Global image segmentation method based on global Gaussian distribution and statistical norm

Abstract

The invention provides a global image segmentation method based on global Gaussian distribution and a statistical norm, and relates to the technical field of image processing and application, and the method comprises the steps: constructing an energy functional based on the global Gaussian distribution and the statistical norm; the energy functional comprises a level set function used for representing an image segmentation result; acquiring pixel coordinates of the to-be-segmented image; inputting pixel coordinates of the to-be-segmented image into the energy functional, and performing iterative operation on a level set function in the energy functional until a difference value between an energy functional value obtained by the current round of iterative calculation and an energy functional value obtained by the previous round of iterative calculation is smaller than a preset precision threshold value, and ending loop iterative calculation; outputting a final level set function of the to-be-segmented image; and after each iterative computation, if the condition of ending the loop iterative computation is not reached, updating the current level set function, and performing the next round of iterative update by using the updated level set function. According to the scheme, the image segmentation precision can be improved.

Description

Technical Field

[0001] The invention relates to the field of image processing and application technology, and in particular to a global image segmentation method based on global Gaussian distribution and statistical norm. Background Art

[0002] Image is one of the carriers of information in people's daily life. At present, people's exploration of image technology has extended to various aspects and has been widely used in target detection, medical image processing, industry, face recognition and intelligent transportation. Image processing has become very important with the continuous advancement of science and technology, and image segmentation is a very important field in image processing. Image segmentation mainly divides an image into several non-overlapping areas and segments the target area in the image, providing a basis for subsequent image analysis and image understanding. Therefore, how to obtain better image segmentation results and improve the accuracy and efficiency of segmentation has become one of the key issues in image segmentation.

[0003] The commonly used image segmentation method at present is the segmentation method based on local level set, which has a good effect in segmenting images with uneven intensity. The use of local statistical features can improve the segmentation accuracy in the case of uneven intensity. However, since this method loses the global features, it is highly sensitive to the initial position of the contour. At the same time, when the algorithm is processing high-noise images, the existing scheme may also perform poorly, because the noise can cause changes in local intensity, which may be mistakenly captured by the contour, resulting in incorrect segmentation. Therefore, the existing image segmentation scheme has poor accuracy in segmenting images. Summary of the invention

[0004] In view of this, in order to address the above shortcomings, it is necessary to propose a global image segmentation method based on global Gaussian distribution and statistical norm to improve the accuracy of image segmentation.

[0005] In a first aspect, the present invention provides a global image segmentation method based on global Gaussian distribution and statistical norm, comprising:

[0006] An energy functional is constructed based on a global Gaussian distribution and a statistical norm; wherein the energy functional includes a level set function for characterizing image segmentation results;

[0007] Get the pixel coordinates of the image to be segmented;

[0008] Inputting the pixel coordinates of the image to be segmented into the energy functional, and iteratively calculating the level set function in the energy functional based on the gradient descent method, until the difference between the energy functional value calculated in the current round of iteration and the energy functional value calculated in the previous round of iteration is less than a preset accuracy threshold, and then terminating the loop iterative calculation;

[0009] Outputting the final level set function of the image to be segmented;

[0010] Among them, after each iterative calculation, if the condition for ending the loop iterative calculation is not met, the current level set function is updated, and the updated level set function is used for the next round of iterative update.

[0011] Preferably, the energy functional is constructed based on the global Gaussian distribution and the statistical norm, including: starting from the global information of the image, for each pixel in the image to be segmented, a global Gaussian distribution fitting energy is defined using the global mean and variance as variables, and the fitting energy of all pixels is integrated to construct the energy functional.

[0012] Preferably, constructing the energy functional specifically includes:

[0013] From a statistical point of view, the fitting energy is described using global Gaussian distributions with different means and variances, and the following primary energy functional is constructed:

[0014]

[0015] in, It is the probability density function of the global regional intensity defined as a Gaussian distribution, i = 1, 2, Ω1 and Ω2 are the inner and outer regions of the image contour C to be segmented, respectively, then the following energy functional can be obtained:

[0016]

[0017] in, represents the energy functional, φ is the level set function, x is the pixel coordinate in the image to be segmented I(x), μ i (x) is the global mean, is the global variance, is the gray value at x and the average intensity value μ i (x), λ(x) is the degree of deviation from μ i (x), α1 and α2 are fixed equilibrium parameters, and H(φ(x)) is the Heaviside function.

[0018] Preferably, the iterative operation of the level set function in the energy functional based on the gradient descent method includes:

[0019] Step S31: configuring initial iteration parameters; wherein the iteration parameters include initial energy functional values;

[0020] Step S32: Calculate the current energy functional with respect to μ i (x), Partial derivatives of λ(x);

[0021] Step S33: Set the calculated partial derivatives to 0, and use the pixel coordinates of the image to be segmented to calculate μ i (x), The calculated value of λ(x);

[0022] Step S34: The calculated μ i (x), Substitute the calculated value of λ(x) into the current energy functional to obtain the energy functional value;

[0023] Step S35: determine whether the difference between the energy functional value currently calculated and the energy functional value calculated in the previous iteration is less than a preset accuracy threshold; if so, end the iterative operation; otherwise, after updating the level set function in the current energy functional, return to step S32 for execution.

[0024] Preferably, in step S35, updating the level set function in the current energy functional includes:

[0025] Calculating the partial derivative of the current energy functional with respect to the level set function;

[0026] The level set function in the current energy functional is updated using the partial derivative of the current energy functional with respect to the level set function, and the new energy functional formed after the level set function is updated is used as the energy functional for the next iterative operation.

[0027] Preferably, the updating of the level set function in the current energy functional by using the partial derivative of the current energy functional with respect to the level set function comprises:

[0028] The level set function is updated using the following calculation formula:

[0029]

[0030] Among them, Δt is the time step.

[0031] Preferably, after constructing the energy functional, the method further comprises:

[0032] The constructed energy functional is modified using the following regularization term to ensure the smoothness of the level set function during the curve evolution process;

[0033]

[0034] Where p(φ(x)) is the regularization term, is the gradient of the level set function.

[0035] Preferably, the preset accuracy threshold is 10 -3 .

[0036] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute any of the methods described in the first aspect.

[0037] In a third aspect, the present invention provides a computing device, including a memory and a processor, wherein executable code is stored in the memory, and when the processor executes the executable code, any method described in the first aspect is implemented.

[0038] It can be seen from the above technical scheme that when the global image is segmented based on this scheme, the energy functional is first constructed based on the global Gaussian distribution and the statistical norm, and then the pixel coordinates of the image to be segmented are obtained and input into the energy functional. The level set function in the energy functional is further iterated based on the gradient descent method until the difference between the energy functional value obtained by the current iteration and the energy functional value obtained in the previous round is less than the preset accuracy threshold, then the loop iteration calculation is terminated, and the final level set function of the image to be segmented is output. It can be seen that in order to solve the problem of high sensitivity to the initial position and noise of the contour due to the lack of global features, this scheme starts from the global information of the image and constructs the energy functional based on the probability density function of the global regional intensity of the Gaussian distribution with statistical regularity characteristics. It can not only consider the features of each pixel point of the image to be segmented with the global mean and variance as variables, but also measure the similarity between pixels through the statistical characteristics of the Gaussian distribution, so as to more accurately identify the boundary of the target area under a complex background. Therefore, this scheme can improve the accuracy of image segmentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flowchart of a global image segmentation method based on global Gaussian distribution and statistical norm is provided in an embodiment of the present invention.

[0040] Figure 2 Comparison results of image segmentation with different initial contours.

[0041] Figure 3 Comparison results of image segmentation with different noise levels.

[0042] Figure 4 Segmentation comparison results of natural images. DETAILED DESCRIPTION

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] like Figure 1 As shown, the present invention provides a global image segmentation method based on global Gaussian distribution and statistical norm, which may include the following steps:

[0045] Step 101: constructing an energy functional based on a global Gaussian distribution and a statistical norm; wherein the energy functional includes a level set function for characterizing an image segmentation result;

[0046] Step 102: Obtain pixel coordinates of the image to be segmented;

[0047] Step 103: Input the pixel coordinates of the image to be segmented into the energy functional, and perform iterative calculation on the level set function in the energy functional based on the gradient descent method until the difference between the energy functional value obtained by the current round of iterative calculation and the energy functional value obtained by the previous round of iterative calculation is less than a preset accuracy threshold, and the loop iterative calculation is terminated; wherein, after each iterative calculation, if the condition for terminating the loop iterative calculation is not met, the current level set function is updated, and the updated level set function is used to perform the next round of iterative update;

[0048] Step 104: Output the final level set function of the image to be segmented.

[0049] In order to solve the problem of being highly sensitive to the initial position and noise of the contour due to the lack of global features, the embodiment of this solution starts from the global information of the image and constructs an energy functional based on the probability density function of the global regional intensity of the Gaussian distribution with statistical regularity characteristics. It can not only consider the features of each pixel point of the image to be segmented with the global mean and variance as variables, but also measure the similarity between pixels through the statistical characteristics of the Gaussian distribution, so as to more accurately identify the boundary of the target area in a complex background. Therefore, this solution can improve the accuracy of image segmentation.

[0050] For step 101, an energy functional is constructed based on a global Gaussian distribution and a statistical norm; wherein the energy functional includes a level set function for characterizing an image segmentation result;

[0051] In this embodiment, in order to solve the problem of being highly sensitive to the initial position and noise of the contour due to the lack of global features, it is considered to start from the global information of the image, and for each pixel in the image to be segmented, a new global Gaussian distribution fitting energy is defined using the global mean and variance as variables, and the fitting energy of all pixels is integrated to form the data item of the energy function, that is, the energy functional. Specifically, when constructing the energy functional, the fitting energy can be described from a statistical perspective using global Gaussian distributions with different means and variances, and the following primary energy functional is constructed:

[0052]

[0053] Among them, p i ,x(I(x),μ i (x), λ(x)) is the probability density function of the global regional intensity defined as a Gaussian distribution, where Ω1 and Ω2 are the inner and outer areas of the image contour C to be segmented, respectively. In this way, the following energy functional can be obtained:

[0054]

[0055] in, represents the energy functional, φ is the level set function, x is the pixel coordinate in the image to be segmented I(x), μ i (x) is the global mean, is the global variance, is the gray value at x and the average intensity value μ i (x), λ(x) is the degree of deviation from μ i (x), α1 and α2 are fixed equilibrium parameters, and H(φ(x)) is the Heaviside function.

[0056] For step 102, the pixel coordinates of the image to be segmented are obtained;

[0057] In this step, first obtain the image to be segmented, then construct the pixel coordinate system of the image to be segmented, and then obtain the pixel coordinates of the image to be segmented. For example, a two-dimensional coordinate system uv with pixels as the unit is established with the upper left corner of the image to be segmented as the origin. The horizontal coordinate u and the vertical coordinate v of the pixel are the column number and row number of the image array respectively. The unit of the pixel coordinate system uv is pixel, which is a discrete image coordinate or pixel coordinate, and the origin is in the upper left corner of the image.

[0058] For step 103, the pixel coordinates of the image to be segmented are input into the energy functional, and the level set function in the energy functional is iteratively calculated based on the gradient descent method until the difference between the energy functional value calculated in the current round of iteration and the energy functional value calculated in the previous round of iteration is less than the preset accuracy threshold, and the loop iterative calculation is terminated; step 104, the final level set function of the image to be segmented is output;

[0059] In this step, the energy functional value obtained after each iterative operation is compared with the energy functional value obtained by the previous iterative calculation to determine whether the difference between the two is less than the preset accuracy threshold. If so, it means that with the progress of the iterative operation, the energy functional has no longer continued to decrease, that is, the iterative operation has met the accuracy requirements, and the output level set function can meet the segmentation accuracy requirements; if not, it means that the iterative operation has not met the accuracy requirements, and then the iterative operation continues. Specifically, when performing iterative operations on the level set function in the energy functional based on the gradient descent method, step 103 can be implemented in the following way:

[0060] Step S31: configuring initial iteration parameters; wherein the iteration parameters include initial energy functional values;

[0061] Step S32: Calculate the current energy functional with respect to μ i (x), Partial derivatives of λ(x);

[0062] Step S33: Set the calculated partial derivatives to 0, and use the pixel coordinates of the image to be segmented to calculate μ i (x), The calculated value of λ(x);

[0063] Step S34: The calculated μ i (x), Substitute the calculated value of λ(x) into the current energy functional to obtain the energy functional value;

[0064] Step S35: determine whether the difference between the energy functional value currently calculated and the energy functional value calculated in the previous iteration is less than the preset accuracy value; if so, end the iterative operation; otherwise, after updating the level set function in the current energy functional, return to step S32 for execution.

[0065] In this embodiment, when performing iterative calculations, each iteration parameter is first initialized, such as giving an initial energy functional value, and then calculating the current energy functional with respect to μ i (x), The partial derivative of λ(x). Specifically, the gradient descent flow is: where δ(φ(x)) is the derivative of H(φ(x)), i.e. In this formula, μ i (x), λ(x) is unknown, so consider taking the partial derivative of the energy functional and setting the derivative equal to zero, then we can get:

[0066]

[0067]

[0068] in:

[0069]

[0070] In this way, the calculated μ i (x), The calculated value of λ(x) is brought into the current energy functional to obtain the current energy functional value; the difference between the current energy functional value and the energy functional value obtained in the previous round and the preset precision value are further determined. Of course, if this is the first iteration, the energy functional value of the previous round is the set initial energy functional value. If the difference in the energy functional value is less than the preset precision threshold, for example, the difference is less than 10 -3 , it means that the preset accuracy requirement has been met. As the iterative operation proceeds, the energy functional no longer evolves towards the minimum. At this time, the level set function in the output energy functional is the final level set function of the image to be segmented, that is, the optimal segmentation curve of the image to be segmented is obtained. If the difference in the energy functional value is greater than the preset accuracy threshold, that is, the difference is greater than 10 -3 , it means that the accuracy requirement is not met, then the level set function is updated and the iterative operation is continued.

[0071] Further, in step S35, when updating the level set function in the current energy functional, consider first calculating the partial derivative of the current energy functional with respect to the level set, then using the partial derivative of the current energy functional with respect to the level set function to update the level set function in the current energy functional, and using the new energy functional formed after the level set function is updated as the energy functional for the next iterative operation. Specifically, when using the partial derivative of the current energy functional with respect to the level set function to update the level set function in the current energy functional, the level set function can be updated using the following calculation formula:

[0072]

[0073] Among them, Δt is the time step.

[0074] In addition, in order to avoid the reinitialization problem and ensure that the level set function remains smooth during the evolution of the curve, it is also possible to consider introducing a regularization term to correct the constructed energy functional. The specific regularization term can be obtained as follows:

[0075]

[0076] Among them, p(φ(x)) is the regularization term, is the gradient of the level set function.

[0077] Then, when the energy functional is corrected using the above regularization term, the following new energy functional can be obtained:

[0078]

[0079] In practice, the regularized corrected energy functional can be used for iterative calculations to obtain more accurate image segmentation results.

[0080] The effect of this solution is further explained below with reference to specific application examples.

[0081] In order to verify the effectiveness of this scheme for natural images, consider selecting 6 natural images from the BSDS500 dataset for experiments, and measure the segmentation performance of this scheme according to the following 4 objective evaluation indicators: Jaccard index (JSC) rate, dice similarity coefficient (DSC) rate, recall rate, and precision rate. Specifically, the evaluation indicators are:

[0082]

[0083] Among the above four objective evaluation indicators, G s is the segmented image, G GT is a GT image, |·| represents the number of pixels in the region, TP represents a true positive, FP represents a false positive, and FN represents a false negative.

[0084] (1) Verify the robustness of the initial contour

[0085] By setting initial contours of different sizes and shapes for experiments, the segmentation results are shown in Figure 2 As shown. Figure 2 It can be seen that in (a0)-(a3), the green rectangle represents the initial contour and the red represents the segmentation result; in (b0)-(b3), the green triangle represents the initial contour and the red represents the segmentation result; in (c0)-(c3), the green circle represents the initial contour and the red represents the segmentation result; in (d0)-(d3), the green trapezoid represents the initial contour and the red represents the segmentation result. The entire segmentation result shows that no matter the shape of the initial contour is a rectangle, circle, triangle or trapezoid, or the initial contour contains one target area, two target areas, three target areas, or even no target area, this scheme can accurately segment the image.

[0086] (2) Robustness to noise

[0087] In order to verify the robustness of this scheme to noise, it is verified from two aspects: subjective evaluation and objective evaluation.

[0088] 1) Subjective evaluation

[0089] The initial contour is set to a circle, a real image is selected, and Gaussian noise with a mean of 0, variances of 0.015, 0.02, 0.025, and 0.03, and salt and pepper noise of 0.01, 0.012, 0.014, and 0.016 are added to the image in sequence. The segmentation results are shown in the figure below. Figure 3 As shown in the figure, (a0)-(a3) are the segmentation results of adding different levels of Gaussian noise to the image, and (b0)-(b3) are the segmentation results of adding different levels of salt and pepper noise to the image. Figure 3 The segmentation results show that this scheme can achieve image segmentation of different levels of Gaussian noise and different levels of salt and pepper noise, and has good robustness to Gaussian noise and salt and pepper noise.

[0090] like Figure 4 The segmentation results of the natural image shown in FIG. 1 are the segmentation results of six different scenes without adding noise, and (b1) to (b6) are the segmentation results of six different scenes with adding different noises. Figure 4 It can be found that this scheme can realize the segmentation of natural images in different scenes, and can also realize the segmentation of noisy natural images in different scenes, thus verifying the feasibility and effectiveness of this scheme for natural image segmentation.

[0091] 2) Objective evaluation

[0092] In order to better analyze the experimental results, four objective evaluation indicators are used to evaluate the above Figure 4 An objective evaluation was conducted, and the evaluation results are shown in Table 1 below:

[0093] Table 1

[0094]

[0095] Table 1 shows the objective evaluation results of natural image segmentation by this scheme. Among them, the value range of the four objective indicators is [0,1]. The closer the objective evaluation value is to 1, the better the segmentation effect is, and the closer it is to 0, the worse the segmentation effect is. As can be seen from Table 1, for the DSC evaluation index, the segmentation accuracy is in the range of 97.79% to 99.48%; for the Precision index, the segmentation accuracy is in the range of 98.67% to 99.99%; for the Recall evaluation index, the segmentation accuracy is in the range of 97.16% to 99.36%; for the JSC index, the segmentation accuracy is in the range of 95.68% to 99.14%, which further verifies the effectiveness of this scheme in image segmentation.

[0096] The present specification also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute a method in any one of the embodiments in the specification.

[0097] The present specification also provides a computing device, including a memory and a processor, wherein executable codes are stored in the memory, and when the processor executes the executable codes, a method in any embodiment of the present specification is implemented.

[0098] The modules or units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. The above disclosure is only the preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Those skilled in the art can understand that all or part of the processes of the above embodiment are implemented, and the equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A global image segmentation method based on global Gaussian distribution and statistical norm, characterized in that: include: An energy functional is constructed based on a global Gaussian distribution and a statistical norm; wherein the energy functional includes a level set function for characterizing image segmentation results; Get the pixel coordinates of the image to be segmented; Inputting the pixel coordinates of the image to be segmented into the energy functional, and iteratively calculating the level set function in the energy functional based on the gradient descent method, until the difference between the energy functional value calculated in the current round of iteration and the energy functional value calculated in the previous round of iteration is less than a preset accuracy threshold, and then terminating the loop iterative calculation; Outputting the final level set function of the image to be segmented; Among them, after each iterative calculation, if the condition for ending the loop iterative calculation is not met, the current level set function is updated, and the updated level set function is used for the next round of iterative update.

2. The global image segmentation method based on global Gaussian distribution and statistical norm according to claim 1, characterized in that: The energy functional is constructed based on the global Gaussian distribution and the statistical norm, including: starting from the global information of the image, for each pixel in the image to be segmented, a global Gaussian distribution fitting energy is defined using the global mean and variance as variables, and the fitting energy of all pixels is integrated to construct the energy functional.

3. The global image segmentation method based on global Gaussian distribution and statistical norm according to claim 2, characterized in that: Constructing the energy functional specifically includes: From a statistical point of view, the fitting energy is described using global Gaussian distributions with different means and variances, and the following primary energy functional is constructed: in, It is the probability density function of the global regional intensity defined as a Gaussian distribution, i = 1, 2, Ω1 and Ω2 are the inner and outer regions of the image contour C to be segmented, respectively, then the following energy functional can be obtained: in, represents the energy functional, φ is the level set function, x is the pixel coordinate in the image to be segmented I(x), μ i (x) is the global mean, is the global variance, is the gray value at x and the average intensity value μ i (x), λ(x) is the degree of deviation from μ i (x), α1 and α2 are fixed equilibrium parameters, and H(φ(x)) is the Heaviside function.

4. The global image segmentation method based on global Gaussian distribution and statistical norm according to claim 3, characterized in that: The iterative operation of the level set function in the energy functional based on the gradient descent method includes: Step S31: configuring initial iteration parameters; wherein the iteration parameters include initial energy functional values; Step S32: Calculate the current energy functional with respect to μ i (x), Partial derivatives of λ(x); Step S33: Set the calculated partial derivatives to 0, and use the pixel coordinates of the image to be segmented to calculate μ i (x), The calculated value of λ(x); Step S34: The calculated μ i (x), Substitute the calculated value of λ(x) into the current energy functional to obtain the energy functional value; Step S35: determine whether the difference between the energy functional value currently calculated and the energy functional value calculated in the previous iteration is less than a preset accuracy threshold; if so, end the iterative operation; otherwise, after updating the level set function in the current energy functional, return to step S32 for execution.

5. The global image segmentation method based on global Gaussian distribution and statistical norm according to claim 4, characterized in that: In step S35, the updating of the level set function in the current energy functional includes: Calculating the partial derivative of the current energy functional with respect to the level set function; The level set function in the current energy functional is updated using the partial derivative of the current energy functional with respect to the level set function, and the new energy functional formed after the level set function is updated is used as the energy functional for the next iterative operation.

6. The global image segmentation method based on global Gaussian distribution and statistical norm according to claim 5, characterized in that: The updating of the level set function in the current energy functional by using the partial derivative of the current energy functional with respect to the level set function comprises: The level set function is updated using the following calculation formula: Among them, Δt is the time step.

7. The global image segmentation method based on global Gaussian distribution and statistical norm according to claim 3, characterized in that: After constructing the energy functional, further comprising: The constructed energy functional is modified using the following regularization term to ensure the smoothness of the level set function during the curve evolution process; Where p(φ(x)) is the regularization term, is the gradient of the level set function.

8. The global image segmentation method based on global Gaussian distribution and statistical norm according to claim 1, characterized in that: The preset accuracy threshold is 10 -3 .

9. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method according to any one of claims 1 to 8.

10. A computing device comprising a memory and a processor, wherein the memory stores executable codes, and when the processor executes the executable codes, the method according to any one of claims 1 to 8 is implemented.

Images