Weight-bearing X-ray imaging for Lisfranc joint injuries

Abstract

The invention relates to the technical field of image enhancement, and in particular to a weight-bearing X-ray image processing method for Lisfranc joint injury. Firstly, the quality credibility reflecting contrast requirements is determined according to the characteristics that when grayscale values ​​are relatively concentrated, the tissue structure distinction is poor and the lack of brightness details will lead to insufficient contrast of joint anatomical structures, as well as exposure interference conditions; then, on the basis of determining the contrast of a joint area, the edge clarity representing the contrast enhancement requirements is determined in combination with the characteristics that different edges in a joint area to be analyzed are easy to distinguish when the contrast requirements are low; thus, a more accurate contrast enhancement coefficient is determined according to the quality credibility and the edge clarity, and then the X-ray grayscale image is adaptively enhanced according to the contrast enhancement coefficient combined with histogram equalization, so that the obtained X-ray enhanced image is more accurate, and the weight-bearing X-ray image processing effect for Lisfranc joint injury is improved.

Description

Technical Field

[0001] The invention relates to the technical field of image enhancement, and in particular to a weight-bearing X-ray image processing method for Lisfranc joint injuries. Background Art

[0002] The Lifranc joint is located in the front of the foot and includes the joints between the first to fifth metatarsal bones and the corresponding cuneiform bones and foot bones. Lifranc joint injuries are usually caused by improper force on the foot, sprains or external trauma, which may affect the patient's walking ability and quality of life. Weight-bearing X-ray imaging is an X-ray imaging technology performed when the patient is weight-bearing, which can better reflect the dynamic state of the foot and the relative position between the joints, making the observation of Lifranc joint injuries clearer.

[0003] At present, the weight-bearing X-ray image processing method for Lifranc joint injuries is generally to use X-ray equipment to take multi-angle images of the foot under weight-bearing state, and then analyze the acquired X-ray images to evaluate the relative position and degree of deformation between joints. Existing X-ray imaging technology is limited by the imaging quality of the equipment, that is, the resolution and contrast are limited, resulting in blurred images in some cases, making it difficult to accurately evaluate slight changes in the joints. Therefore, after obtaining the X-ray image of the patient's Lisfranc joint under weight-bearing state, image enhancement processing is usually required to make the image clearer. Existing technologies usually use a contrast-limited adaptive histogram equalization algorithm to improve the contrast and detail texture of the image, thereby improving the clarity of observation of Lifranc joint injuries.

[0004] For the contrast-limited adaptive histogram equalization algorithm, the cliplimit parameter is a very important parameter, which is used to control the cumulative intensity of pixels in the histogram equalization process. When the cliplimit parameter is set too large, the contrast will be over-enhanced, resulting in noise amplification and possible artifacts, making the enhanced image less clear. When the cliplimit parameter set a priori is too small, the contrast will be insufficiently enhanced, resulting in missing image details and failure to effectively highlight important structural features. The cliplimit parameter in the contrast-limited adaptive histogram equalization algorithm in the prior art is usually a fixed parameter set manually. However, due to the influence of the external environment, the texture details of X-ray images of different Lisfranc joints are usually different, resulting in different requirements for contrast enhancement. Therefore, in order to make the enhanced X-ray image clearer, different cliplimit parameters are usually required to be set; therefore, the prior art uses a fixed cliplimit parameter in combination with a contrast-limited adaptive histogram equalization algorithm to enhance different X-ray images, but does not take into account the contrast enhancement requirements of X-ray images under the influence of the external environment, resulting in low image enhancement accuracy for X-ray images of Lisfranc joints, and poor weight-bearing X-ray image processing effect for Lisfranc joint injuries. Summary of the invention

[0005] The present application provides a weight-bearing X-ray image processing method for Lisfranc joint injury. The method first determines the grayscale distribution validity according to the characteristics that the tissue structure distinction is poor when the grayscale value is relatively concentrated and the lack of brightness details will lead to insufficient contrast of the joint anatomical structure; then determines the exposure interference according to the grayscale characteristics of underexposure or overexposure; thereby combining the grayscale distribution validity and the exposure interference to determine the quality credibility that reflects the contrast requirement to a certain extent; then, according to the grayscale distribution disorder in the joint area to be analyzed and the corresponding number of edge lines, determines the contrast of the joint area in the initial situation, and on this basis, combined with the characteristics that different edges in the joint area to be analyzed are easy to distinguish when the contrast requirement is low, determines the quality credibility that reflects the contrast requirement to a certain extent. The invention relates to an X-ray image processing method for enhancing the image of a Lisfranc joint by using a fixed cliplimit parameter combined with a contrast-limited adaptive histogram equalization algorithm. ...

[0006] The first aspect of the present application provides a weight-bearing X-ray image processing method for Lisfranc joint injury, comprising:

[0007] Acquire an X-ray grayscale image of the Lisfranc joint of the patient under weight-bearing condition; obtain a joint area to be analyzed in the X-ray grayscale image;

[0008] Determine the corresponding grayscale distribution validity according to the grayscale value distribution concentration and the lack of brightness details in the X-ray grayscale image; determine the corresponding exposure interference according to the grayscale distribution extreme situation in the X-ray grayscale image; determine the quality credibility of the X-ray grayscale image according to the grayscale distribution validity and the exposure interference;

[0009] In the X-ray grayscale image, the joint area contrast of the joint area to be analyzed is determined according to the grayscale distribution disorder in the joint area to be analyzed and the number of edge lines after edge detection; the edge clarity of the joint area to be analyzed is determined according to the overall deviation of the grayscale texture between the edge lines in the joint area to be analyzed and the joint area contrast;

[0010] The contrast enhancement coefficient of the X-ray grayscale image is determined according to the quality credibility and the edge clarity; the X-ray grayscale image is enhanced according to the contrast enhancement coefficient in combination with histogram equalization to determine an X-ray enhanced image.

[0011] Furthermore, the grayscale distribution validity acquisition process includes:

[0012] The mean grayscale value of all pixels in the X-ray grayscale image is used as the reference grayscale mean; the difference between the grayscale value of each pixel in the X-ray grayscale image and the reference grayscale mean is used as the reference standard grayscale deviation of each pixel; the mean of the reference standard grayscale deviations of all pixels in the X-ray grayscale image is used as the grayscale distribution discreteness of the X-ray grayscale image;

[0013] In the X-ray grayscale image, taking pixels whose grayscale values ​​are greater than a preset upper grayscale threshold or whose grayscale values ​​are less than a preset lower grayscale threshold as brightness detail pixels; and counting the number of brightness detail pixels in the X-ray grayscale image;

[0014] The grayscale distribution validity of the X-ray grayscale image is determined according to the grayscale distribution discreteness and the number of brightness detail pixels; the grayscale distribution discreteness and the number of brightness detail pixels are both positively correlated with the grayscale distribution validity.

[0015] Furthermore, the process of determining the grayscale distribution validity of the X-ray grayscale image according to the grayscale distribution discreteness and the number of brightness detail pixels includes:

[0016] The product of the grayscale distribution discreteness and the number of brightness detail pixels is normalized to determine the grayscale distribution validity of the X-ray grayscale image.

[0017] Furthermore, the exposure interference degree acquisition process includes:

[0018] The mean value between the maximum grayscale value and the minimum grayscale value of all pixels in the X-ray grayscale image is used as the reference grayscale middle value; the difference between the grayscale value of each pixel in the X-ray grayscale image and the reference grayscale middle value is used as the exposure-affected grayscale deviation of each pixel; the mean value of the exposure-affected grayscale deviation of all pixels is normalized to determine the exposure interference degree of the X-ray grayscale image.

[0019] Furthermore, the process of obtaining the quality credibility includes:

[0020] Normalization is performed according to the product between the negative correlation mapping value of the exposure interference and the grayscale distribution validity to determine the quality credibility of the X-ray grayscale image.

[0021] Furthermore, the process of acquiring the contrast of the joint area includes:

[0022] The product of the gray value variance of all pixels in the joint area to be analyzed and the number of edge lines is used as the joint area contrast of the joint area to be analyzed.

[0023] Furthermore, the process of obtaining edge definition includes:

[0024] The average grayscale value of all edge pixels on each edge line is taken as the edge grayscale feature value of each edge line;

[0025] Select each edge line and each other edge line in turn as an edge line binary group; count all edge line binary groups corresponding to all edge lines;

[0026] The difference between the edge grayscale feature values ​​of the two edge lines in each edge line binary group is used as the edge line local discrimination degree of each edge line binary group; the average of the edge line local discrimination degrees of all edge line binary groups is used as the edge line overall discrimination degree of the joint area to be analyzed;

[0027] The edge clarity of the joint area to be analyzed is determined according to the overall discrimination degree of the edge line and the contrast of the joint area; both the overall discrimination degree of the edge line and the contrast of the joint area are positively correlated with the edge clarity.

[0028] Furthermore, the process of determining the edge clarity of the joint area to be analyzed according to the overall distinction of the edge line and the contrast of the joint area includes:

[0029] The product of the overall discrimination degree of the edge line and the contrast of the joint area is normalized to determine the edge clarity of the joint area to be analyzed.

[0030] Furthermore, the contrast enhancement coefficient acquisition process includes:

[0031] A contrast enhancement coefficient of the X-ray grayscale image is determined according to a mean value between the quality confidence and the edge definition.

[0032] Furthermore, the process of acquiring the X-ray enhanced image includes:

[0033] The positive correlation mapping value of the contrast enhancement coefficient is used as the cliplimit parameter in the contrast limited adaptive histogram equalization algorithm; the X-ray grayscale image is enhanced according to the contrast limited adaptive histogram equalization algorithm combined with the cliplimit parameter to determine the X-ray enhanced image.

[0034] This application includes the following benefits:

[0035] For X-ray grayscale images, different tissue structures usually have different grayscale values. If the overall grayscale value is concentrated in a relatively narrow grayscale range, it means that the overall contrast of the image is low and the details between different tissue structures are difficult to distinguish; and if the image lacks low grayscale values ​​or high grayscale values ​​as a whole, it means that the image lacks dark or bright details, resulting in insufficient contrast of the joint anatomical structure; therefore, in combination with the above characteristics, the present application first determines the corresponding grayscale distribution validity according to the grayscale value distribution concentration and the lack of brightness details in the X-ray grayscale image, thereby indirectly reflecting the image contrast requirements based on the grayscale distribution validity. Then it is necessary to consider that the grayscale distribution will be affected by the exposure of the X-ray equipment, and the credibility of the grayscale distribution validity will be reduced when the exposure quality is poor. Therefore, according to the relatively extreme characteristics of the corresponding grayscale distribution when underexposure or overexposure occurs, the exposure interference degree that characterizes the impact of exposure quality is determined, thereby combining the exposure interference degree and the grayscale distribution validity to comprehensively determine the quality credibility that can characterize the image contrast requirements.

[0036] Then, the joint area to be analyzed in the X-ray grayscale image is analyzed. First, according to the characteristics that the light-dark contrast represented by the concentrated grayscale distribution in the joint area to be analyzed is usually not obvious, and the characteristics that the fewer edge lines in the joint area to be analyzed usually indicate unclear boundaries between different structures and regions in the image, the joint area contrast of the joint area to be analyzed is determined according to the chaotic grayscale distribution in the joint area to be analyzed and the number of edge lines after edge detection, so that the lower the joint area contrast, the greater the need for contrast enhancement; and when the overall deviation of the grayscale texture between different edge lines is greater, it can be explained that the distinction of the edge lines in the corresponding joint area to be analyzed is clearer, and the feature representation of the details in the key area is more obvious, then the damaged area or damaged state can be observed more clearly in the joint area to be analyzed, and the corresponding contrast requirement will be smaller; therefore, on the basis of the overall deviation of the grayscale texture between the edge lines in the joint area to be analyzed and combined with the joint area contrast, the edge clarity representing the need for contrast enhancement is comprehensively determined.

[0037] Both quality credibility and edge clarity can characterize the contrast enhancement requirements to a certain extent. Therefore, a more accurate contrast enhancement coefficient is finally determined according to the quality credibility and edge clarity, so that the X-ray grayscale image is adaptively enhanced according to the contrast enhancement coefficient combined with histogram equalization, so that the obtained X-ray enhanced image is more accurate and the weight-bearing X-ray image processing effect for Lisfranc joint injury is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A flow chart of a weight-bearing X-ray image processing method for Lisfranc joint injury provided by one embodiment of the present invention;

[0040] Figure 2 A schematic diagram of an X-ray grayscale image of a Lisfranc joint of a patient in a weight-bearing state provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following is a detailed description of a weight-bearing X-ray image processing method for Lisfranc joint injury proposed according to the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0042] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0043] The following is a detailed description of a specific scheme of a weight-bearing X-ray image processing method for Lisfranc joint injury provided by the present invention in conjunction with the accompanying drawings.

[0044] The present application embodiment provides a weight-bearing X-ray image processing method for Lisfranc joint injury, please refer to Figure 1 , which shows a flow chart of a weight-bearing X-ray image processing method for Lisfranc joint injury provided by an embodiment of the present invention, the method comprising:

[0045] Step S101: Acquire an X-ray grayscale image of the Lisfranc joint of the patient in a weight-bearing state; and obtain a joint region to be analyzed in the X-ray grayscale image.

[0046] In a specific implementation of the embodiment of the present invention, when the patient is standing or sitting in a specific weight-bearing state, the Lisfranc joint is photographed by an X-ray device to obtain a corresponding X-ray grayscale image, and the weight-bearing device selects a foot pedal to ensure that the patient remains stable during the photographing process; then the X-ray grayscale image is grayscaled to determine the required X-ray grayscale image; please refer to Figure 2 , which shows a schematic diagram of an X-ray grayscale image of a patient's Lisfranc joint under a weight-bearing state provided by an embodiment of the present invention. Considering that the purpose of the embodiment of the present invention is to determine the degree to which the contrast of the X-ray grayscale image needs to be enhanced, and the enhanced image is ultimately used to observe Lisfranc joint injuries, the degree to which the contrast needs to be enhanced can be measured by analyzing the contrast requirements of the Lisfranc joint area; therefore, the joint area to be analyzed in the X-ray grayscale image, that is, the area corresponding to the Lisfranc joint, is further obtained.

[0047] In a specific implementation of an embodiment of the present invention, the process of acquiring the joint area to be analyzed includes: inputting the X-ray grayscale image into a trained U-Net network model, outputting the Lisfranc joint area in the X-ray grayscale image, and using the Lisfranc joint area as the area to be analyzed; the U-Net network model can be trained with the data of the Lisfranc joint area in the authorized medical database. It should be noted that the U-Net network model is a technical means well known to those skilled in the art, and the implementer can choose other neural network models according to the specific implementation environment, which will not be further limited or elaborated here.

[0048] Step S102: Determine the corresponding grayscale distribution validity according to the grayscale value distribution concentration and the lack of brightness details in the X-ray grayscale image; determine the corresponding exposure interference according to the extreme grayscale distribution in the X-ray grayscale image; determine the quality credibility of the X-ray grayscale image according to the grayscale distribution validity and the exposure interference.

[0049] For X-ray grayscale images, different tissue structures usually have different grayscale values. If the overall grayscale value is concentrated in a relatively narrow grayscale range, it means that the overall contrast of the image is low and the details between different tissue structures are difficult to distinguish; and if the image lacks low grayscale values ​​or high grayscale values ​​as a whole, it means that the image lacks dark or bright details, resulting in insufficient contrast of the joint anatomical structure; therefore, in combination with the above characteristics, this application first determines the corresponding grayscale distribution validity based on the grayscale value distribution concentration and the lack of brightness details in the X-ray grayscale image, thereby indirectly reflecting the image contrast requirements based on the grayscale distribution validity. Then it is necessary to consider that the grayscale distribution will be affected by the exposure of the X-ray equipment, and the credibility of the grayscale distribution validity will be reduced when the exposure quality is poor. Therefore, according to the relatively extreme characteristics of the corresponding grayscale distribution when underexposure or overexposure occurs, the exposure interference degree that characterizes the impact of exposure quality is determined, thereby combining the exposure interference degree and the grayscale distribution validity to comprehensively determine the quality credibility that can characterize the image contrast requirements.

[0050] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the grayscale distribution validity includes:

[0051] The mean grayscale value of all pixels in the X-ray grayscale image is taken as the reference grayscale mean; the difference between the grayscale value of each pixel in the X-ray grayscale image and the reference grayscale mean is taken as the reference standard grayscale deviation of each pixel; the mean of the reference standard grayscale deviation of all pixels in the X-ray grayscale image is taken as the discrete degree of grayscale distribution of the X-ray grayscale image.

[0052] For X-ray grayscale images, if the overall grayscale values ​​are concentrated in a relatively narrow grayscale range, it means that the grayscale values ​​of each pixel are generally closer to the grayscale mean; therefore, the smaller the reference standard grayscale deviation of all pixels in the X-ray grayscale image, the more concentrated the overall grayscale value of the X-ray grayscale image, that is, the lower the overall contrast of the X-ray grayscale image, the more difficult it is to distinguish the details between different tissue structures, and the greater the corresponding contrast enhancement demand.

[0053] In the X-ray grayscale image, the pixel points whose grayscale value is greater than the preset high grayscale threshold or whose grayscale value is less than the preset low grayscale threshold are regarded as brightness detail pixel points; the number of brightness detail pixel points in the X-ray grayscale image is counted. In a specific implementation of an embodiment of the present invention, the preset high grayscale threshold is set to 200, and the preset low grayscale threshold is set to 50, which can be adjusted according to the specific implementation environment. It should be noted that the grayscale value range of the embodiment of the present invention is 0-255. When setting the preset high grayscale threshold and the preset low grayscale threshold, it is necessary to set them specifically in combination with the grayscale value range. Brightness detail pixel points are pixel points with higher or lower grayscale values, which conform to the brightness characteristics of the bright or dark parts in the image; because the lack of dark or bright details in the image will lead to insufficient contrast of the joint anatomical structure, the fewer the number of brightness detail pixel points, the greater the demand for contrast enhancement of the X-ray grayscale image.

[0054] Further combining the relationship between the discrete degree of grayscale distribution and the number of brightness detail pixels and the contrast enhancement requirement, the grayscale distribution effectiveness of the X-ray grayscale image is determined according to the discrete degree of grayscale distribution and the number of brightness detail pixels; when the grayscale distribution effectiveness is greater, the corresponding contrast in the dimension of grayscale distribution is more effective, and the corresponding contrast requirement is smaller; therefore, the discrete degree of grayscale distribution and the number of brightness detail pixels are positively correlated with the grayscale distribution effectiveness.

[0055] Preferably, in some possible implementations of the embodiments of the present invention, the process of determining the grayscale distribution validity of the X-ray grayscale image according to the grayscale distribution discreteness and the number of brightness detail pixels includes:

[0056] The product between the discrete degree of grayscale distribution and the number of brightness detail pixels is normalized to determine the grayscale distribution validity of the X-ray grayscale image. It should be noted that in addition to the product, the implementer can also calculate the grayscale distribution validity by other methods according to the correlation, such as mean, sum, etc., which will not be further elaborated here. The purpose of normalization is to suppress the influence of different dimensions on subsequent calculations.

[0057] In a specific implementation of the embodiment of the present invention, the process of obtaining the grayscale distribution validity is expressed by the formula: ;in, The grayscale distribution validity of X-ray grayscale image; is the number of brightness detail pixels in the X-ray grayscale image; is the number of pixels in the X-ray grayscale image; is the first The gray value of each pixel; is the mean gray value of all pixels in the X-ray grayscale image, that is, the reference grayscale mean; is the first The reference standard grayscale deviation of each pixel; is the discrete degree of grayscale distribution of X-ray grayscale image; It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment.

[0058] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the exposure interference degree includes:

[0059] The average value between the maximum grayscale value and the minimum grayscale value of all pixels in the X-ray grayscale image is taken as the reference grayscale middle value; the difference between the grayscale value of each pixel in the X-ray grayscale image and the reference grayscale middle value is taken as the exposure-affected grayscale deviation of each pixel; the average value of the exposure-affected grayscale deviation of all pixels is normalized to determine the exposure interference degree of the X-ray grayscale image.

[0060] When underexposure or overexposure occurs, a large number of pixels with extreme grayscale values ​​will be formed, and the grayscale values ​​of these pixels are close to the maximum grayscale value or the minimum grayscale value; therefore, for X-ray grayscale images, taking the reference grayscale middle value as the standard, the greater the overall deviation between the grayscale value of each pixel and the reference grayscale middle value, the more pixels with extreme grayscale values ​​there are, and the greater the impact or interference caused by underexposure or overexposure; that is, when the overall exposure impact grayscale deviation of all pixels is greater, the corresponding exposure interference is greater, which means that the credibility of the effectiveness of the corresponding grayscale distribution is lower.

[0061] In a specific implementation of the embodiment of the present invention, the process of obtaining the exposure interference degree is expressed by the formula: ;in, is the exposure interference of the X-ray grayscale image; is the number of pixels in the X-ray grayscale image; is the first The gray value of each pixel; is the maximum gray value of all pixels in the X-ray gray image; It is the minimum gray value of all pixels in the X-ray gray image; is the reference grayscale median value of the X-ray grayscale image; is the first The exposure of each pixel affects the grayscale deviation; is a linear normalization function, and the implementer can adjust the normalization method according to the specific implementation environment. It should be noted that the exposure in the embodiment of the present invention does not affect all pixels and cause all pixels to be at a higher or lower grayscale value, so there is no need to consider such extreme cases.

[0062] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the quality credibility includes:

[0063] Normalization is performed based on the product of the negative correlation mapping value of the exposure interference and the grayscale distribution validity to determine the quality credibility of the X-ray grayscale image. Since the greater the exposure interference, the lower the credibility of the corresponding grayscale distribution validity, the grayscale distribution validity is weighted after the exposure interference negative correlation mapping to determine the grayscale distribution validity after weighted credibility, that is, the quality credibility required by the embodiment of the present invention; since the greater the grayscale distribution validity, the more effective the corresponding contrast in the grayscale distribution dimension, and the smaller the corresponding contrast requirement; therefore, the greater the quality credibility, the smaller the corresponding contrast requirement.

[0064] In a specific implementation of the embodiment of the present invention, the process of obtaining the quality credibility is expressed by the formula: ;in, is the quality reliability of the X-ray grayscale image; is the exposure interference of the X-ray grayscale image; The grayscale distribution validity of X-ray grayscale image; It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment. is an exponential function with a natural constant as the base. Implementers can adopt other negative correlation mapping methods according to the specific implementation environment, such as , and the reciprocal, where is a hyperbolic tangent function, which will not be further described here.

[0065] Step S103: In the X-ray grayscale image, the joint area contrast of the joint area to be analyzed is determined according to the grayscale distribution disorder in the joint area to be analyzed and the number of edge lines after edge detection; the edge clarity of the joint area to be analyzed is determined according to the overall deviation of the grayscale texture between the edge lines in the joint area to be analyzed and the joint area contrast.

[0066] Then, the joint area to be analyzed in the X-ray grayscale image is analyzed. First, according to the characteristics that the light-dark contrast represented by the concentrated grayscale distribution in the joint area to be analyzed is usually not obvious, and the characteristics that the fewer edge lines in the joint area to be analyzed usually indicate that the boundaries between different structures and regions in the image are not clear, the joint area contrast of the joint area to be analyzed is determined according to the chaotic grayscale distribution in the joint area to be analyzed and the number of edge lines after edge detection, so that the lower the joint area contrast, the greater the need for contrast enhancement; and when the overall deviation of the grayscale texture between different edge lines is greater, it can be explained that the distinction of the edge lines in the corresponding joint area to be analyzed is clearer, and the feature representation of the details in the key area is more obvious, then the damaged area or damaged state can be observed more clearly in the joint area to be analyzed, and the corresponding contrast requirement will be smaller; therefore, on the basis of the overall deviation of the grayscale texture between the edge lines in the joint area to be analyzed and combined with the contrast of the joint area, the edge clarity representing the need for contrast enhancement is comprehensively determined.

[0067] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the contrast of the joint area includes:

[0068] The product between the grayscale value variance of all pixels in the joint area to be analyzed and the number of edge lines is used as the joint area contrast of the joint area to be analyzed. Since it is necessary to determine the joint area contrast of the joint area to be analyzed based on the grayscale distribution disorder in the joint area to be analyzed and the number of edge lines after edge detection; therefore, according to the characteristic that the variance can characterize the degree of discreteness of a set of data, the grayscale value variance of all pixels in the joint area to be analyzed is used to characterize the corresponding grayscale distribution disorder. In other possible implementation methods of the embodiments of the present invention, the grayscale value numerical entropy of all pixels in the joint area to be analyzed can also be used to replace the grayscale value variance to characterize the grayscale distribution disorder, and the numerical entropy and variance are both technical means well known to those skilled in the art, and will not be further elaborated here.

[0069] In a specific implementation of the embodiment of the present invention, after performing canny edge detection on all joint areas to be analyzed, the number of corresponding edge lines is determined. Among them, canny edge detection is a technical means well known to those skilled in the art, and the implementer can use other methods other than canny edge detection for edge detection according to the specific implementation environment, which will not be further described here.

[0070] In a specific implementation of the embodiment of the present invention, the process of obtaining the contrast of the joint area is expressed by the formula: ;in, The joint area contrast is the joint area to be analyzed; is the number of edge lines in the joint area to be analyzed; is the gray value variance of all pixels in the joint area to be analyzed.

[0071] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring edge definition includes:

[0072] The mean grayscale value of all edge pixels on each edge line is used as the edge grayscale feature value of each edge line; each edge line and each other edge are selected as an edge line binary; all edge line binary groups corresponding to all edge lines are counted; the difference between the edge grayscale feature values ​​of the two edge lines in each edge line binary group is used as the edge line local discrimination of each edge line binary group; the mean of the edge line local discrimination of all edge line binary groups is used as the edge line overall discrimination of the joint area to be analyzed. Since the greater the overall deviation of the grayscale texture between different edge lines, the clearer the distinction of the corresponding edge lines in the joint area to be analyzed; therefore, the greater the difference in the grayscale value mean of the two edge lines of the same edge line binary group, the greater the grayscale feature deviation of the two edge lines, and the greater the corresponding grayscale texture deviation; therefore, the greater the overall local discrimination of the edge line obtained according to the difference between the edge grayscale feature values ​​of the two edge lines in each edge line binary group, the clearer the distinction of the corresponding edge line, that is, the greater the overall discrimination of the edge line.

[0073] The greater the overall discrimination of the edge line, the clearer the discrimination of the edge line in the joint area to be analyzed; the greater the contrast of the joint area, the clearer the boundary between different structures and areas in the image; therefore, the overall discrimination of the edge line and the contrast of the joint area can jointly characterize the edge clarity of the corresponding joint area to be analyzed; therefore, the edge clarity of the joint area to be analyzed is further determined according to the overall discrimination of the edge line and the contrast of the joint area; the overall discrimination of the edge line and the contrast of the joint area are both positively correlated with the edge clarity. Since the greater the overall discrimination of the edge line, the clearer the discrimination of the edge line in the corresponding joint area to be analyzed, the more obvious the feature representation of the details in the key area, then the more clearly the damaged area or damaged state can be observed in the joint area to be analyzed, and the corresponding contrast requirement will be smaller; and the lower the contrast of the joint area of ​​the joint area to be analyzed, the greater the need for contrast enhancement. Therefore, the greater the edge clarity, the smaller the need for contrast enhancement.

[0074] Preferably, in some possible implementations of the embodiments of the present invention, the process of determining the edge clarity of the joint area to be analyzed according to the overall distinction of the edge line and the contrast of the joint area includes:

[0075] Since the overall discrimination of the edge line and the contrast of the joint area are both positively correlated with the edge clarity, the product of the overall discrimination of the edge line and the contrast of the joint area is further normalized to determine the edge clarity of the joint area to be analyzed. By normalization, the edge clarity value range is limited to 0 to 1, reducing the impact of different dimensions on subsequent calculations. In a specific implementation of an embodiment of the present invention, the process of obtaining edge clarity is expressed by the formula: ;in, is the edge clarity of the joint area to be analyzed; The joint area contrast is the joint area to be analyzed; is the number of edge line pairs in the joint area to be analyzed; is the first joint area to be analyzed The local discrimination of the edge line of the edge line binary group, that is, the first The difference between the edge grayscale feature values ​​of two edge lines in an edge line binary group; is the overall discrimination degree of the edge line of the joint area to be analyzed; It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment.

[0076] Step S104: determining a contrast enhancement coefficient of the X-ray grayscale image according to the quality credibility and the edge clarity; performing image enhancement on the X-ray grayscale image according to the contrast enhancement coefficient in combination with histogram equalization to determine an X-ray enhanced image.

[0077] Both quality credibility and edge clarity can characterize the contrast enhancement requirements to a certain extent. Therefore, a more accurate contrast enhancement coefficient is determined based on the quality credibility and edge clarity. Then, the X-ray grayscale image is adaptively enhanced based on the contrast enhancement coefficient combined with histogram equalization, making the obtained X-ray enhanced image more accurate and improving the weight-bearing X-ray image processing effect for Lisfranc joint injuries.

[0078] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the contrast enhancement coefficient includes:

[0079] According to step S102 and step S103, the smaller the quality confidence and the smaller the edge definition, the greater the need for corresponding contrast enhancement; therefore, after the need for contrast enhancement is characterized by the contrast enhancement coefficient, the contrast enhancement coefficient of the X-ray grayscale image is determined according to the average value between the quality confidence and the edge definition. In a specific implementation of the embodiment of the present invention, the process of obtaining the contrast enhancement coefficient is expressed by the formula: ;in, is the contrast enhancement factor of the X-ray grayscale image; is the quality reliability of the X-ray grayscale image; It is the edge clarity of the joint area to be analyzed; it should be noted that, in addition to the mean, the implementer can also calculate the contrast enhancement coefficient through other methods according to the correlation, such as the normalized value of the product or sum, etc., which will not be further elaborated here.

[0080] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the X-ray enhanced image includes:

[0081] The positive correlation mapping value of the contrast enhancement coefficient is used as the cliplimit parameter in the contrast-limited adaptive histogram equalization algorithm; the X-ray grayscale image is enhanced according to the contrast-limited adaptive histogram equalization algorithm combined with the cliplimit parameter to determine the X-ray enhanced image. Since the quality confidence and edge clarity are both normalized values, the value range of the quality confidence is also 0 to 1; however, the empirical value of the cliplimit parameter in the contrast-limited adaptive histogram equalization algorithm is usually not within the range of 0 to 1, so it is necessary to determine the cliplimit parameter after positive correlation mapping of the contrast enhancement coefficient, so as to optimize the image enhancement process of the histogram equalization.

[0082] In a specific implementation of the embodiment of the present invention, the process of obtaining the cliplimit parameter is expressed by the formula: ;in, It is the cliplimit parameter; is the contrast enhancement factor of the X-ray grayscale image; To preset positive correlation mapping parameters, in a specific implementation of the embodiment of the present invention, the positive correlation mapping parameters are preset It is set to 7 and can be adjusted according to the specific implementation environment. After determining the cliplimit parameter, the X-ray grayscale image is further enhanced in combination with the contrast-limited adaptive histogram equalization algorithm to determine the X-ray enhanced image required by this application. It should be noted that the contrast-limited adaptive histogram equalization algorithm is a histogram equalization method well known to those skilled in the art, and the cliplimit parameter is a parameter in the contrast-limited adaptive histogram equalization algorithm, which will not be further defined or elaborated here. After determining the X-ray enhanced image with better image enhancement effect, since the X-ray enhanced image has a clearer display of the Lisfranc joint, it can be used in medical teaching or neural network training and other processes.

[0083] In summary, the weight-bearing X-ray image processing method for Lisfranc joint injury first determines the grayscale distribution validity according to the characteristics that the tissue structure distinction is poor when the grayscale value is relatively concentrated and the lack of brightness details will lead to insufficient contrast of the joint anatomical structure; then determines the exposure interference according to the grayscale characteristics of underexposure or overexposure; thus, the quality credibility that reflects the contrast requirement to a certain extent is determined by combining the grayscale distribution validity and the exposure interference; then, the joint area contrast in the initial situation is determined according to the grayscale distribution chaos in the joint area to be analyzed and the corresponding number of edge lines, and on this basis, combined with the characteristics that different edges in the joint area to be analyzed are easy to distinguish when the contrast requirement is low, determines the contrast enhancement. The invention can determine the edge clarity required according to the quality credibility and the edge clarity, thereby determining a more accurate contrast enhancement coefficient according to the contrast enhancement coefficient in combination with histogram equalization to adaptively enhance the X-ray grayscale image, thereby solving the problem that the prior art uses a fixed cliplimit parameter in combination with a contrast-limited adaptive histogram equalization algorithm to enhance different X-ray images without considering the contrast enhancement requirement of the X-ray image under the influence of the external environment, resulting in low image enhancement accuracy of the X-ray image of the Lisfranc joint, making the obtained X-ray enhanced image more accurate and improving the weight-bearing X-ray image processing effect for Lisfranc joint injuries.

[0084] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A weight-bearing X-ray image processing method for Lisfranc joint injury, characterized in that: The method comprises: Acquire an X-ray grayscale image of the Lisfranc joint of the patient under weight-bearing condition; obtain a joint area to be analyzed in the X-ray grayscale image; Determine the corresponding grayscale distribution validity according to the grayscale value distribution concentration and the lack of brightness details in the X-ray grayscale image; determine the corresponding exposure interference according to the grayscale distribution extreme situation in the X-ray grayscale image; determine the quality credibility of the X-ray grayscale image according to the grayscale distribution validity and the exposure interference; In the X-ray grayscale image, the joint area contrast of the joint area to be analyzed is determined according to the grayscale distribution disorder in the joint area to be analyzed and the number of edge lines after edge detection; the edge clarity of the joint area to be analyzed is determined according to the overall deviation of the grayscale texture between the edge lines in the joint area to be analyzed and the joint area contrast; Determining a contrast enhancement coefficient of the X-ray grayscale image according to the quality credibility and the edge clarity; performing image enhancement on the X-ray grayscale image according to the contrast enhancement coefficient in combination with histogram equalization to determine an X-ray enhanced image; The process of acquiring the X-ray enhanced image includes: Using the positive correlation mapping value of the contrast enhancement coefficient as a cliplimit parameter in a contrast-limited adaptive histogram equalization algorithm; performing image enhancement on the X-ray grayscale image according to the contrast-limited adaptive histogram equalization algorithm in combination with the cliplimit parameter to determine an X-ray enhanced image; The process of obtaining the grayscale distribution validity includes: The mean grayscale value of all pixels in the X-ray grayscale image is used as the reference grayscale mean; the difference between the grayscale value of each pixel in the X-ray grayscale image and the reference grayscale mean is used as the reference standard grayscale deviation of each pixel; the mean of the reference standard grayscale deviations of all pixels in the X-ray grayscale image is used as the grayscale distribution discreteness of the X-ray grayscale image; In the X-ray grayscale image, taking pixels whose grayscale values ​​are greater than a preset upper grayscale threshold or whose grayscale values ​​are less than a preset lower grayscale threshold as brightness detail pixels; and counting the number of brightness detail pixels in the X-ray grayscale image; Determining the grayscale distribution validity of the X-ray grayscale image according to the grayscale distribution discreteness and the number of brightness detail pixels; the grayscale distribution discreteness and the number of brightness detail pixels are both positively correlated with the grayscale distribution validity; The exposure interference degree acquisition process includes: The average value between the maximum grayscale value and the minimum grayscale value of all pixels in the X-ray grayscale image is used as the reference grayscale intermediate value; the difference between the grayscale value of each pixel in the X-ray grayscale image and the reference grayscale intermediate value is used as the exposure-affected grayscale deviation of each pixel; the average value of the exposure-affected grayscale deviation of all pixels is normalized to determine the exposure interference degree of the X-ray grayscale image; The process of obtaining the edge definition includes: The average grayscale value of all edge pixels on each edge line is taken as the edge grayscale feature value of each edge line; Select each edge line and each other edge line in turn as an edge line binary group; count all edge line binary groups corresponding to all edge lines; The difference between the edge grayscale feature values ​​of the two edge lines in each edge line binary group is used as the edge line local discrimination degree of each edge line binary group; the average of the edge line local discrimination degrees of all edge line binary groups is used as the edge line overall discrimination degree of the joint area to be analyzed; The edge clarity of the joint area to be analyzed is determined according to the overall discrimination degree of the edge line and the contrast of the joint area; both the overall discrimination degree of the edge line and the contrast of the joint area are positively correlated with the edge clarity.

2. A weight-bearing X-ray image processing method for Lisfranc joint injury according to claim 1, characterized in that: The process of determining the grayscale distribution validity of the X-ray grayscale image according to the grayscale distribution discreteness and the number of brightness detail pixels includes: The product of the grayscale distribution discreteness and the number of brightness detail pixels is normalized to determine the grayscale distribution validity of the X-ray grayscale image.

3. A weight-bearing X-ray image processing method for Lisfranc joint injury according to claim 1, characterized in that: The process of obtaining the quality credibility includes: Normalization is performed according to the product between the negative correlation mapping value of the exposure interference and the grayscale distribution validity to determine the quality credibility of the X-ray grayscale image.

4. The method for processing weight-bearing X-ray images of Lisfranc joint injuries according to claim 1, characterized in that: The process of acquiring the contrast of the joint area includes: The product of the gray value variance of all pixels in the joint area to be analyzed and the number of edge lines is used as the joint area contrast of the joint area to be analyzed.

5. The method for processing weight-bearing X-ray images of Lisfranc joint injuries according to claim 1, characterized in that: The process of determining the edge clarity of the joint area to be analyzed according to the overall distinction of the edge line and the contrast of the joint area includes: The product of the overall discrimination degree of the edge line and the contrast of the joint area is normalized to determine the edge clarity of the joint area to be analyzed.

6. The method for processing weight-bearing X-ray images of Lisfranc joint injuries according to claim 1, characterized in that: The process of obtaining the contrast enhancement coefficient includes: A contrast enhancement coefficient of the X-ray grayscale image is determined according to a mean value between the quality confidence and the edge definition.

Images