gray value histogram, thereby obtaining the patient information acquisition image.

7. The method for optimizing image information of acute abdomen patients according to claim 4, characterized in that: The adaptive threshold adjustment coefficient of each frequency sub-band image is calculated using the following formula: in, , are the median and mean of the contourlet coefficients after pixel decomposition in the frequency subband image, is the empirical value of small error probability, is the adaptive threshold adjustment coefficient of the Lth frequency sub-band image; The adaptive threshold of each frequency subband image is calculated using the following formula: in, is the adaptive threshold of the Lth frequency subband image, They respectively represent the number of row pixels and column pixels of the Lth frequency subband image as a percentage of the total number of pixels. Expressed The logarithmic function with base .

8. The method for optimizing image information of acute abdomen patients according to claim 3, characterized in that: The highlight coefficient of each pixel is calculated using the following formula: in, represents the highlight coefficient of pixel i, Represents the difference index distribution histogram of the image to be detected. The numerical value of the frequency of the difference index distribution of the horizontal axis interval is calculated as follows: , Indicates that the difference index value in the difference distribution histogram falls into the The number of pixels in the horizontal axis interval, j represents the jth horizontal axis interval, It represents the maximum grayscale value in the image to be detected, and Z represents the size of the neighborhood area of ​​the pixel point. Represents the pixel point in the image to be detected The gray value of the pixel at the position, m represents the number of horizontal axis intervals in the difference index distribution histogram of the image to be detected, Represents the pixel point in the image to be detected The difference index of the position relative to the neighboring pixels, Represents the difference index distribution histogram of the image to be detected. The mean of the difference index corresponding to the horizontal axis interval is Express request The absolute value of .

Images