Method and device for enhancing X-ray low-signal image, storage medium and terminal

Abstract

The invention provides an enhancement method and device for an X-ray low-signal image, a storage medium and a terminal, and the method comprises the steps: carrying out the band-pass filtering processing of the X-ray low-signal image through a preset band-pass function, so as to obtain a filtered image; performing foreground segmentation on the filtered image to obtain a segmentation threshold value, and obtaining a foreground and background discrimination map corresponding to the filtered image based on the segmentation threshold value; and performing fusion enhancement on the filtered image by using the discrimination map to obtain an enhanced image. According to the invention, the contrast of the x-ray low-signal image can be effectively enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and relates to a method for enhancing X-ray low-signal images, in particular to a method and device for enhancing X-ray low-signal images, a storage medium, and a terminal. Background Art

[0002] In the field of X-ray imaging, X-ray images are well-known for their wide grayscale distribution and rich detail information, and are important tools for applications such as medical imaging and non-destructive testing. However, due to the influence of ray intensity, object absorption characteristics, and sensor response characteristics during the generation process of X-ray images, the original images often have problems such as low contrast and large noise interference. Therefore, enhancing the contrast of the images has become a key link in improving image quality. Image contrast enhancement aims to make the foreground objects more prominent by adjusting the grayscale distribution, thereby improving the visibility and resolvability of the signals. Currently, common image contrast enhancement methods mainly include histogram equalization, histogram stretching, and multi-scale enhancement techniques, etc. Histogram equalization adjusts the grayscale value distribution to make the image grayscale cover a wider range and improve the overall contrast; histogram stretching linearly expands the original grayscale range to enhance the local contrast; multi-scale enhancement techniques decompose the feature information of the image at different scales and enhance the significant objects step by step.

[0003] However, the effectiveness of these methods usually depends on the intensity of the direct component signal in the X-ray image. That is to say, when the ray dose is moderate and the direct signal dominates, these methods can significantly improve the image quality. Therefore, for X-ray images with weak signals, such as low-dose imaging or scenes with strong target absorption, these traditional algorithms often fail to meet the requirements, resulting in limited or ineffective enhancement effects.

[0004] Although the existing contrast enhancement technologies have achieved certain results in enhancing X-ray images, they still face many challenges when dealing with X-ray images with weak signals. First, in weak-signal X-ray images, the direct X-ray signal is weak, and the grayscale contrast between the foreground objects and the background in the image is already weak. When traditional algorithms enhance the overall contrast of the image, they cannot effectively distinguish the foreground from the background. In this case, the enhancement of the background area will further obscure the foreground objects, making the objects more difficult to distinguish and even possibly losing key information. Second, weak-signal X-ray images are usually accompanied by large noise interference, and the noise signals are often further amplified during the enhancement process, thus having an adverse impact on the image quality. Traditional algorithms lack a targeted noise suppression mechanism and rely only on global grayscale distribution adjustment, making it difficult to highlight the effective signals. Third, the existing technologies have poor enhancement effects on low-signal-to-noise ratio images. Especially when the foreground signal is submerged in the background noise, the enhancement results may show problems such as excessive blurring or loss of texture details.

[0005] Generally speaking, current image enhancement methods have the defects of low efficiency and poor robustness when dealing with the special requirements of weak signal X-ray images. There is an urgent need to develop more targeted and intelligent enhancement technologies to improve the imaging quality in weak signal scenarios and meet the actual application requirements in the medical and industrial fields. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and device for enhancing X-ray low-signal images, a storage medium, and a terminal, which are used to solve the technical problem of poor enhancement effect of weak signal X-ray images in the prior art.

[0007] In a first aspect, the present invention provides a method for enhancing X-ray low-signal images, including:

[0008] Performing band-pass filtering on the X-ray low-signal image by using a preset band-pass function to obtain a filtered image;

[0009] Performing foreground segmentation on the filtered image to obtain a segmentation threshold, and obtaining a discrimination map of the foreground and background corresponding to the filtered image based on the segmentation threshold;

[0010] Performing fusion enhancement on the filtered image by using the discrimination map to obtain an enhanced image.

[0011] In an embodiment of the present invention,

[0012] The method for performing foreground segmentation on the filtered image includes: Otsu threshold segmentation method, triangle threshold segmentation method, and adaptive threshold segmentation method.

[0013] In an embodiment of the present invention,

[0014] Obtaining the discrimination map of the foreground and background corresponding to the filtered image based on the segmentation threshold includes:

[0015]

[0016] Where map(x) represents the weight value of the discrimination map of the foreground and background at pixel x, T represents the segmentation threshold, Max represents the maximum pixel value of the filtered image, and I bandpass (x) represents the gray value of the filtered image at pixel x.

[0017] In an embodiment of the present invention, it further includes:

[0018] Performing further gray processing on the discrimination map of the foreground and background by using a preset method, and using the processed discrimination map as the discrimination map. The gray processing includes gray stretching and gray compression.

[0019] In one embodiment of the present invention, the preset method is the gamma transformation method.

[0020] In one embodiment of the present invention, using the discrimination map to fuse and enhance the filtered image to obtain the enhanced image includes:

[0021] Taking the discrimination map as the weight value of the corresponding region in the filtered image, multiplying the weight value with the corresponding region point by point to obtain the enhanced image.

[0022] In a second aspect, the present invention also provides an enhancement device for X-ray low-signal images, which is characterized by including:

[0023] A band-pass filtering module, configured to perform band-pass filtering on the X-ray low-signal image by using a preset band-pass function to obtain a filtered image;

[0024] A discrimination module, configured to perform foreground segmentation on the filtered image to obtain a segmentation threshold, and obtain a discrimination map of the foreground and background corresponding to the filtered image based on the segmentation threshold;

[0025] A fusion enhancement module, configured to use the discrimination map to fuse and enhance the filtered image to obtain an enhanced image.

[0026] In a third aspect, the present invention also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the enhancement method for X-ray low-signal images as described above.

[0027] In a fourth aspect, the present invention also provides a terminal, including a processor and a memory, and the memory is communicatively connected to the processor;

[0028] 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 executes the enhancement method for X-ray low-signal images as described above.

[0029] As described above, the enhancement method and device, storage medium, and terminal for X-ray low-signal images according to the present invention have the following beneficial effects:

[0030] 1. In the present invention, first, band-pass filtering is performed on the X-ray low-signal image as the first step of enhancement, which can reduce noise and also reduce some low-frequency components while highlighting the target edge; then, foreground and background discrimination maps are obtained through threshold segmentation, and the discrimination maps are used as weights to perform point multiplication with the corresponding regions for fusion enhancement as the second step of enhancement, further increasing the contrast between the foreground and the background. Through the above two-step fusion enhancement method, the contrast enhancement of the X-ray low-signal image is realized.

[0031] 2. The enhancement of the X-ray low-signal image in the present invention does not require an increase in dose or a change in shooting conditions, and has a wider application range. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The flowchart of the method for enhancing the X-ray low-signal image according to the embodiment of the present invention is shown.

[0033] Figure 2 The comparison of the enhancement effect diagrams of the method for enhancing the X-ray low-signal image according to the embodiment of the present invention is shown.

[0034] Figure 3 The structural schematic diagram of the device for enhancing the X-ray low-signal image according to the embodiment of the present invention is shown.

[0035] Figure 4 The structural schematic diagram of the terminal according to the embodiment of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The following specific examples illustrate the embodiments of the present invention. Those skilled in the art can easily understand the 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, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0037] The principles and embodiments of the method and device for enhancing the X-ray low-signal image, the storage medium, and the terminal in this embodiment will be elaborated in detail below, so that those skilled in the art can understand the method, device, storage medium, and terminal for enhancing the X-ray low-signal image in this embodiment without creative labor.

[0038] To solve the above technical problems existing in the prior art, the embodiment of the present invention provides a method for enhancing an X-ray low-signal image.

[0039] Figure 1 The flowchart of the method for enhancing the X-ray low-signal image according to the embodiment of the present invention is shown. Refer to Figure 1 As shown, the method for enhancing the X-ray low-signal image in the embodiment of the present invention mainly includes steps S100 to S300.

[0040] Step S100: Perform band-pass filtering on the X-ray low-signal image using a preset band-pass function to obtain a filtered image.

[0041] Band-pass filtering is a signal processing technique that preserves the effective signals in an image by selecting a specific frequency range (band-pass range), while filtering out low-frequency components (such as background illumination changes) and high-frequency components (such as random noise). Preset band-pass functions are used to define the characteristics of the band-pass filter, such as the frequency range and filtering intensity. Preset band-pass functions include Gaussian functions and Butterworth filters.

[0042] The input data in step S100 is an X-ray low-signal image, which is an X-ray image with a low signal-to-noise ratio, unclear details, and a limited gray-scale dynamic range. Specifically, perform a frequency-domain transformation (such as fast Fourier transform, FFT) on the input data to convert the X-ray low-signal image from the spatial domain to the frequency domain to obtain a frequency-domain image. Use a preset band-pass function to process the frequency-domain image to obtain a filtered frequency-domain image. Perform an inverse Fourier transform (IFFT) on the filtered frequency-domain image to restore it to the filtered image in the spatial domain as the filtered image.

[0043] Through the above band-pass filtering steps, the low-signal target area in the X-ray low-signal image is enhanced, the structural or texture information within a specific frequency range is enhanced, and while reducing noise, some low-frequency components can also be reduced, thereby highlighting the target edge. The enhanced image has better contrast and resolution in the target area, laying a foundation for subsequent foreground segmentation and fusion enhancement.

[0044] Step S200: Perform foreground segmentation on the filtered image to obtain a segmentation threshold, and obtain the discrimination map of the foreground and background corresponding to the filtered image based on the segmentation threshold.

[0045] Specifically, statistically analyze the gray-scale distribution of the filtered image to generate a histogram. The histogram uses the gray-scale value as the horizontal axis and the frequency of pixel points as the vertical axis, and the histogram reflects the gray-scale distribution characteristics of the image. The histogram usually has multiple peaks and valleys, corresponding to the gray-scale distribution differences between the foreground and the background. Determine the segmentation threshold T based on the histogram characteristics using a segmentation algorithm to maximize the distinction between the foreground and the background. The discrimination map is a weight distribution map that describes the gray-scale difference degree between the foreground and background regions in the filtered image. For each pixel point x, calculate its discrimination based on the difference between its gray-scale value and the segmentation threshold T. When the pixel gray-scale value is close to T, the discrimination map value is low; when the pixel gray-scale value is far from T (closer to the foreground or background extreme values), the discrimination map value is high. The above process provides an accurate regional weight distribution basis for subsequent fusion enhancement by obtaining the discrimination map.

[0046] In a specific implementation manner of step S200, the methods for foreground segmentation of the filtered image include: Otsu threshold segmentation method, triangle threshold segmentation method, and adaptive threshold segmentation method. The Otsu method automatically finds the optimal segmentation threshold by maximizing the between-class variance of the foreground and background. Its main process is as follows: construct the gray-level histogram of the filtered image, traverse each possible gray value as the segmentation threshold T and calculate the between-class variance of the foreground and background, so as to obtain the threshold T that maximizes the between-class variance as the segmentation threshold. The triangle threshold segmentation method finds the segmentation threshold through the morphological characteristics of the histogram and is applicable to images with single-peak or skewed gray-level distributions. Its main process is as follows: obtain the position of the maximum peak of the histogram (the gray value corresponding to the maximum gray frequency) in the gray-level histogram of the filtered image, connect the peak point with the end of the histogram to form a straight line based on the maximum peak point of the histogram, and find the point on the histogram with the largest distance from this straight line, and its gray value is the segmentation threshold T. The adaptive threshold segmentation method dynamically adjusts the segmentation threshold according to the gray characteristics of local regions of the image and is suitable for images with uneven gray-level distributions. Its main process is as follows: divide the filtered image into multiple small regions (windows), calculate the statistical quantity of the local gray characteristics (such as local average gray value or median) for each small region, and dynamically determine the threshold according to the local statistics. Through the above three segmentation methods, the appropriate foreground segmentation method can be flexibly selected according to the characteristics of the actual filtered image, providing an accurate segmentation threshold for generating the discrimination map of the foreground and background.

[0047] In a specific implementation manner of step S200, obtaining the discrimination map of the foreground and background corresponding to the filtered image based on the segmentation threshold includes:

[0048]

[0049] Among them, map(x) represents the weight value of the discrimination map of the foreground and background at pixel x, T represents the segmentation threshold, Max represents the maximum pixel value of the filtered image, and I bandpass (x) represents the gray value of the filtered image at pixel x. In this method, when the value of I bandpass (x) is close to the foreground region (greater than the threshold T), the value of map(x) tends to 1; when the value of I bandpass (x) is close to the background region (close to or less than T), the value of map(x) tends to -1. Therefore, this method can obtain an accurate discrimination map.

[0050] Optionally, it further includes: further gray-scale processing the foreground-background discrimination map using a preset method, and using the processed discrimination map as the discrimination map. The gray-scale processing includes gray-scale stretching and gray-scale compression. After obtaining the initial discrimination map, it is optimized using a preset gray-scale processing method to generate a processed discrimination map (i.e., the improved discrimination map), providing a more accurate basis for subsequent enhancement processing. Gray-scale stretching further expands the value range of the discrimination map to [0, 1] or a larger range, making the gray-scale difference between the foreground and the background more significant, so as to enhance the contrast between the foreground and the background in the image. When the foreground signal is weak and easily submerged by the background signal, gray-scale stretching helps to improve the saliency of the foreground signal. Gray-scale compression uses a non-linear function (such as a logarithmic or power function) to compress the gray-scale values of the discrimination map, making the gray-scale distribution of the image more uniform, so as to weaken the prominence of some high-gray-scale or low-gray-scale regions in the image. When the gray-scale range of the image is too large and some regions (such as the highlighted part) are over-enhanced, gray-scale compression helps to balance the processing effect.

[0051] Optionally, the preset method is the gamma transformation method. Specifically, the mathematical expression of the Gamma transformation is:

[0052]

[0053] where, I in represents the input gray-scale value (i.e., the gray-scale value of the foreground-background discrimination map), I out represents the transformed gray-scale value (i.e., the gray-scale value of the processed discrimination map), γ represents the Gamma value, which is used to control the non-linearity degree of the transformation, and c represents a preset coefficient, which is used to adjust the overall gray-scale value range. The Gamma transformation can dynamically adjust the gray-scale distribution according to the characteristics of the discrimination map. By controlling the Gamma value, it enhances the gray-scale details in the dark part or suppresses the gray-scale details in the bright part, making the separation between the foreground and the background more significant. Selecting an appropriate Gamma value for non-linear gray-scale transformation according to the image to generate a processed discrimination map is beneficial to improving the effect of subsequent fusion enhancement.

[0054] Step S300: Use the discrimination map to perform fusion enhancement on the filtered image to obtain the enhanced image.

[0055] Specifically, based on the discrimination map, perform fusion enhancement on the filtered image (i.e., the image after band-pass filtering), and finally output the image with the target signal enhanced.

[0056] In a specific implementation of step S300, the filtered image is fused and enhanced using a discrimination map to obtain the enhanced image, which includes: using the discrimination map as the weight value of the corresponding region in the filtered image, and multiplying the weight value with the corresponding region to obtain the enhanced image. Specifically, refer to the following formula:

[0057] I enh (x) = map(x) · I bandpass (x)

[0058] Where, I bandpass (x) represents the gray value of the filtered image at pixel x, I enh (x) represents the gray value of the enhanced image at pixel x, and map(x) represents the discrimination map at pixel x. By adaptively assigning weights to different regions through the discrimination map to adapt to the changes in complex image scenes, through the weighting effect of the discrimination map, the enhancement of foreground targets and the suppression of background signals are effectively achieved, the signal-to-noise ratio of the target region is improved, and the image quality of X-ray low-signal images can be significantly enhanced.

[0059] Figure 2 The comparison of the enhancement effect diagrams of the X-ray low-signal image enhancement method described in the embodiments of the present invention is shown. Refer to Figure 2 as shown. Among them, the left figure is the image before enhancement, and the right figure is the image after enhancement. It can be seen that the enhancement method of the present invention significantly enhances the image contrast.

[0060] The protection scope of the X-ray low-signal image enhancement method in the embodiments of the present invention is not limited to the execution order of the steps listed in this embodiment. Any solution achieved by adding or subtracting steps of the prior art and replacing steps according to the principle of the present invention is included in the protection scope of the present invention.

[0061] The X-ray low-signal image enhancement method in the embodiments of the present invention first performs band-pass filtering on the X-ray low-signal image as the first step of enhancement, which can reduce noise and also reduce some low-frequency components to highlight the target edge. Then, the foreground-background discrimination map is obtained through threshold segmentation, and the discrimination map is used as the weight to multiply with the corresponding region for fusion enhancement as the second step of enhancement, further increasing the contrast between the foreground and the background. Through the above two-step fusion enhancement method, the contrast enhancement of the X-ray low-signal image is achieved. In addition, the enhancement of the X-ray low-signal image in the present invention does not require an increase in dose and does not require a change in shooting conditions, and has a wider application range.

[0062] To solve the above technical problems existing in the prior art, the embodiments of the present invention also provide an X-ray low-signal image enhancement device.

[0063] Figure 3The structural schematic diagram of the X-ray low-signal image enhancement device according to the embodiment of the present invention is shown. Refer to Figure 3 As shown, the X-ray low-signal image enhancement device according to the embodiment of the present invention includes a band-pass filtering module, a discrimination module, and a fusion enhancement module.

[0064] The band-pass filtering module is used to perform band-pass filtering on the X-ray low-signal image by using a preset band-pass function to obtain a filtered image.

[0065] The discrimination module is used to perform foreground segmentation on the filtered image to obtain a segmentation threshold, and obtain a discrimination map of the foreground and background corresponding to the filtered image based on the segmentation threshold.

[0066] The fusion enhancement module is used to perform fusion enhancement on the filtered image by using the discrimination map to obtain an enhanced image.

[0067] The X-ray low-signal image enhancement device according to the embodiment of the present invention first performs band-pass filtering on the X-ray low-signal image as the first step of enhancement, which can reduce noise and also reduce some low-frequency components while highlighting the target edge. Then, the discrimination map of the foreground and background is obtained through threshold segmentation, and the discrimination map is used as a weight to perform dot multiplication with the corresponding region for fusion enhancement as the second step of enhancement, further increasing the contrast between the foreground and the background. The contrast enhancement of the X-ray low-signal image is achieved through the above two-step fusion enhancement method. In addition, the enhancement of the X-ray low-signal image by the present invention does not require increasing the dose and does not require changing the shooting conditions, and has a wider application range.

[0068] To solve the above technical problems existing in the prior art, the embodiment of the present invention also provides a storage medium, on which a computer program is stored, and is characterized in that when the program is executed by a processor, all steps of the X-ray low-signal image enhancement method according to the embodiment are realized.

[0069] The specific steps of the X-ray low-signal image enhancement method and the beneficial effects obtained by using the readable storage medium provided by the embodiment of the present invention are the same as those of the above embodiment, and will not be elaborated herein.

[0070] Those of ordinary skill in the art can understand that all or part of the steps in the methods of the above embodiments can be completed by instructing a processor through a program. The program can be stored in a computer-readable storage medium. The storage medium 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 disc, and any combination thereof. The above storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that integrates one or more available media. The 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 disk (SSD)), etc.

[0071] To solve the above technical problems existing in the prior art, an embodiment of the present invention further provides a terminal. Figure 4 The structural schematic diagram of the terminal according to the embodiment of the present invention is shown. Refer to Figure 4 As shown, the terminal according to the embodiment of the present invention includes a processor and a memory, and the memory is 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 so that the terminal executes all the steps of the method for enhancing the X-ray low-signal image in the above embodiment.

[0072] The specific steps of the method for enhancing the X-ray low-signal image and the beneficial effects obtained by applying the terminal provided by the embodiment of the present invention are the same as those in the above embodiment, and will not be elaborated herein.

[0073] It should be noted that the memory may include a Random Access Memory (RAM), and may also include non-volatile memory, such as at least one disk memory. Similarly, the processor may also be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may 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, discrete hardware components.

[0074] Although the embodiments disclosed in the present invention are as described above, the above content is only an embodiment adopted for the convenience of understanding the present invention, and is not intended to limit the present invention. Any person skilled in the art within the technical field to which the present invention pertains may make any modifications and changes in the form of implementation and details without departing from the spirit and scope disclosed by the present invention. However, the protection scope of the present invention shall still be subject to the scope defined by the appended claims.

Claims

1. A method for enhancing an X-ray low signal image, comprising: Performing bandpass filtering on the X-ray low signal image using a preset bandpass function to obtain a filtered image; Performing foreground segmentation on the filtered image to obtain a segmentation threshold, and obtaining a foreground-background distinction map corresponding to the filtered image based on the segmentation threshold; The filtered image is fused and enhanced using the discrimination map to obtain an enhanced image.

2. The method for enhancing an X-ray low signal image according to claim 1, characterized in that: The methods of performing foreground segmentation on the filtered image include: Otsu threshold segmentation method, triangle threshold segmentation method, and adaptive threshold segmentation method.

3. The method for enhancing an X-ray low signal image according to claim 1, characterized in that: Acquiring a foreground-background distinction map corresponding to the filtered image based on the segmentation threshold includes: Wherein, map(x) represents the weight value of the foreground-background distinction map at pixel x, T represents the segmentation threshold, Max represents the maximum pixel value of the filtered image, and I bandpass (x) represents the grayscale value of the filtered image at pixel x.

4. The method for enhancing an X-ray low signal image according to claim 3, characterized in that: Also includes: The foreground-background discrimination map is further gray-scale processed using a preset method, and the processed discrimination map is used as the discrimination map, wherein the gray-scale processing includes gray-scale stretching and gray-scale compression.

5. The method for enhancing an X-ray low signal image according to claim 4, characterized in that: The preset method is a gamma transformation method.

6. The method for enhancing an X-ray low signal image according to claim 1, characterized in that: Using the discrimination map to perform fusion enhancement on the filtered image to obtain an enhanced image includes: The discrimination map is used as the weight value of the corresponding area in the filtered image, and the weight value is multiplied by the corresponding area to obtain the enhanced image.

7. An X-ray low signal image enhancement device, comprising: A bandpass filtering module is used to perform bandpass filtering on the X-ray low signal image using a preset bandpass function to obtain a filtered image; A distinguishability module, used to perform foreground segmentation on the filtered image to obtain a segmentation threshold, and obtain a distinguishability map of the foreground and background corresponding to the filtered image based on the segmentation threshold; The fusion enhancement module is used to use the discrimination map to perform fusion enhancement on the filtered image to obtain an enhanced image.

8. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for enhancing an X-ray low signal image according to any one of claims 1 to 6 is implemented.

9. A terminal, characterized in that: It includes a processor and a memory, and the memory is 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, so that the terminal executes the method for enhancing an X-ray low signal image as described in any one of claims 1 to 6.

Images