Point target infrared quantitative processing method based on diffusion model fitting

Abstract

The invention relates to the technical field of photoelectricity, in particular to a point target infrared quantitative processing method based on diffusion model fitting. The embodiment of the invention provides a point target infrared quantitative processing method based on diffusion model fitting, and the method comprises the steps: carrying out the modeling of a diffusion process of a point target through a Gaussian formula, and obtaining a target formula; setting a random noise formula, and establishing a background radiation formula according to the random noise formula; performing fitting solution on the target formula by using measurement data and the background radiation formula to obtain a variable of the target formula; and substituting the variable into the target formula, and obtaining the radiation intensity of the point target through characteristic inversion. The embodiment of the invention provides a point target infrared quantitative processing method based on diffusion model fitting, and the measurement precision of a point target is improved.

Description

Technical Field

[0001] The present invention relates to the field of optoelectronic technologies, and particularly to a method for infrared quantitative processing of point targets based on diffusion model fitting. Background Art

[0002] In actual range measurements, airborne targets are often relatively far from the optical system. For example, when measuring radiation characteristics, the target is generally about 1×10 3 km away, and the image formed by the target passing through the optical system is generally a point target, that is, the area of the actual image on the detector surface is smaller than one pixel of the detector. In this case, due to the diffraction of the optical system, its energy is diffused. For this situation, when using traditional radiation characteristic measurement methods to measure its radiation characteristics, the accuracy of the radiation characteristic measurement will be affected.

[0003] Therefore, in view of the above deficiencies, the present application proposes a method for measuring the radiation characteristics of point targets based on diffusion model fitting, which improves the measurement accuracy of point targets. Summary of the Invention

[0004] An embodiment of the present invention provides a method for infrared quantitative processing of point targets based on diffusion model fitting, which improves the measurement accuracy of point targets.

[0005] An embodiment of the present invention provides a method for infrared quantitative processing of point targets based on diffusion model fitting, including:

[0006] Modeling the diffusion process of the point target through the Gaussian formula to obtain the target formula;

[0007] Setting a random noise formula, and establishing a background radiation formula according to the random noise formula;

[0008] Using the measurement data and the background radiation formula to perform fitting and solving on the target formula to obtain the variables of the target formula;

[0009] After substituting the variables into the target formula, the radiation intensity of the point target is obtained through characteristic inversion.

[0010] In a possible design, using the measurement data and the background radiation formula to perform fitting and solving on the target formula to obtain the variables of the target formula, including:

[0011] Substituting the measurement data, the background radiation formula, and the target formula into the loss function, and obtaining the variables of the target formula under the limitation of the loss function.

[0012] In a possible design, the loss function is as follows:

[0013]

[0014] Among them, L 测 is the measurement data, L m is the target formula, L bg is the background radiation formula, Γ 目 is the target area, Γ 背 is the background area, and x, y are coordinate values.

[0015] In a possible design, the target area is obtained in the following manner:

[0016] The target area is obtained by using a target detection and tracking algorithm.

[0017] In a possible design, the background area is obtained in the following manner:

[0018] Pixels are expanded outside the target area as the background area.

[0019] In a possible design, the target formula is as follows:

[0020] L m (x,y) = α * exp(-(x - μ x ) 2 / σ x -(y - μ y ) 2 / σ y )

[0021] Among them, x, y are coordinate values, α is the coefficient of the formula, μ x and μ y are the means of x and y respectively.

[0022] In a possible design, the background radiation formula is as follows:

[0023]

[0024] Among them, μ is the mean of the uniform background, is random noise.

[0025] In a possible design, the measurement data is obtained in the following manner:

[0026] Load a calibration file to convert the grayscale image collected by the measurement device into a radiance image;

[0027] Use the radiance images within the target area and the background area as the measurement data.

[0028] The present invention has at least the following beneficial effects compared with the prior art:

[0029] First, establish a model for the diffusion process of point targets through Gauss's formula to obtain the target formula. At this time, the variables in the target formula are not determined. Set the random noise formula and establish the background radiation formula based on the random noise formula. After superimposing the background radiation formula and the target formula, it is the total radiation luminance formula. The measurement data obtained from the test equipment is the total luminance that superimposes the target luminance and the background radiation. Use the obtained measurement data combined with the background radiation formula to fit the target formula to obtain the five variables of the target formula, that is, obtain the exact target formula, and then the radiation intensity of the point target can be inversely obtained through characteristics. Brief Description of the Drawings

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0031] Figure 1 It is a flowchart of a method for infrared quantitative processing of point targets based on diffusion model fitting provided by an embodiment of the present invention;

[0032] Figure 2a It is a schematic diagram of target modeling in the related art provided by an embodiment of the present invention;

[0033] Figure 2b It is a schematic diagram of background modeling in the related art provided by an embodiment of the present invention;

[0034] Figure 2c It is a schematic diagram of superimposed modeling in the related art provided by an embodiment of the present invention. Detailed Description of the Embodiments

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0036] In the description of the embodiments of the present invention, unless otherwise clearly specified and defined, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance; unless otherwise specified or stated, the term "plural" means two or more; the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, an integral connection, or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0037] In the description of this specification, it should be understood that the orientation terms such as "upper" and "lower" described in the embodiments of the present invention are described from the angles shown in the drawings and should not be construed as limiting the embodiments of the present invention. In addition, in the context, it should also be understood that when it is mentioned that an element is connected "above" or "below" another element, it can not only be directly connected "above" or "below" another element, but also be indirectly connected "above" or "below" another element through an intermediate element.

[0038] As described above, when quantitatively measuring the radiation characteristics of a point target, after the point target passes through the optical system of the measuring device, a diffraction effect will occur. After diffraction, it will be diffused into a spot on the imaging target surface and then superimposed on the background. Assuming that the energy distribution after the point target is diffused follows a Gaussian distribution, as Figure 2a shown, to obtain the accurate target energy, it is necessary to integrate the entire Gaussian function. Assuming that the background area is a uniform background, after being collected by the measuring device, a background image is obtained. Due to the influence of the dark current of the imaging circuit (assuming that the dark current is random noise, etc.), the obtained background image is as Figure 2b shown, and the point target image obtained after the target is superimposed on the background is as Figure 2c shown.

[0039] In order to accurately calculate the radiation characteristics of the target, usually the threshold segmentation method is used to segment the target area. The segmentation threshold is generally set to the mean value of the background area plus 5 times the standard deviation of the background area, as shown by the red dashed line in the figure. After segmentation, the target area is the area from D1 to D2, and then the radiation energy of the target is inverted. The radiation energy of the inverted target corresponds to the Gaussian function of the target as the I1 area, and theoretically the radiation energy of the target is equal to I1 + I2 + I3. Currently, the radiation energy of the target calculated by the conventional method is less than the radiation energy of the theoretical target, especially when the target signal-to-noise ratio is small, the error is greater. Therefore, it is necessary to study a new algorithm for inverting the radiation characteristic data of point targets.

[0040] To solve the above problems, please refer to Figure 1 , the embodiments of the present invention provide a point target infrared quantitative processing method based on diffusion model fitting, including:

[0041] Model the diffusion process of a point target through Gauss's formula to obtain the target formula;

[0042] Set the random noise formula and establish the background radiation formula according to the random noise formula;

[0043] Use the measurement data and the background radiation formula to fit and solve the target formula to obtain the variables of the target formula;

[0044] After substituting the variables into the target formula, the radiation intensity of the point target is obtained through characteristic inversion.

[0045] First, model the diffusion process of the point target through Gauss's formula to obtain the target formula. At this time, the variables of the target formula are not determined. Set the random noise formula and establish the background radiation formula according to the random noise formula. After superimposing the background radiation formula and the target formula, it is the total radiation luminance formula, and the measurement data obtained by the test equipment is the total luminance superimposing the target luminance and the background radiation. Use the obtained measurement data combined with the background radiation formula to fit the target formula to obtain 5 variables of the target formula, that is, obtain the exact target formula, and then the radiation intensity of the point target can be obtained through characteristic inversion.

[0046] In some embodiments of the present invention, use the measurement data and the background radiation formula to fit and solve the target formula to obtain the variables of the target formula, including:

[0047] Substitute the measurement data, the background radiation formula and the target formula into the loss function, and find the variables of the target formula under the limitation of the loss function.

[0048] In some embodiments of the present invention, the loss function is as follows:

[0049]

[0050] where L 测 is the measurement data, L m is the target formula, L bg is the background radiation formula, Γ 目 is the target area, Γ 背 is the background area, and x, y are coordinate values.

[0051] In some embodiments of the present invention, the target area is obtained in the following way:

[0052] Use the target detection and tracking algorithm to obtain the target area.

[0053] In some embodiments of the present invention, the background area is obtained in the following way:

[0054] Expand the pixels on the basis of the target area as the background area.

[0055] In some embodiments of the present invention, the target formula is as follows:

[0056] L m (x,y) = α * exp(-(x - μ x ) 2 / σ x -(y - μ y ) 2 / σ y )

[0057] where x and y are coordinate values, α is the coefficient of the formula, μ x and μ y are the means of x and y respectively.

[0058] In some embodiments of the present invention, the background radiation formula is as follows:

[0059]

[0060] where μ is the mean of the uniform background, is random noise.

[0061] The total radiance is L(x, y) = L m + L bg . The dark current of the measuring device and the like introduce image noise effects.

[0062] In some embodiments of the present invention, the measurement data is obtained in the following manner:

[0063] Load the calibration file, and convert the grayscale image collected by the measuring device into a radiance image;

[0064] Use the radiance images in the target area and the background area as the measurement data.

[0065] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A point target infrared quantitative processing method based on diffusion model fitting, characterized in that: include: The diffusion process of the point target is modeled by the Gaussian formula to obtain the target formula; Setting a random noise formula, and establishing a background radiation formula according to the random noise formula; Using the measurement data and the background radiation formula to fit and solve the target formula, and obtain variables of the target formula; After substituting the variables into the target formula, the radiation intensity of the point target is obtained through characteristic inversion.

2. The method according to claim 1, characterized in that The target formula is fitted and solved using the measurement data and the background radiation formula to obtain variables of the target formula, including: The measurement data, the background radiation formula and the target formula are substituted into the loss function, and under the limitation of the loss function, the variables of the target formula are obtained.

3. The method according to claim 2, characterized in that The loss function is as follows: Among them, L 测 is the measured data, L m is the target formula, L bg is the background radiation formula, Γ 目 is the target area, Γ 背 is the background area, and x and y are the coordinate values.

4. The method according to claim 3, characterized in that: The target area is obtained by: The target area is acquired using a target detection and tracking algorithm.

5. The method according to claim 3, characterized in that: The background area is obtained by: Pixels are expanded on the basis of the target area as a background area.

6. The method according to claim 1, characterized in that The target formula is as follows: L m (x,y)=α*exp(-(x-μ x ) 2 / s x -(y-μ y ) 2 / s y ) Among them, x and y are coordinate values, α is the coefficient of the formula, and μ x and μ y are the means of x and y respectively.

7. The method according to claim 1, characterized in that The background radiation formula is as follows: Where μ is the mean value of the uniform background, is random noise.

8. The method according to claim 3, characterized in that The measurement data is obtained in the following manner: Load the calibration file to convert the grayscale image collected by the measurement device into a radiometric brightness image; The radiation brightness images in the target area and the background area are used as the measurement data.

Images