Large animal X-ray image enhancement method and enhancement device, storage medium and terminal

Abstract

The invention provides a large animal X-ray image enhancement method and device, a storage medium and a terminal, and the method comprises the steps: carrying out the preprocessing of an original X-ray image, and obtaining a preprocessing image with the enhanced target details; scattering quantity in the preprocessed image is removed, and an effective image is obtained; performing multi-scale enhancement on the effective image to obtain a multi-scale enhanced image; and carrying out sharpening processing on the multi-scale enhanced image. According to the method, vertebra details and edge contours in the x-ray image of the large animal can be effectively highlighted and enhanced, and the enhancement quality of the image is ensured.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology and relates to a method, device, storage medium and terminal for enhancing X-ray images of large animals. Background Technology

[0002] In fields such as non-destructive testing and clinical diagnosis, X-rays are often used to penetrate the subject, and the penetrated X-rays are received by a detector to generate an X-ray image of the subject. However, when the subject is a large animal such as a pig, horse, or cattle, a large detector and a large SID (short distance between the X-ray source and the detector) are required to obtain a complete picture of the animal's body. Furthermore, to avoid overexposure of the image edges and the resulting "burn-in" effect, the radiation dose must be reduced. However, reducing the radiation dose inevitably lowers the image signal-to-noise ratio, leading to a deterioration in image quality and affecting the identification of key information such as animal edges and vertebrae. Therefore, it is necessary to perform image enhancement on the generated X-ray image to highlight key information such as the animal outline and small edges of vertebrae, thus facilitating human observation or further analysis (such as fat thickness measurement).

[0003] Currently, commonly used image enhancement algorithms include histogram equalization, CLAHE algorithm, and Retinex enhancement. Histogram equalization enhances images by adjusting the grayscale values ​​of pixels. However, this adjustment often leads to over-enhancing of certain information, resulting in amplified image noise or reduced grayscale levels in local areas, causing a loss of detail and failing to highlight crucial information such as vertebral details. CLAHE is a method for enhancing local contrast. It divides the image into small blocks and performs histogram equalization on each block to improve local contrast. This algorithm may produce block artifacts, blurring image edges and / or masking image details, affecting image quality. Retinex enhancement aims to adjust contrast and brightness while preserving image details. However, the large kernel filtering introduced during enhancement can cause strong overshoot artifacts, leading to loss of image details and hindering the prominence of edge information.

[0004] Therefore, a new enhancement method is needed to highlight details such as animal edges and vertebrae in the image while avoiding the introduction of new artifacts and ensuring the enhancement quality of the image. Summary of the Invention

[0005] The purpose of this invention is to provide a method, device, storage medium, and terminal for enhancing X-ray images of large animals, in order to solve the technical problem of poor enhancement quality of X-ray images of large animals in the prior art.

[0006] In a first aspect, the present invention provides a method for enhancing X-ray images of large animals, comprising:

[0007] S1. Preprocess the original X-ray image to obtain a preprocessed image with enhanced target details; the target details are the details of the low-absorption region.

[0008] S2. Remove the scattering from the preprocessed image to obtain a valid image;

[0009] S3. Perform multi-scale enhancement on the effective image to obtain a multi-scale enhanced image;

[0010] S4. Sharpen the multi-scale enhanced image.

[0011] In one embodiment of the present invention,

[0012] Step S1 includes performing a logarithmic transformation on the original X-ray image.

[0013] In one embodiment of the present invention,

[0014] Step S1 also includes inverting the colors of the logarithmically transformed image.

[0015] In one embodiment of the present invention, the scattering amount in step S2 is a low-frequency component; the method for removing scattering amount from the preprocessed image includes:

[0016] Extract low-frequency components from the preprocessed image to obtain a low-frequency image;

[0017] The effective image is calculated based on the preprocessed image and the extracted low-frequency image.

[0018] In one embodiment of the present invention, the multi-scale enhancement step in step S3 includes:

[0019] S31. Perform multi-scale decomposition on the effective image to obtain low-frequency sub-bands and high-frequency sub-bands at different scales;

[0020] S32. Enhance the high-frequency subband at different scales;

[0021] S33. Reconstruct the image based on the low-frequency subbands at different scales and the enhanced high-frequency subbands to obtain a multi-scale enhanced image.

[0022] In one embodiment of the present invention, the sharpening process is performed using an unsharpening mask or Laplacian sharpening.

[0023] Secondly, the present invention also provides an enhancement device for X-ray images of large animals, comprising:

[0024] The preprocessing module is used to perform logarithmic transformation and color inversion on the original X-ray image to obtain a preprocessed image;

[0025] The high-frequency enhancement module is used to extract low-frequency components from the preprocessed image and obtain an effective image with high-frequency enhancement based on the preprocessed image and the extracted low-frequency components.

[0026] The multi-scale enhancement module is used to perform multi-scale enhancement on the high-frequency components in the effective image to obtain a multi-scale enhanced image.

[0027] The sharpening module is used to sharpen multi-scale images to obtain the final enhanced image.

[0028] Thirdly, the present invention also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for enhancing X-ray images of large animals.

[0029] Fourthly, the present invention also provides a terminal, including a processor and a memory, wherein the memory and the processor are communicatively connected;

[0030] The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to enable the terminal to perform the above-described method for enhancing X-ray images of large animals.

[0031] As described above, the method, apparatus, storage medium, and terminal for enhancing X-ray images of large animals according to the present invention have the following beneficial effects:

[0032] This invention first performs logarithmic transformation and color inversion on X-ray images of large animals. The logarithmic transformation highlights target details in low-absorption regions, improving the overall image contrast. Then, color inversion brightens the low-absorption regions containing target details, enhancing visual perception and further highlighting the target details. Next, some invalid low-frequency components caused by scattering are removed to enhance high-frequency highlights and brightness diffusion. This not only makes high-frequency information such as bone boundaries clearer but also suppresses artifacts, improving image quality. Subsequently, a multi-scale decomposition method is used to decompose the image into low-frequency and high-frequency components at different scales, and the decomposed high-frequency components are enhanced to further highlight small parts of the target details. Finally, sharpening is performed to reduce noise and suppress edge artifacts, thereby enhancing contours and improving contrast. This invention, through the above four steps of fusion enhancement of X-ray images of large animals, can highlight target details and edge contours such as vertebrae while avoiding the introduction of new artifacts, thus improving the quality of image enhancement. Attached Figure Description

[0033] Figure 1A schematic flowchart of the method for enhancing X-ray images of large animals according to an embodiment of the present invention is shown.

[0034] Figure 2 The enhancement curves of the high-frequency subbands at various scales are shown.

[0035] Figure 3 The illustration shows a comparison of the enhancement effects of the method for enhancing X-ray images of large animals according to embodiments of the present invention.

[0036] Figure 4 A schematic diagram of the structure of the enhancement device for X-ray images of large animals according to an embodiment of the present invention is shown.

[0037] Figure 5 A schematic diagram of the terminal according to an embodiment of the present invention is shown. Detailed Implementation

[0038] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0039] The following will describe in detail the principles and implementation methods of the X-ray low-signal image enhancement method, enhancement device, storage medium and terminal of this embodiment, so that those skilled in the art can understand the X-ray low-signal image enhancement method, device, storage medium and terminal of this embodiment without creative effort.

[0040] To address the aforementioned technical problems in the prior art, embodiments of the present invention provide a method for enhancing X-ray images of large animals.

[0041] Figure 1 A schematic flowchart of the method for enhancing X-ray images of large animals according to an embodiment of the present invention is shown. (Refer to...) Figure 1 As shown, the method for enhancing X-ray images of large animals according to an embodiment of the present invention includes the following steps.

[0042] S1. Preprocess the original X-ray image to obtain a preprocessed image with enhanced target details; where the target details are the details of the low absorption region.

[0043] Based on the imaging characteristics of X-rays, when X-rays pass through tissues of varying densities and thicknesses in large animals, they are absorbed to varying degrees by the tissues, resulting in different energies of the X-rays reaching the detector and thus forming a black-and-white X-ray image. Specifically, the thicker the tissue, the less X-rays pass through, appearing as whiter areas in the image; conversely, untended background areas transmit more X-rays and appear as black areas. Therefore, bright areas in the original X-ray image can be referred to as high-absorption areas, and dark areas as low-absorption areas.

[0044] It should be noted that the original X-ray image is the initial detector image that has undergone full calibration.

[0045] The preprocessing in step S1 includes performing a logarithmic transformation on the original X-rays. The purpose of the logarithmic transformation is to enhance the image contrast and highlight the target details in the dark areas. The target details include, but are not limited to, key details such as soft tissue boundaries and rib boundaries.

[0046] Specifically, the formula for the logarithmic transformation is:

[0047] I log (x) = log e (I(x))

[0048] Where I(x) is the gray value of the original x-ray image at pixel x; I log (x) is the gray value of the image at pixel x after logarithmic transformation.

[0049] Optionally, the preprocessing in step S1 may also include inverting the colors of the logarithmically transformed image (i.e., the object image) to obtain the final preprocessed image.

[0050] Specifically, the formula for color inversion processing in this application is as follows:

[0051] I invert (x)=max(I log (x))-I log (x)+min(I log (x))

[0052] Among them, I log (x) represents the gray value at pixel x in the image after logarithmic transformation; max(I log (x) represents the maximum gray value in the image after logarithmic transformation; min(I log (x) represents the minimum gray value in the image after logarithmic transformation; I invert (x) is the gray value of the inverted image at pixel x.

[0053] The inverse color formula used in this application can not only brighten the low absorption area where the target details are located without changing the grayscale range and brightness reference (brightness-darkness contrast), thus improving the visual effect, but also provide a basis for subsequent high-frequency enhancement, ensuring the processing effect of high-frequency enhancement.

[0054] S2. Remove the scattering from the preprocessed image to obtain a valid image.

[0055] In X-ray imaging, thick tissues (such as the back region) or irregular shapes on the animal's body surface increase photon scattering (i.e., generate low-frequency noise), resulting in reduced image contrast. Therefore, it is necessary to remove the blurred background caused by scattering in order to highlight high-frequency information such as bones and organ boundaries.

[0056] Since the scattering amount can be regarded as the low-frequency component of the image, high-frequency enhancement can be achieved by removing the low-frequency component.

[0057] Based on this, high-frequency enhancement methods include, but are not limited to, the following two.

[0058] Method 1:

[0059] Directly applying high-pass filtering to the preprocessed image is quick and convenient, but while it enhances high-frequency components, it also amplifies noise in the image, which may lead to the loss of key structural information.

[0060] Method 2:

[0061] S21. Extract the low-frequency components from the preprocessed image to obtain a low-frequency image;

[0062] S22. Calculate the effective image based on the preprocessed image and the extracted low-frequency image.

[0063] In step S21, the extraction of low-frequency components can be performed using spatial domain filtering methods such as mean filtering and Gaussian filtering, or frequency domain filtering methods such as Fourier transform, without limitation; in this embodiment, Gaussian filtering is preferred for the extraction of low-frequency components.

[0064] In step S22, the formulas involved in acquiring the valid image are as follows:

[0065] I valid (x)=I invert (x)-ω×I lowpass (x)

[0066] Among them, I invert (x) is the gray value at pixel x in the inverted image; lowpass(x) represents the grayscale value of the low-frequency image at pixel x; ω is a weighting parameter (i.e., the intensity of low-frequency removal), satisfying 0 ≤ ω ≤ 1, and ω is determined according to the shooting scene (e.g., when shooting the back, ω needs to be greater than 0.5; when shooting the limbs, ω needs to be less than 0.5); I invert (x) is the gray value of the inverted image at pixel x.

[0067] This method can suppress noise interference on the image by adjusting the weight parameter ω. It can not only remove some of the invalid low-frequency information brought by scattering, so as to retain and highlight more high-frequency details such as edges and textures, but also diffuse the image brightness to protect the edges such as bones in the target and suppress the generation of artifacts.

[0068] Since Method 2 can flexibly select the weight parameter ω for low-noise suppression based on the shooting scene, it effectively ensures the image quality of the acquired valid image. Therefore, in this embodiment, Method 2 is preferred for high-frequency enhancement.

[0069] S3. Perform multi-scale enhancement on the effective image to obtain a multi-scale enhanced image.

[0070] Specifically, the steps of multi-scale enhancement include:

[0071] S31. Perform multi-scale decomposition on the effective image to obtain low-frequency sub-bands and high-frequency sub-bands at different scales;

[0072] Among them, the multi-scale decomposition method includes, but is not limited to, various existing multi-scale decomposition methods such as Gaussian Laplace pyramid decomposition or wavelet decomposition; in this embodiment, the multi-scale decomposition method used is Gaussian Laplace pyramid decomposition.

[0073] Specifically, sub-band images at various scales of the effective image are obtained through Gaussian Laplacian pyramid decomposition. The specific decomposition principle is as follows:

[0074] The effective image is used as the 0th layer low-frequency subband of the Gaussian pyramid;

[0075] The low-frequency subband of the current layer is low-pass filtered and then downsampled to obtain the low-frequency subband of the next layer. This process is repeated until the required number of layers is reached, thus completing the construction of the Gaussian pyramid. At this point, the low-frequency subbands of different layers of the Gaussian pyramid represent low-frequency subbands at different scales.

[0076] After upsampling each low-frequency subband of the Gaussian pyramid, Gaussian filtering is performed, and the difference between the low-frequency subband of the current layer of the Gaussian pyramid and the high-frequency subband of the current layer of the Laplace pyramid is obtained. The above steps are repeated until a complete Laplace pyramid is constructed. At this point, the high-frequency subbands of different layers of the Laplace pyramid represent high-frequency subbands at different scales.

[0077] S32. Enhance the high-frequency subband at different scales;

[0078] To highlight subtle details in an image, high-frequency subbands at different scales need to be enhanced. The enhancement formulas involved are as follows:

[0079]

[0080] Among them, L k (x) represents the grayscale value of the high-frequency subband at the k-th scale at pixel x; β is an adjustable parameter, typically ranging from 1 to 100, set according to actual conditions; CV is an enhancement parameter, satisfying CV = mean(L k (x)) / std(L k (x)), mean(L k (x)) represents the average gray level of the high-frequency sub-band at the k-th scale, std(L) k (x)) represents the standard deviation of the gray level of the high-frequency sub-band at the k-th scale; This refers to the enhanced k-th scale high-frequency subband.

[0081] The curve diagram of the above enhancement formula is shown below. Figure 2 As shown, when the grayscale in the detail image is low, a larger parameter is applied for enhancement. When encountering strong edges, appropriately reducing the enhancement coefficient can suppress some strong edge artifacts and noise. Through the above calculations, small details in the image are enhanced, highlighting key information. Compared with traditional linear high-frequency enhancement methods, the high-frequency enhancement method used in this application can adaptively adjust the enhancement degree of each layer to ensure that small details in the image are clearly highlighted, thus guaranteeing the accuracy of subsequent processing.

[0082] S33. Reconstruct the image based on the low-frequency subbands at different scales and the enhanced high-frequency subbands to obtain a multi-scale enhanced image.

[0083] Since image reconstruction is a current technology, it will not be elaborated upon further.

[0084] S4. Sharpen the multi-scale enhanced image to highlight the complete contours.

[0085] It should be noted that the sharpening process includes, but is not limited to, existing sharpening methods such as unsharpening masks or Laplacian sharpening; in this embodiment, the preferred sharpening method is an unsharpening mask.

[0086] Specifically, the calculation formula involved in the desharpening mask is as follows:

[0087] I final (x)=λ×I multi (x)+(1-λ)×I smooth (x)

[0088] Among them, I multi (x) represents the gray value at pixel x in the multi-scale enhanced image; smooth (x) represents the grayscale value at pixel x in the image obtained after multi-scale enhanced image filtering; λ is the sharpening intensity, satisfying 1≤λ, and adjusted according to the actual situation; I final (x) represents the grayscale value of the final enhanced image at pixel x.

[0089] In this embodiment, the filtering process for the multi-scale enhanced image is edge-preserving filtering, which enhances the edges while improving image contrast.

[0090] Figure 3 The accompanying diagrams show a comparison of the enhancement effects of the method for enhancing X-ray images of large animals according to embodiments of the present invention. Figure 3 As shown in the figure, the left image is the image before enhancement and the right image is the image after enhancement. It can be seen that the enhancement method of the present invention significantly enhances the image contrast.

[0091] The scope of protection of the method for enhancing X-ray images of large animals according to the embodiments of the present invention is not limited to the order of steps listed in this embodiment. Any solution achieved by adding, subtracting, or replacing steps in the prior art based on the principles of the present invention is included within the scope of protection of the present invention.

[0092] The enhancement method for large animal X-ray images in this invention performs logarithmic transformation and color inversion on the original X-ray image to enhance visual effects while highlighting details in dark areas. Then, it removes some low-frequency components caused by scattering to retain and highlight more edge details by reducing low-frequency noise, while suppressing artifacts. Next, multi-scale enhancement is used to further highlight fine details in the image. Finally, sharpening is applied to suppress edge artifacts and highlight complete contours. This four-step fusion enhancement method significantly enhances a large number of details and edge contours in large animal X-ray images, effectively improving the image enhancement quality.

[0093] To address the aforementioned technical problems in the prior art, embodiments of the present invention also provide an enhancement device for X-ray images of large animals.

[0094] Figure 4 A schematic diagram of the structure of the enhancement device for X-ray images of large animals according to an embodiment of the present invention is shown. (Refer to...) Figure 4 As shown, the enhancement device for X-ray images of large animals in this embodiment of the invention includes a preprocessing module, a high-frequency emphasis module, a multi-scale enhancement module, and a sharpening module.

[0095] The preprocessing module is used to perform logarithmic transformation and color inversion on the original X-ray image to obtain a preprocessed image;

[0096] The high-frequency enhancement module is used to extract low-frequency components from the preprocessed image and obtain an effective image with high-frequency enhancement based on the preprocessed image and the extracted low-frequency components.

[0097] The multi-scale enhancement module is used to perform multi-scale enhancement on the high-frequency components in the effective image to obtain a multi-scale enhanced image.

[0098] The sharpening module is used to sharpen multi-scale images to obtain the final enhanced image.

[0099] The enhancement device for large animal X-ray images in this embodiment of the invention first performs logarithmic transformation and color inversion on the original X-ray image to enhance visual effects while highlighting details in dark areas. Then, it removes some low-frequency components caused by scattering to retain and highlight more edge details by reducing low-frequency noise, while suppressing artifacts. Next, it uses multi-scale enhancement to further highlight fine details in the image. Finally, it uses sharpening to suppress edge artifacts and highlight complete contours. This four-step fusion enhancement method significantly highlights a large number of details and edge contours in large animal X-ray images, effectively improving the enhancement quality of the image.

[0100] To address the aforementioned technical problems in the prior art, embodiments of the present invention also provide a storage medium storing a computer program, characterized in that, when executed by a processor, the program implements all steps of the method for enhancing X-ray images of large animals according to the embodiments.

[0101] The specific steps of the method for enhancing X-ray images of large animals and the beneficial effects obtained by applying the readable storage medium provided in the embodiments of the present invention are the same as those in the above embodiments, and will not be repeated here.

[0102] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof. The storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. This available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state drive (SSD)).

[0103] To address the aforementioned technical problems in the prior art, embodiments of the present invention also provide a terminal. Figure 5 A schematic diagram of the terminal structure according to an embodiment of the present invention is shown. (Refer to...) Figure 5 As shown, the terminal in this embodiment of the invention includes a processor and a memory, with the memory and processor being communicatively connected; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal performs all the steps of the X-ray low-signal image enhancement method of the above embodiment.

[0104] The specific steps of the method for enhancing low-signal X-ray images and the beneficial effects obtained by applying the terminal provided in this embodiment are the same as those in the above embodiments, and will not be repeated here.

[0105] It should be noted that the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Similarly, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0106] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of protection of this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A method for enhancing X-ray images of large animals, characterized in that, include: S1. Preprocess the original X-ray image to obtain a preprocessed image with enhanced target details; the target details are the details of the low-absorption region. S2. Remove the scattering from the preprocessed image to obtain a valid image; S3. Perform multi-scale enhancement on the effective image to obtain a multi-scale enhanced image; S4. Sharpen the multi-scale enhanced image.

2. The enhancement method according to claim 1, characterized in that, Step S1 includes performing a logarithmic transformation on the original X-ray image.

3. The enhancement method according to claim 2, characterized in that, Step S1 also includes inverting the colors of the logarithmically transformed image.

4. The enhancement method according to any one of claims 1 to 3, characterized in that, The scattering in step S2 is a low-frequency component; Methods for removing scattering from preprocessed images include: Extract low-frequency components from the preprocessed image to obtain a low-frequency image; The effective image is calculated based on the preprocessed image and the extracted low-frequency image.

5. The enhancement method according to any one of claims 1 to 3, characterized in that, In the S3 step, the multi-scale enhancement steps include: S31. Perform multi-scale decomposition on the effective image to obtain low-frequency sub-bands and high-frequency sub-bands at different scales; S32. Enhance the high-frequency subband at different scales; S33. Reconstruct the image based on the low-frequency subbands at different scales and the enhanced high-frequency subbands to obtain a multi-scale enhanced image.

6. The enhancement method according to any one of claims 1 to 3, characterized in that, The sharpening process can be achieved using an unsharpening mask or Laplacian sharpening.

7. An enhancement device for X-ray images of large animals, comprising: The preprocessing module is used to perform logarithmic transformation and color inversion on the original X-ray image to obtain a preprocessed image; The high-frequency enhancement module is used to extract low-frequency components from the preprocessed image and obtain an effective image with high-frequency enhancement based on the preprocessed image and the extracted low-frequency components. The multi-scale enhancement module is used to perform multi-scale enhancement on the high-frequency components in the effective image to obtain a multi-scale enhanced image. The sharpening module is used to sharpen multi-scale images to obtain the final enhanced image.

8. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the enhanced method as described in any one of claims 1 to 6.

9. A terminal, characterized in that, The device includes a processor and a memory, the memory being communicatively connected to the processor; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the terminal to perform the enhanced method as described in any one of claims 1 to 6.

Images