Method and device for processing metacarpal and phalangeal bone image

Abstract

The invention relates to the technical field of image processing, in particular to a metacarpal and phalanx image processing method and device. The method comprises the following steps: acquiring an initial metacarpal and phalanx image, and performing brightness processing on the initial metacarpal and phalanx image to obtain a first metacarpal and phalanx image; performing contrast processing on the first metacarpal and phalanx image to obtain a second metacarpal and phalanx image; constructing a preset convolution kernel, and performing definition processing on the second metacarpal and phalanx image based on the preset convolution kernel to obtain a metacarpal and phalanx target image; and inputting the metacarpal and phalanx target image into the bone age recognition model, and outputting a bone age recognition result. According to the method, dynamic image enhancement optimization can be carried out on metacarpal and phalanx X-ray images of different devices and different qualities through brightness, contrast and definition processing, so that the processed metacarpal and phalanx target images can highlight morphological characteristics of epiphysis, the bone age recognition accuracy is improved, and the bone age recognition efficiency is improved. The problem of low bone age recognition accuracy caused by different image qualities in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and device for processing metacarpal and phalangeal images. Background Art

[0002] Bone age assessment plays a central role in pediatric growth and development monitoring and the diagnosis and treatment of endocrine diseases. It provides a key basis for assessing children's growth and development and diagnosing related diseases. With advancements in medical technology and growing medical needs, the development of bone age assessment technology has become increasingly important.

[0003] Currently, bone age assessment uses traditional methods, such as the Greulich–Pyle (GP) atlas and scoring methods. These methods rely on doctors' personal experience based on metacarpal and phalangeal X-rays, but are time-consuming and labor-intensive. Another approach is automated assessment, represented by artificial intelligence, such as deep learning models, which can speed up bone age assessment.

[0004] However, both of the above-mentioned bone age assessment methods require X-rays of the metacarpals and phalanges to perform bone age assessment. When different equipment is used to take X-rays of the metacarpals and phalanges, the image quality of the metacarpals and phalanges is uneven due to reasons such as exposure degree or shooting angle. For example, the key observation area of ​​the image is not clear and there is severe noise, which makes the subsequent reading of the metacarpals and phalanges X-rays more difficult and makes the bone age assessment results inaccurate. Summary of the Invention

[0005] The embodiments of the present invention provide a method and apparatus for processing metacarpal and phalangeal images to solve the problems in the prior art of difficulty in reading bone age assessment images and low accuracy of assessment results.

[0006] In a first aspect, an embodiment of the present invention provides a method for processing metacarpal and phalangeal images, comprising:

[0007] Obtain initial images of the metacarpal bones;

[0008] Performing brightness processing on the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image;

[0009] performing contrast processing on the first metacarpal and phalangeal bone image to obtain a second metacarpal and phalangeal bone image;

[0010] Constructing a preset convolution kernel, and performing clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image;

[0011] The metacarpal and phalangeal target images are input into the bone age recognition model, and the bone age recognition results are output.

[0012] In a possible implementation, performing brightness processing on the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image includes:

[0013] Converting the initial metacarpal and phalangeal image from RGB image to HSV space to obtain brightness component;

[0014] Calculating a global brightness mean according to the brightness component;

[0015] Determine the Gamma value according to the global brightness mean;

[0016] The gamma value is used to adjust the brightness of the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image.

[0017] In a possible implementation, determining a gamma value according to the global brightness mean includes:

[0018] If the global brightness mean is greater than or equal to the first brightness value and less than or equal to the second brightness value, determine the Gamma value to be a first preset value;

[0019] If the global brightness mean is less than the first brightness value, the Gamma value is determined according to the first formula, wherein the first formula is γ=0.5+0.02×(30-B avg ), where γ represents the determined Gamma value, B avg represents the global brightness mean;

[0020] If the global brightness mean is greater than the second brightness value, the Gamma value is determined according to the second formula, wherein the second formula is γ=1.5+0.03×(B avg -180).

[0021] In a possible implementation, performing contrast processing on the first metacarpal and phalangeal bone image to obtain a second metacarpal and phalangeal bone image includes:

[0022] Performing detail enhancement on different regions of the first metacarpal and phalangeal bone images, extracting epiphyseal regions from the processed images, and calculating epiphyseal area ratios based on the epiphyseal regions;

[0023] Determining a contrast limitation parameter of the real-time first metacarpal and phalanx image according to the epiphyseal area ratio;

[0024] The first metacarpal and phalangeal bone image is divided into a plurality of small blocks, a local histogram is calculated based on each small block, and the local histogram of the corresponding small block is cropped according to the contrast limit parameter to obtain a second metacarpal and phalangeal bone image.

[0025] In a possible implementation, determining a contrast limitation parameter according to the epiphyseal area ratio includes:

[0026] According to ClipLimit=2.0+0.1×(50-Area%), the contrast limit parameter is determined;

[0027] Wherein, ClipLimit represents the determined contrast limit parameter, and Area% represents the epiphyseal area ratio.

[0028] In a possible implementation, constructing a preset convolution kernel includes:

[0029] The Laplace kernel is processed, and a preset convolution kernel is constructed based on the processed Laplace kernel and the original image.

[0030] In a possible implementation, performing clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image includes:

[0031] Using the preset convolution kernel to perform clarity processing on each pixel in the second metacarpal and phalangeal bone image to determine a preliminary sharpened image;

[0032] The gradient intensity of each pixel in the preliminary sharpened image is calculated, and the gradient intensity greater than the preset gradient intensity is deleted, and the metacarpal and phalangeal target images are obtained according to the pixels corresponding to the remaining gradient intensities and the corresponding pixels in the preliminary sharpened image.

[0033] In a second aspect, an embodiment of the present invention provides a metacarpal and phalangeal image processing device, comprising:

[0034] An acquisition module, used for acquiring initial images of the metacarpal bones and phalanges;

[0035] a processing module, configured to perform brightness processing on the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image;

[0036] The processing module is further configured to perform contrast processing on the first metacarpal and phalangeal bone image to obtain a second metacarpal and phalangeal bone image;

[0037] The processing module is further configured to construct a preset convolution kernel, and perform clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image;

[0038] The bone age prediction module is used to input the metacarpal and phalangeal target images into the bone age recognition model and output the bone age recognition results.

[0039] In a third aspect, an embodiment of the present invention provides a terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the metacarpal and phalangeal image processing method described in the first aspect or any possible implementation of the first aspect are implemented.

[0040] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the metacarpal and phalangeal image processing method as described in the first aspect or any possible implementation method of the first aspect.

[0041] An embodiment of the present invention provides a method and apparatus for processing metacarpal and phalangeal images. The method comprises obtaining an initial metacarpal and phalangeal image, performing brightness processing on the initial metacarpal and phalangeal image, obtaining a first metacarpal and phalangeal image, performing contrast processing on the first metacarpal and phalangeal image, obtaining a second metacarpal and phalangeal image, constructing a preset convolution kernel, and performing clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image, and inputting the metacarpal and phalangeal target image into a bone age recognition model to obtain a bone age assessment result. In this embodiment, adaptive brightness, contrast, and clarity processing can be used for metacarpal and phalangeal X-ray images of different qualities from different devices, and dynamic image enhancement optimization can be performed, so that the processed metacarpal and phalangeal target image can highlight the morphological features of the epiphysis, provide an accurate data basis for subsequent bone age recognition, and improve the accuracy of metacarpal and phalangeal bone age recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 any creative work.

[0043] Figure 1 is a flowchart of an implementation method for processing metacarpal and phalangeal images provided by an embodiment of the present invention;

[0044] Figure 2 is a schematic diagram of a metacarpal and phalangeal image processing method provided by an embodiment of the present invention;

[0045] Figure 3 This is a flowchart of a method for determining a gamma value based on a global brightness mean value provided by an embodiment of the present invention;

[0046] Figure 4 is a structural diagram of a metacarpal and phalangeal image processing device provided by an embodiment of the present invention;

[0047] Figure 5 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In the following description, specific details such as particular system structures and techniques are provided for purposes of illustration, not limitation, to facilitate a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below with reference to the accompanying drawings.

[0050] In the prior art, bone age assessment technology mainly adopts the Greulich–Pyle (GP) atlas method, scoring method, etc. These methods are mainly based on X-rays of the metacarpal and phalangeal bones, and use manual or deep learning models to perform bone age assessment. Whether it is manual or artificial intelligence, it is necessary to rely on clear images to achieve accurate bone age assessment. However, in the prior art, the X-rays of the metacarpal and phalangeal bones taken by different equipment have different clarity and angles, which makes the key observation position of the metacarpal and phalangeal bones unclear and the image noise is large, resulting in inaccurate results of subsequent bone age assessment. In order to solve the above problems, an embodiment of the present invention provides a metacarpal and phalangeal image processing method, Figure 1 The following is a flowchart of the implementation of the metacarpal and phalangeal image processing method, which is detailed as follows:

[0051] Step 101: Acquire initial images of metacarpal bones and phalanges.

[0052] Initial metacarpal and phalangeal images can be X-rays of the palm taken with different equipment. Due to the different equipment and shooting angles, the quality of the X-rays varies. Therefore, preprocessing of these heterogeneous and non-uniform metacarpal and phalangeal X-rays is necessary to ensure accurate bone age recognition results using the bone age recognition model.

[0053] Step 102: Perform brightness processing on the initial metacarpal and phalangeal bone image to obtain a first metacarpal and phalangeal bone image.

[0054] In this embodiment, see Figure 2 As shown, adaptive dynamic image enhancement is performed on the initial metacarpal and phalangeal images to improve image clarity and highlight epiphyseal features. The processing flow is a three-level cascade adjustment of brightness, contrast, and clarity to achieve cross-device X-ray enhancement. Among them, step 102 is the brightness processing of the initial metacarpal and phalangeal images.

[0055] In one embodiment, brightness processing is performed on the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image, which may include: converting the initial metacarpal and phalangeal image from a red, green, blue (RGB) image to a hue, saturation, value (HSV) space to obtain a brightness component; calculating a global brightness mean based on the brightness component; determining a gamma value (γ value) based on the global brightness mean; and using the gamma value to adjust the brightness of the initial metacarpal and phalangeal image to obtain the first metacarpal and phalangeal image.

[0056] By converting the initial metacarpal and phalangeal bone images from RGB to HSV space, we can obtain the hue, saturation, and brightness components. In this embodiment, we extract the V channel histogram, i.e., the brightness component. Then, we calculate the global brightness mean and automatically adjust the gamma value based on the brightness distribution.

[0057] In the fields of image processing and computer vision, the gamma value is an important parameter used to describe the nonlinear transformation relationship between image brightness and pixel value. Its core function is to correct the brightness and contrast of the image to make it more consistent with the visual characteristics of the human eye or the requirements of specific display devices.

[0058] In one embodiment, see Figure 3 As shown, the Gamma value is determined according to the global brightness mean, which may include:

[0059] If the global brightness mean is greater than or equal to the first brightness value and less than or equal to the second brightness value, determine the Gamma value to be the first preset value;

[0060] If the global brightness mean is less than the first brightness value, the Gamma value is determined according to the first formula, wherein the first formula is γ=0.5+0.02×(30-B avg ), where γ represents the determined Gamma value, B avg represents the global brightness mean;

[0061] If the global brightness mean is greater than the second brightness value, the Gamma value is determined according to the second formula, wherein the second formula is γ=1.5+0.03×(B avg -180).

[0062] After the Gamma value is determined by the above calculation method, the Gamma value is used to perform brightness correction on the initial metacarpal and phalangeal images, so that the brightness standard deviation of the initial metacarpal and phalangeal images becomes smaller, thereby solving the problem of image overexposure or underexposure.

[0063] Understandably, Figure 2As shown, after obtaining the global brightness mean, it is detected whether the global brightness mean is greater than or equal to the first brightness value and less than or equal to the second brightness value. If the global brightness mean is greater than or equal to the first brightness value and less than or equal to the second brightness value, the gamma value can be directly determined, for example, the gamma value can be 1.0. If not, it is detected whether the global brightness mean is less than the first brightness value. If so, the gamma value is determined according to the first formula. If not, it is necessary to detect whether the global brightness mean is greater than the second brightness value. If so, the gamma value is determined according to the second formula.

[0064] It should be noted that the first brightness value is the brightness threshold for underexposure, and the second brightness value is the brightness threshold for overexposure. In this embodiment, the values ​​of the first brightness value and the second brightness value are not limited. Optionally, the first brightness value can be set to 30 and the second brightness value can be set to 180.

[0065] Step 103: Perform contrast processing on the first metacarpal and phalangeal bone image to obtain a second metacarpal and phalangeal bone image.

[0066] In this embodiment, contrast limited adaptive histogram equalization (CLAHE) is performed on the first metacarpal and phalangeal image after brightness adjustment to enhance details in different areas of the first metacarpal and phalangeal image, for example, the texture of the dark area and the edge of the bright area are enhanced. Then, the contrast limit parameter is adjusted based on the area percentage of the epiphyseal area (Area%). After this correction, image details are enhanced, noise control is balanced, and a natural visual effect is improved.

[0067] In one embodiment, performing contrast processing on the first metacarpal and phalangeal bone image to obtain the second metacarpal and phalangeal bone image may include:

[0068] Perform detail enhancement on different regions of the first metacarpal phalanx image, extract the epiphyseal region from the processed image, and calculate the epiphyseal area ratio based on the epiphyseal region;

[0069] Determine the contrast limit parameter of the real-time first metacarpal phalanx image according to the epiphyseal area ratio;

[0070] The first metacarpal and phalangeal bone image is divided into multiple small blocks, a local histogram is calculated based on each small block, and the local histogram of the corresponding small block is cropped according to a contrast limit parameter to obtain a second metacarpal and phalangeal bone image.

[0071] Optionally, when performing detail enhancement processing on different areas in the first metacarpal and phalangeal image, the first metacarpal and phalangeal image can be divided into multiple overlapping or non-overlapping small blocks, for example, common block sizes are 8×8 pixels or 16×16 pixels; a histogram is calculated independently for each small block, and equalization is performed to enhance local contrast.

[0072] The contrast limit parameter is used to limit the proportion of pixels at each grayscale level in the histogram. If the number of pixels at a grayscale level exceeds ClipLimit, the excess will be "clipped" and evenly distributed among the other grayscale levels. This prevents local noise from being amplified by excessive contrast enhancement. For example, random noise points in dark areas may be overbrightened in HE, forming noise points. In this solution, the contrast limit parameter is used to limit excessive brightening.

[0073] Finally, bilinear interpolation is performed on each processed small block to smooth the boundaries between blocks and avoid the "chessboard effect" caused by block processing. Ultimately, a globally enhanced and noise-balanced image is generated, which improves the natural visual effect.

[0074] In one embodiment, determining the contrast limitation parameter according to the epiphyseal area ratio may include:

[0075] According to ClipLimit=2.0+0.1×(50-Area%), the contrast limit parameter is determined;

[0076] Wherein, ClipLimit represents the determined contrast limit parameter, and Area% represents the epiphyseal area ratio.

[0077] Step 104: construct a preset convolution kernel, and perform clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image.

[0078] During the clarity processing, the epiphyseal edges in the metacarpal and phalangeal X-rays are further enhanced to highlight the epiphyseal features, so as to better identify the epiphyseal features and assist in the bone age diagnosis of the metacarpal and phalangeal epiphysis. (This part can be placed in the background)

[0079] Clarity processing is the process of sharpening.

[0080] In this embodiment, constructing the preset convolution kernel may include: processing the Laplace kernel, and constructing the preset convolution kernel according to the processed Laplace kernel and the original image.

[0081] The original image is a 3*3 convolution kernel with a center of 1 and surrounding 0s, that is,

[0082] There are usually two forms of Laplacian kernel representation: one is a 3*3 convolution kernel with a center of -4 and a surrounding of 1, and the second is a 3*3 convolution kernel with a center of -8 and a surrounding of 1. In this embodiment, the second form of the Laplacian kernel is negated as a whole to obtain the processed Laplacian kernel, that is, the processed Laplacian kernel

[0083] Preset convolution kernel The Laplacian operator is essentially a second-order differential operator, used in image processing to measure the difference between a point and its surrounding points. If the values ​​of the pixels surrounding a pixel differ significantly from its surroundings (such as at an edge), the result of the convolution will be a large positive or negative number. If the values ​​of the surrounding pixels are similar (such as in a smooth area), the result of the convolution will be close to zero. Therefore, after using the Laplacian kernel, the edges of the image will be enhanced, while the smooth areas will be suppressed. Adding a certain multiple of the Laplacian kernel to the original image can enhance the edge details of the image, making the image appear clearer.

[0084] The preset convolution kernel satisfies the brightness conservation constraint, where the brightness conservation constraint is in, Represents the average of the 8 pixels surrounding the current pixel (x,y). This ensures that the overall brightness remains unchanged after sharpening to avoid conflicts with gamma correction. I(x,y) represents the value of the current pixel (x,y).

[0085] In one embodiment, performing clarity processing on the second metacarpal and phalangeal bone image based on a preset convolution kernel to obtain a metacarpal and phalangeal bone target image may include:

[0086] Using a preset convolution kernel to perform clarity processing on each pixel in the second metacarpal and phalangeal bone image to determine a preliminary sharpened image;

[0087] The details of the sharpened image are more obvious, but in order to further improve the image signal-to-noise ratio, increase the gradient intensity of the epiphyseal edge in the metacarpophalangeal bones, make the edges clearer, and make the morphological features more obvious, the high-frequency components of the sharpened image can be suppressed, that is, the gradient intensity of each pixel in the preliminary sharpened image is calculated, and the gradient intensity greater than the preset gradient intensity is deleted. The metacarpophalangeal target image is obtained according to the pixels corresponding to the remaining gradient intensities and the corresponding pixels in the preliminary sharpened image.

[0088] Step 105: input the metacarpal and phalangeal target images into the bone age recognition model and output the bone age recognition results.

[0089] The bone age recognition model can be an EfficientNetB3 classification model or an improved EfficientNetB3.

[0090] The embodiment of the present invention obtains an initial image of the metacarpophalangeal bones, performs brightness processing on the initial image of the metacarpophalangeal bones, and obtains a first metacarpophalangeal bone image; performs contrast processing on the first metacarpophalangeal bone image to obtain a second metacarpophalangeal bone image; constructs a preset convolution kernel, and performs clarity processing on the second metacarpophalangeal bone image based on the preset convolution kernel to obtain a metacarpophalangeal bone target image; inputs the metacarpophalangeal bone target image into a bone age recognition model, and outputs a bone age recognition result. In this embodiment, adaptive brightness, contrast, and clarity processing are used for multi-source heterogeneous and non-uniform quality metacarpophalangeal bone X-ray images, and dynamic image enhancement optimization is performed, so that the processed metacarpophalangeal bone target image can highlight the morphological features of the epiphysis, improve the accuracy of bone age recognition, and solve the problem of low accuracy of bone age recognition due to different image quality in the prior art. Subsequent application to the bone age recognition model for bone age recognition can improve the speed of bone age recognition compared with the manual bone age recognition in the prior art.

[0091] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0092] The following are device embodiments of the present invention. For details not fully described therein, reference may be made to the corresponding method embodiments described above.

[0093] Figure 4 The structure diagram of the metacarpal and phalangeal image processing device provided by an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, which are detailed as follows:

[0094] like Figure 4 As shown, the metacarpal and phalangeal image processing device 4 includes: an acquisition module 41 , a processing module 42 and a bone age prediction module 43 .

[0095] An acquisition module 41 is used to acquire an initial image of the metacarpal bones;

[0096] The processing module 42 is used to perform brightness processing on the initial metacarpal and phalangeal bone image to obtain a first metacarpal and phalangeal bone image;

[0097] The processing module 42 is further configured to perform contrast processing on the first metacarpal and phalangeal bone image to obtain a second metacarpal and phalangeal bone image;

[0098] The processing module 42 is further configured to construct a preset convolution kernel, and perform clarity processing on the second metacarpal and phalangeal bone image based on the preset convolution kernel to obtain a metacarpal and phalangeal bone target image;

[0099] The bone age prediction module 43 is used to input the metacarpal and phalangeal target images into the bone age recognition model and output the bone age recognition results.

[0100] In a possible implementation, the processing module 42 performs brightness processing on the initial metacarpal and phalangeal image to obtain the first metacarpal and phalangeal image, and is used to:

[0101] Convert the initial metacarpal and phalangeal image from RGB image to HSV space to obtain the brightness component;

[0102] According to the brightness component, calculate the global brightness mean;

[0103] Determine the Gamma value based on the global brightness mean;

[0104] The brightness of the initial metacarpal and phalangeal image is adjusted using the Gamma value to obtain the first metacarpal and phalangeal image.

[0105] In a possible implementation, when the processing module 42 determines the gamma value based on the global brightness mean, it is configured to:

[0106] If the global brightness mean is greater than or equal to the first brightness value and less than or equal to the second brightness value, determine the Gamma value to be the first preset value;

[0107] If the global brightness mean is less than the first brightness value, the Gamma value is determined according to the first formula, wherein the first formula is γ=0.5+0.02×(30-B avg ), where γ represents the determined Gamma value, B avg represents the global brightness mean;

[0108] If the global brightness mean is greater than the second brightness value, the Gamma value is determined according to the second formula, wherein the second formula is γ=1.5+0.03×(B avg -180).

[0109] In a possible implementation, when the processing module 42 performs contrast processing on the first metacarpal and phalangeal bone image to obtain the second metacarpal and phalangeal bone image, it is configured to:

[0110] Perform detail enhancement on different regions of the first metacarpal phalanx image, extract the epiphyseal region from the processed image, and calculate the epiphyseal area ratio based on the epiphyseal region;

[0111] Determine the contrast limit parameter of the real-time first metacarpal phalanx image according to the epiphyseal area ratio;

[0112] The first metacarpal and phalangeal bone image is divided into multiple small blocks, a local histogram is calculated based on each small block, and the local histogram of the corresponding small block is cropped according to a contrast limit parameter to obtain a second metacarpal and phalangeal bone image.

[0113] In a possible implementation, when the processing module 42 determines the contrast limitation parameter based on the epiphyseal area ratio, it is configured to:

[0114] According to ClipLimit=2.0+0.1×(50-Area%), the contrast limit parameter is determined;

[0115] Wherein, ClipLimit represents the determined contrast limit parameter, and Area% represents the epiphyseal area ratio.

[0116] In one possible implementation, when the processing module 42 constructs the preset convolution kernel, it is used to:

[0117] The Laplace kernel is processed, and a preset convolution kernel is constructed based on the processed Laplace kernel and the original image.

[0118] In a possible implementation, the processing module 42 performs clarity processing on the second metacarpal and phalangeal bone image based on a preset convolution kernel to obtain a metacarpal and phalangeal bone target image, and is used to:

[0119] Using a preset convolution kernel to perform clarity processing on each pixel in the second metacarpal and phalangeal bone image to determine a preliminary sharpened image;

[0120] The gradient intensity of each pixel in the preliminary sharpened image is calculated, and the gradient intensity greater than the preset gradient intensity is deleted. The metacarpal and phalangeal target images are obtained based on the pixels corresponding to the remaining gradient intensities and the corresponding pixels in the preliminary sharpened image.

[0121] The above-mentioned metacarpophalangeal image processing device obtains the initial metacarpophalangeal image through the acquisition module; the processing module performs brightness processing on the initial metacarpophalangeal image to obtain a first metacarpophalangeal image, then performs contrast processing on the first metacarpophalangeal image to obtain a second metacarpophalangeal image, and constructs a preset convolution kernel. Based on the preset convolution kernel, the second metacarpophalangeal image is processed for clarity to obtain a metacarpophalangeal target image; the bone age prediction module inputs the metacarpophalangeal target image into the bone age recognition model and outputs the bone age recognition result. In this embodiment, adaptive brightness, contrast, and clarity processing are used for metacarpophalangeal X-ray images of different qualities from different devices, and dynamic image enhancement optimization is performed, so that the processed metacarpophalangeal target image can highlight the morphological characteristics of the epiphysis and improve the accuracy of bone age recognition.

[0122] Figure 5 Schematic diagram of a terminal provided by an embodiment of the present invention. Figure 5 As shown, the terminal 5 of this embodiment includes: a processor 50, a memory 51, and a computer program 52 stored in the memory 51 and executable on the processor 50. When the processor 50 executes the computer program 52, the steps in the above-mentioned embodiments of the metacarpal bone image processing method are implemented, for example Figure 1Alternatively, when the processor 50 executes the computer program 52, the functions of the modules / units in the above-mentioned device embodiments are realized, for example, Figure 4 The functions of each module / unit are shown.

[0123] Exemplarily, the computer program 52 may be divided into one or more modules / units, which are stored in the memory 51 and executed by the processor 50 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program 52 in the terminal 5. For example, the computer program 52 may be divided into Figure 4 The modules / units shown.

[0124] The terminal 5 may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art will understand that Figure 5 It is only an example of terminal 5 and does not constitute a limitation on terminal 5. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0125] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0126] The memory 51 can be an internal storage unit of the terminal 5, such as a hard disk or memory of the terminal 5. The memory 51 can also be an external storage device of the terminal 5, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal 5. Furthermore, the memory 51 can also include both an internal storage unit of the terminal 5 and an external storage device. The memory 51 is used to store the computer program and other programs and data required by the terminal. The memory 51 can also be used to temporarily store data that has been output or is about to be output.

[0127] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0128] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0129] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0130] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, and can be electrical, mechanical, or other forms.

[0131] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0132] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0133] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned metacarpal bone image processing method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0134] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A metacarpal bone image processing method, characterized in that: include: Obtain initial images of the metacarpal bones; Performing brightness processing on the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image; performing contrast processing on the first metacarpal and phalangeal bone image to obtain a second metacarpal and phalangeal bone image; Constructing a preset convolution kernel, and performing clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image; The metacarpal and phalangeal target images are input into the bone age recognition model, and the bone age recognition results are output.

2. The method for processing metacarpal and phalangeal images according to claim 1, wherein: Performing brightness processing on the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image includes: Converting the initial metacarpal and phalangeal image from RGB image to HSV space to obtain brightness component; Calculating a global brightness mean according to the brightness component; Determine a Gamma value according to the global brightness mean; The gamma value is used to adjust the brightness of the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image.

3. The method for processing metacarpal and phalangeal images according to claim 2, wherein: Determining a Gamma value according to the global brightness mean includes: If the global brightness mean is greater than or equal to the first brightness value and less than or equal to the second brightness value, determine the Gamma value to be a first preset value; If the global brightness mean is less than the first brightness value, the Gamma value is determined according to the first formula, wherein the first formula is γ=0.5+0.02×(30-B avg ), where γ represents the determined Gamma value, B avg represents the global brightness mean; If the global brightness mean is greater than the second brightness value, the Gamma value is determined according to the second formula, wherein the second formula is γ=1.5+0.03×(B avg -180).

4. The method for processing metacarpal and phalangeal images according to any one of claims 1 to 3, characterized in that: Performing contrast processing on the first metacarpal and phalangeal bone images to obtain second metacarpal and phalangeal bone images includes: Performing detail enhancement on different regions of the first metacarpal and phalangeal bone images, extracting epiphyseal regions from the processed images, and calculating epiphyseal area ratios based on the epiphyseal regions; Determining a contrast limitation parameter of the real-time first metacarpal and phalanx image according to the epiphyseal area ratio; The first metacarpal and phalangeal bone image is divided into a plurality of small blocks, a local histogram is calculated based on each small block, and the local histogram of the corresponding small block is cropped according to the contrast limit parameter to obtain a second metacarpal and phalangeal bone image.

5. The method for processing metacarpal and phalangeal images according to claim 4, characterized in that: Determining the contrast limitation parameter according to the epiphyseal area ratio includes: According to ClipLimit=2.0+0.1×(50-Area%), the contrast limit parameter is determined; Wherein, ClipLimit represents the determined contrast limit parameter, and Area% represents the epiphyseal area ratio.

6. The method for processing metacarpal and phalangeal images according to any one of claims 1 to 3, characterized in that: The step of constructing a preset convolution kernel includes: The Laplace kernel is processed, and a preset convolution kernel is constructed based on the processed Laplace kernel and the original image.

7. The method for processing metacarpal and phalangeal images according to claim 6, characterized in that: Performing clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image includes: Using the preset convolution kernel to perform clarity processing on each pixel in the second metacarpal and phalangeal bone image to determine a preliminary sharpened image; The gradient intensity of each pixel in the preliminary sharpened image is calculated, and the gradient intensity greater than the preset gradient intensity is deleted, and the metacarpal and phalangeal target images are obtained according to the pixels corresponding to the remaining gradient intensities and the corresponding pixels in the preliminary sharpened image.

8. A metacarpal bone image processing device, characterized in that: include: An acquisition module, used for acquiring initial images of the metacarpal bones and phalanges; a processing module, configured to perform brightness processing on the initial metacarpal and phalangeal image to obtain a first metacarpal and phalangeal image; The processing module is further configured to perform contrast processing on the first metacarpal and phalangeal bone image to obtain a second metacarpal and phalangeal bone image; The processing module is further configured to construct a preset convolution kernel, and perform clarity processing on the second metacarpal and phalangeal image based on the preset convolution kernel to obtain a metacarpal and phalangeal target image; The bone age prediction module is used to input the metacarpal and phalangeal target images into the bone age recognition model and output the bone age recognition results.

9. A terminal comprising a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory, characterized in that: When the processor executes the computer program, the steps of the metacarpal and phalangeal image processing method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the metacarpal and phalangeal image processing method according to any one of claims 1 to 7 are implemented.

Images