A contrast enhancement method suitable for complex special-shaped workpiece X-ray image

Abstract

The present application relates to X-ray image contrast enhancement technology, in particular to a kind of contrast enhancement method suitable for complex special-shaped workpiece X-ray image.The present application solves the problem that the existing X-ray image contrast enhancement method cannot give consideration to high contrast and low noise when applied to the X-ray image of complex special-shaped workpiece, is difficult to realize, and noise is serious.A kind of contrast enhancement method suitable for complex special-shaped workpiece X-ray image, the method is realized using the following steps: step one: collect the X-ray image of complex special-shaped workpiece;Step two: calculate the local deviation of X-ray image;Step three: calculate the enhanced contrast field;Step four: solve energy functional, obtain the iterative form of target image;Step five: the initial value of iteration is substituted into the iterative form of target image, and the target image is obtained by iteration.The present application is suitable for the X-ray image of complex special-shaped workpiece.

Description

Technical Field

[0001] This invention relates to X-ray image contrast enhancement technology, specifically a method for enhancing the contrast of X-ray images of complex, irregularly shaped workpieces. Background Technology

[0002] X-ray imaging systems are widely used for non-destructive testing and evaluation of products and systems in aerospace, weaponry, and automotive industries due to their advantages such as ease of operation, strong penetration, and wide dynamic range. However, for complex and irregularly shaped workpieces, X-ray images acquired by X-ray imaging systems often exhibit low contrast, low brightness, and severe noise. Therefore, to ensure the accuracy of non-destructive testing and evaluation results, contrast enhancement of X-ray images of complex and irregularly shaped workpieces is necessary. Currently, X-ray image contrast enhancement methods are mainly divided into three types: The first type is the X-ray image contrast enhancement method based on distribution mapping. When applied to X-ray images of complex and irregularly shaped workpieces, this method cannot effectively suppress the influence of noise while enhancing contrast, thus it suffers from the problem of not being able to simultaneously achieve high contrast and low noise. The second type is the X-ray image contrast enhancement method based on deep learning. When applied to X-ray images of complex and irregularly shaped workpieces, this method requires a large dataset to train the model, but large datasets are difficult to obtain in industry, thus posing a significant implementation challenge. The third method is a model-optimized X-ray image contrast enhancement method. However, this method cannot effectively suppress noise when applied to X-ray images of complex and irregularly shaped workpieces, resulting in severe noise issues. Therefore, it is necessary to invent a contrast enhancement method suitable for X-ray images of complex and irregularly shaped workpieces to solve the problems of existing X-ray image contrast enhancement methods being unable to balance high contrast and low noise, being difficult to implement, and suffering from severe noise when applied to X-ray images of complex and irregularly shaped workpieces. Summary of the Invention

[0003] To address the problems of existing X-ray image contrast enhancement methods being unable to simultaneously achieve high contrast and low noise when applied to X-ray images of complex and irregularly shaped workpieces, as well as being difficult to implement and suffering from severe noise, this invention provides a contrast enhancement method suitable for X-ray images of complex and irregularly shaped workpieces.

[0004] This invention is achieved using the following technical solution:

[0005] A contrast enhancement method for X-ray images of complex, irregularly shaped workpieces is implemented through the following steps:

[0006] Step 1: Acquire X-ray images of complex, irregularly shaped workpieces using an X-ray imaging system;

[0007] Step 2: Extract gradient information from the X-ray image, construct the gradient field and contrast field of the X-ray image based on the gradient information, and then calculate the local deviation of the X-ray image;

[0008] Step 3: Calculate the adaptive magnification factor of the contrast field based on the local deviation, and calculate the enhanced contrast field based on the adaptive magnification factor;

[0009] Step 4: Construct an energy functional between the enhanced contrast field and the target image, and solve the energy functional using fractional variational method and gradient descent method to obtain the iterative form of the target image;

[0010] Step 5: Perform a logarithmic transformation on the X-ray image to obtain the initial value for iteration. Then, substitute the initial value into the iterative form of the target image and obtain the target image through iteration.

[0011] Compared with existing X-ray image contrast enhancement methods, the contrast enhancement method for X-ray images of complex irregularly shaped workpieces described in this invention utilizes image gradients and local biases to achieve contrast enhancement, thus possessing the following advantages: First, compared with X-ray image contrast enhancement methods based on distribution mapping, this invention, when applied to X-ray images of complex irregularly shaped workpieces, can effectively suppress the influence of noise while enhancing contrast, thus effectively balancing high contrast and low noise. Second, compared with X-ray image contrast enhancement methods based on deep learning, this invention, when applied to X-ray images of complex irregularly shaped workpieces, does not require a large dataset, thus possessing the advantage of low implementation difficulty. Third, compared with X-ray image contrast enhancement methods based on model optimization, this invention, when applied to X-ray images of complex irregularly shaped workpieces, can effectively suppress the influence of noise, thus possessing the advantage of low noise.

[0012] This invention effectively solves the problems of existing X-ray image contrast enhancement methods being unable to balance high contrast and low noise when applied to X-ray images of complex and irregularly shaped workpieces, being difficult to implement, and having severe noise. It is suitable for X-ray images of complex and irregularly shaped workpieces. Detailed Implementation

[0013] A contrast enhancement method for X-ray images of complex, irregularly shaped workpieces is implemented through the following steps:

[0014] Step 1: Acquire X-ray images of complex, irregularly shaped workpieces using an X-ray imaging system;

[0015] Step 2: Extract gradient information from the X-ray image, construct the gradient field and contrast field of the X-ray image based on the gradient information, and then calculate the local deviation of the X-ray image;

[0016] Step 3: Calculate the adaptive magnification factor of the contrast field based on the local deviation, and calculate the enhanced contrast field based on the adaptive magnification factor;

[0017] Step 4: Construct an energy functional between the enhanced contrast field and the target image, and solve the energy functional using fractional variational method and gradient descent method to obtain the iterative form of the target image;

[0018] Step 5: Perform a logarithmic transformation on the X-ray image to obtain the initial value for iteration. Then, substitute the initial value into the iterative form of the target image and obtain the target image through iteration.

[0019] In step one, the X-ray image is denoted as:

[0020] R(i,j),(i,j)∈Ω={0≤i≤W-1,0≤j≤H-1};

[0021] In the formula: R represents the X-ray image; Ω represents the image domain; W represents the width of the X-ray image; H represents the height of the X-ray image;

[0022] In step two, the gradient field and contrast field are represented as follows:

[0023]

[0024] In the formula: C R Represents the contrast field; C R (p) represents the contrast of point p in the X-ray image; Represents the gradient field; This represents the gradient of point p in an X-ray image;

[0025] In step two, the calculation steps for local deviations are as follows:

[0026] First, the least squares method is used in the local region. Fit a smooth plane;

[0027] Among them, local areas It is symmetrical, and in local areas Contained in the image domain Ω, x0 and y0 represent local regions respectively. The x and y coordinates of the center point;

[0028] Let the smooth plane be denoted as:

[0029] f0(x,y) = ax + by + c;

[0030] In the formula: f0 represents a smooth plane; a, b, and c all represent coefficients;

[0031] At the same time, a smooth plane satisfies:

[0032]

[0033] In the formula: f represents the actual plane;

[0034] Then, the local deviation is calculated using the following formula:

[0035]

[0036] In the formula: LD represents local deviation; Indicates local area The set of discrete points after uniform sampling, and Indicates the cardinality of the set;

[0037] In step three, the adaptive amplification factor is calculated using the following formula:

[0038]

[0039] In the formula: k represents the adaptive magnification coefficient; μ represents the magnification factor, and μ > 0; the value of parameter C is 90% of the image gradient magnitude histogram;

[0040] In step three, the enhanced contrast field is calculated using the following formula:

[0041]

[0042] In the formula: G represents the enhanced contrast field;

[0043] In step four, the energy functional is expressed as follows:

[0044]

[0045] In the formula: E represents the energy functional; I represents the target image; D α This represents a fractional-order operator; the parameter β is used to balance image enhancement and denoising. Indicates the smoothing term; Indicates the fidelity item;

[0046] In step five, the logarithmic transformation formula is as follows:

[0047] I0 = γ·log(1+R);

[0048] In the formula: I0 represents the initial value of the iteration; γ represents the amplification factor.

[0049] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but all such changes and modifications fall within the scope of protection of the present invention.

Claims

1. A contrast enhancement method for X-ray images of complex, irregularly shaped workpieces, characterized in that: This method is implemented using the following steps: Step 1: Acquire X-ray images of complex, irregularly shaped workpieces using an X-ray imaging system; Step 2: Extract gradient information from the X-ray image, construct the gradient field and contrast field of the X-ray image based on the gradient information, and then calculate the local deviation of the X-ray image; Step 3: Calculate the adaptive magnification factor of the contrast field based on the local deviation, and calculate the enhanced contrast field based on the adaptive magnification factor; Step 4: Construct an energy functional between the enhanced contrast field and the target image, and solve the energy functional using fractional variational method and gradient descent method to obtain the iterative form of the target image; Step 5: Perform a logarithmic transformation on the X-ray image to obtain the initial value for iteration. Then, substitute the initial value into the iterative form of the target image and obtain the target image through iteration. In step two, the calculation steps for local deviations are as follows: First, the least squares method is used in the local region. Fit a smooth plane; Among them, local areas It is symmetrical, and in local areas Included in the image domain middle, and Representing local areas respectively The x and y coordinates of the center point; Let the smooth plane be denoted as: ； In the formula: Represents a smooth plane; , , All represent coefficients; At the same time, a smooth plane satisfies: ； In the formula: Represents the actual plane; Then, the local deviation is calculated using the following formula: ； In the formula: Indicates local deviation; Indicates local area The set of discrete points after uniform sampling, and This represents the cardinality of the set.

2. The contrast enhancement method for X-ray images of complex irregularly shaped workpieces according to claim 1, characterized in that: In step one, the X-ray image is denoted as: ； In the formula: Represents an X-ray image; Represents the image domain; Indicates the width of the X-ray image; Indicates the height of the X-ray image; In step two, the gradient field and contrast field are represented as follows: ； In the formula: Represents the contrast field; Indicates the midpoint of an X-ray image Contrast; Represents the gradient field; Indicates the midpoint of an X-ray image The gradient; In step three, the adaptive amplification factor is calculated using the following formula: ； In the formula: Indicates the adaptive amplification factor; Indicates the magnification factor, and ;parameter The value is 90% of the image gradient magnitude histogram; In step three, the enhanced contrast field is calculated using the following formula: ； In the formula: Indicates an enhanced contrast field; In step four, the energy functional is expressed as follows: ； In the formula: Represents the energy functional; Represents the target image; Represents a fractional operator; parameters Used to balance image enhancement and denoising; Indicates the smoothing term; Indicates the fidelity item; In step five, the logarithmic transformation formula is as follows: ； In the formula: Indicates the initial value for the iteration; This indicates the magnification factor.

Images