A method, device, equipment and storage medium for screening target detection spectrum

Abstract

The present invention discloses a method, device, equipment and storage medium for screening target detection spectrum, which relates to the field of target detection technology. It includes: obtaining the background fluctuation scale of the target, quantifying the background fluctuation scale with clutter and establishing a clutter quantitative characterization model; calculating the signal-to-clutter ratio projection diagram of the target relative to different backgrounds through the clutter quantitative characterization model, and selecting the peak signal-to-clutter ratio spectrum as the preliminary detection spectrum; combining the clutter quantitative characterization model with the space-based detection link to obtain the target detectability characterization model; and screening the preferred detection spectrum in the preliminary detection spectrum through the target detectability characterization model. The present invention takes into account the multi-scale coupling effect of the space-based detection link and its various links, and can solve the problems existing in the multi-scale coupling matching of the detectability performance of space-based targets. On the premise of ensuring the preferred performance of the spectrum, the efficiency of selecting the preferred detection spectrum is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of target detection technology, and in particular to a method, device, equipment and storage medium for screening a target detection spectrum. Background Art

[0002] Space-based optical detection has the advantages of all-day, high timeliness, and long-distance wide-area monitoring. It is an important means for early detection, continuous tracking and positioning of faint targets in the air. However, space-based detection scenes and transmission links are complex, and targets are easily interfered by complex environmental backgrounds such as atmospheric transmission paths, cloud clutter, and surface radiation. Its radiation brightness is likely to be similar to or even weaker than the background. After the target signal is transmitted to the entrance pupil of the detection camera, the camera's own receiving spectrum, geometric scale, and spectrum scale also affect the detectability of the target. The impact of these factors on the detectability of the target is not a simple positive or negative relationship. There are also cross-coupling characteristics between the scales. For example, the higher the geometric resolution, the better the detectability of the target.

[0003] Existing detection spectrum selection methods are mostly based on the differences in target and background radiation characteristics. Different models are established to characterize the target / background differences and achieve spectrum design optimization.

[0004] The defects of the above-mentioned existing technology are: it does not take into account the multi-scale coupling effect of the space-based detection link and its various links, resulting in the selected detection spectrum being difficult to adapt to the high-probability detection of targets under different detection system parameters. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device, equipment and storage medium for screening target detection spectrum bands in response to the above technical problems.

[0006] An embodiment of the present invention provides a method for screening a target detection spectrum, comprising:

[0007] Obtain the background fluctuation scale of the target, quantify the clutter on the background fluctuation scale and establish a clutter quantitative characterization model;

[0008] The signal-to-clutter ratio projection diagram of the target corresponding to different backgrounds is determined through the clutter quantitative characterization model, and the spectrum segment including the peak signal-to-clutter ratio is selected as the primary detection spectrum segment;

[0009] The clutter quantitative characterization model is combined with the space-based detection link to obtain the target detectability characterization model;

[0010] The preferred detection spectrum is screened from the preliminary detection spectrum using the target detectability characterization model.

[0011] In addition, the clutter quantification and establishment of a clutter quantitative characterization model include:

[0012] The SWmetric measurement method is used to quantify the clutter of the background fluctuation scale, and the clutter distribution in different regions is statistically analyzed by the regionalization method. The clutter distribution calculation formula is:

[0013]

[0014] Where N represents the number of regions where the background is divided; σ i represents the variance of the ith region;

[0015] The target signal-to-clutter ratio is determined by clutter distribution to establish a target detectability model, and the clutter quantitative characterization model is established by maximizing the difference between the target detectability model and the background.

[0016] In addition, the target signal-to-noise ratio calculation formula is:

[0017]

[0018] Among them, SCR Obj represents the target signal-to-noise ratio, Ω Obj represents the target area, T Obj represents the target physical energy, σ Set It represents the mean square error of the scene radiation energy at the detection scale.

[0019] In addition, the signal-to-clutter ratio projection of the calculated target relative to different backgrounds includes two cases of 0 km and 10 km.

[0020] In addition, the combination of the clutter quantitative characterization model and the space-based detection link includes:

[0021] Calculate the target signal-to-noise ratio:

[0022]

[0023] Among them, SNR IMG represents the signal-to-noise ratio of the target in the image, T IMG is the energy component of the target in the image;

[0024] During imaging, T IMG With T TAR The relationship is:

[0025]

[0026] Where τ represents the transmittance of the optical system, t inte represents the integration time, η represents the detector photoelectric conversion efficiency, Ω Pro represents the photosensitive element area of ​​the detector, and PSF represents the point spread function of the optical system;

[0027] During the detection process, G Pro=t inte τη

[0028] Among them, G DET represents the detection gain coefficient, τ represents the transmittance of the optical system, t inte represents the integration time, η represents the detector photoelectric conversion efficiency;

[0029] Then T IMG With T TAR The relationship changes to:

[0030]

[0031] Through the space-based detection link, we get:

[0032]

[0033] When the center of the target coincides with the center of the detector, the received target energy is highest:

[0034]

[0035] The geometric dispersion of the target in the image plane is smaller than the geometric scale of the detection pixel, that is, Ω Obj ≤Ω Pro , the above formula is simplified to:

[0036]

[0037] The target signal-to-noise ratio is characterized as:

[0038]

[0039] Among them, SNR Obj Indicates the signal-to-noise ratio of the detected target, SNR Set represents the background signal-to-noise ratio, σ Set It represents the mean square error of the scene radiation energy at the detection scale.

[0040] In addition, the calculation formula for the signal-to-noise ratio in the target detectability characterization model is:

[0041] The scene signal-to-noise ratio is but

[0042]

[0043] in, represents the average signal-to-noise ratio of the target in the image, σ Set Represents the mean square error of the scene radiation energy at the detection scale, SNR Obj represents the signal-to-noise ratio of the detected target, G Pro It represents the detection gain coefficient, COE represents the imaging performance of the optical system, and can be expressed as energy concentration, SNRSet Represents the background signal-to-noise ratio, SCR Obj Represents the signal-to-noise ratio of the detected target.

[0044] In addition, the method of screening the preferred detection spectrum segments from the preliminary detection spectrum segments by using the target detectability characterization model includes:

[0045] Selecting a working spectrum band from the primary detection spectrum band, and determining the central wavelength of the working spectrum band by the peak wavelength of the signal-to-noise ratio;

[0046] The central wavelength of the working spectrum is matched with the actual performance of the detector in terms of spectral scale and geometric scale to obtain the optimal detection spectrum.

[0047] In addition, a device for screening a target detection spectrum includes:

[0048] The scale acquisition module is used to obtain the background fluctuation scale of the target, quantify the background fluctuation scale and establish a clutter quantitative characterization model;

[0049] The detection spectrum segment preliminary selection module is used to determine the signal-to-clutter ratio projection diagram of the target corresponding to different backgrounds through the clutter quantitative characterization model, and select the spectrum segment including the peak signal-to-clutter ratio as the preliminary detection spectrum segment;

[0050] The model building module is used to combine the clutter quantitative characterization model with the space-based detection link to obtain the target detectability characterization model;

[0051] The detection spectrum segment optimization module is used to screen out the preferred detection spectrum segment from the preliminary detection spectrum segments through the target detectability characterization model.

[0052] In addition, a device includes: a memory and a processor, wherein the memory stores readable instructions, and when the readable instructions are executed by the processor, one or more processors execute the above-mentioned target detection spectrum screening method.

[0053] In addition, a storage medium includes one or more non-volatile storage media storing readable instructions, and when the readable instructions are executed by one or more processors, the one or more processors execute the above-mentioned target detection spectrum screening method.

[0054] Compared with the prior art, the above-mentioned method, device, equipment and storage medium for screening target detection spectrum provided by the embodiment of the present invention have the following beneficial effects:

[0055] The present invention quantifies the clutter on the background fluctuation scale and establishes a clutter quantitative characterization model, combines the clutter quantitative characterization model with the space-based detection link to obtain a target detectability characterization model, and selects the preferred detection spectrum segment from the preliminary detection spectrum segment through the target detectability characterization model. Compared with the prior art, the multi-scale coupling effect of the space-based detection link and its various links is considered, and the clutter quantitative characterization model is combined with the space-based detection link, which can solve the multi-scale coupling matching problem of the detectability performance of space-based targets, and greatly improves the efficiency of selecting the preferred detection spectrum segment while ensuring the performance of the spectrum segment optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flowchart of a method for screening a target detection spectrum provided in one embodiment;

[0057] Figure 2 A projection diagram showing changes in the signal-to-clutter ratio of three detection targets under different backgrounds at 0 km in a method for screening a target detection spectrum provided in an embodiment;

[0058] Figure 3 This is a projection diagram of the changes in the signal-to-noise ratio of three detection targets under different backgrounds at 10 km in a method for screening target detection spectrum bands provided in an embodiment. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] In one embodiment, a method for screening a target detection spectrum is provided, such as Figure 1 As shown, including:

[0061] Step 1: Use the SW (Schmieder and Weathersby) metric method to measure clutter, establish a quantitative characterization model for background clutter, and further establish a target detectability characterization model;

[0062] The specific steps for the initial selection of target detection spectrum are as follows:

[0063] (1) Clutter quantification is achieved based on the background fluctuation scale. The quantification adopts the SW (Schmieder and Weathersby) metric method and adopts the regionalization idea to statistically analyze the clutter distribution in different regions.

[0064] The clutter distribution calculation formula is:

[0065]

[0066] Where: N represents the number of regions where the background is divided; σ i represents the variance of the ith region.

[0067] (2) Based on the measurement of spatial clutter in complex backgrounds, the target signal-to-clutter ratio is calculated to establish a target detectability model, and a clutter quantitative characterization model is established by maximizing the difference between the target and the background.

[0068] The target signal-to-noise ratio is defined as follows:

[0069]

[0070] Where: SCR Obj represents the target signal-to-noise ratio, Ω Obj represents the target area, T Obj represents the target physical energy, σ Set It represents the mean square error of the scene radiation energy at the detection scale.

[0071] Step 2: Based on the above model, calculate the signal-to-clutter ratio projection diagram of various targets relative to different backgrounds to obtain the preliminary detection spectrum;

[0072] Based on the known target and background radiation characteristics, the projection diagrams of signal-to-clutter ratio changes of various detection targets under different backgrounds are calculated, including 0km and 10km, such as Figure 2 and Figure 3 shown.

[0073] The spectrum segment including the peak signal-to-noise ratio is selected as the primary spectrum segment.

[0074] Step 3: Establish a target detectability characterization model in combination with the space-based detection link

[0075] Since there is no correlation between the target and the actual background, and the target detection process is performed in the image, the target signal-to-noise ratio is used to characterize the target energy in the image:

[0076]

[0077] Where: SNR IMG represents the signal-to-noise ratio of the target in the image, T IMG is the energy component of the target in the image.

[0078] During the imaging process, due to the sensitivity and noise of the detector, the target detectability performance is reduced. Considering the optical dispersion and discrete sampling of the detector in the imaging process, T IMG With T TAR The relationship can be expressed as:

[0079]

[0080] Where: τ represents the transmittance of the optical system, t inte represents the integration time, η represents the detector photoelectric conversion efficiency, Ω Pro represents the photosensitive element area of ​​the detector, and PSF represents the point spread function of the optical system.

[0081] In the detection process, it is reflected as the gain of target energy. Here, it is assumed that:

[0082] G Pro =t inte τη;

[0083] Where: G Pro represents the detection gain coefficient, τ represents the transmittance of the optical system, t inte represents the integration time, and η represents the detector photoelectric conversion efficiency. In the space-based detection system, the imaging performance of the optical system can be characterized by the energy concentration (CoE), and then:

[0084]

[0085] It is easy to see that when the center of the target coincides with the center of the detector, the received target energy is the highest:

[0086]

[0087] Taking the oversampling system as an example, the geometric dispersion of the target in the image plane is smaller than the geometric scale of the detection pixel, that is, Ω Obj ≤Ω Pro , then the above formula is simplified to:

[0088]

[0089] Combining the above formulas, the target signal-to-noise ratio can be expressed as:

[0090]

[0091] Among them, SNR Obj Indicates the signal-to-noise ratio of the detected target, SNR Set represents the background signal-to-noise ratio, σ Set It represents the mean square error of the scene radiation energy at the detection scale.

[0092] Here, the scene signal-to-noise ratio is set Then the target detectability representation model, that is, formula (8), can be simplified as:

[0093]

[0094] in, represents the average signal-to-noise ratio of the target in the image, σ Set Represents the mean square error of the scene radiation energy at the detection scale, SNRObj represents the signal-to-noise ratio of the detected target, G Pro It represents the detection gain coefficient, COE represents the imaging performance of the optical system, and can be expressed as energy concentration, SNR Set Represents the background signal-to-noise ratio, SCR Obj Represents the signal-to-noise ratio of the detected target.

[0095] Step 4: Based on the preliminary spectral data set, select the detection spectrum based on the parameters of the typical space-based platform target detection system.

[0096] A working spectrum band is selected from the preliminary spectrum band set, and the central wavelength of the working spectrum band is determined by selecting the signal-to-noise ratio peak wavelength; the signal-to-noise ratio calculation is shown in Table 1.

[0097] Table 1 Signal-to-noise ratio results at different central wavelengths

[0098]

[0099] The matching design of spectral scale and geometric scale is carried out in combination with the actual performance of the detector. On the basis of optimizing the central wavelength and spectral width of the detection spectrum, combined with practical applications, the matching relationship between the detection space scales is studied, so as to give the detectable spatial resolution of the target. The spectral width is determined to be 50nm, and the calculation results of each detection spectrum under different ground resolution conditions are shown in Table 2.

[0100] Table 2 lists the signal-to-noise ratios of the detection spectrum at different ground resolutions (10m, 15m, 20m, 25m, 30m, 40m, 50m) when the central wavelength of the detection spectrum is 2.76μm, 3.34μm, and 4.26μm, respectively.

[0101] Table 2 Signal-to-noise ratio results at different resolutions

[0102]

[0103] A device for screening a target detection spectrum, comprising:

[0104] The scale acquisition module is used to obtain the background fluctuation scale of the target, quantify the background fluctuation scale and establish a clutter quantitative characterization model;

[0105] A detection spectrum segment preliminary selection module is used to determine the signal-to-clutter ratio projection diagram of the target corresponding to different backgrounds through a clutter quantitative characterization model, and select a spectrum segment including a peak signal-to-clutter ratio as a preliminary detection spectrum segment;

[0106] The model building module is used to combine the clutter quantitative characterization model with the space-based detection link to obtain the target detectability characterization model;

[0107] The detection spectrum segment optimization module is used to screen out the preferred detection spectrum segment from the preliminary detection spectrum segments through the target detectability characterization model.

[0108] A device includes: a memory and a processor, wherein readable instructions are stored in the memory, and when the readable instructions are executed by the processor, one or more processors execute a method for screening a target detection spectrum.

[0109] A storage medium includes one or more non-volatile storage media storing readable instructions. When the readable instructions are executed by one or more processors, the one or more processors execute a method for screening a target detection spectrum.

[0110] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for screening a target detection spectrum, characterized in that: include: Obtain the background fluctuation scale of the target, quantify the clutter on the background fluctuation scale and establish a clutter quantitative characterization model; The clutter quantification and establishment of a clutter quantitative characterization model comprises: The SWmetric measurement method is used to quantify the clutter of the background fluctuation scale, and the clutter distribution in different regions is statistically analyzed by the regionalization method. The clutter distribution calculation formula is: Among them, C is the clutter distribution, N is the number of regions where the background is divided, and σ i represents the variance of the ith region; The target signal-to-clutter ratio is determined through clutter distribution to establish a target detectability model, and a clutter quantitative characterization model is established by maximizing the difference between the target detectability model and the background; The signal-to-clutter ratio projection diagram of the target corresponding to different backgrounds is determined through the clutter quantitative characterization model, and the spectrum segment including the peak signal-to-clutter ratio is selected as the primary detection spectrum segment; The clutter quantitative characterization model is combined with the space-based detection link to obtain the target detectability characterization model; The combining of the clutter quantitative characterization model with the space-based detection link includes: Calculate the target signal-to-noise ratio: Among them, SNR IMG represents the target signal-to-noise ratio, T IMG is the energy component of the target in the image; During imaging, T IMG With T TAR The relationship is: Where τ represents the transmittance of the optical system, t inte represents the integration time, η represents the detector photoelectric conversion efficiency, Ω Pro represents the photosensitive element area of ​​the detector, and PSF represents the point spread function of the optical system; During the detection process, G Pro =t inte τη Among them, G Pro represents the detection gain coefficient, τ represents the transmittance of the optical system, t inte represents the integration time, η represents the detector photoelectric conversion efficiency, and CoE is used to represent the energy concentration of the optical system imaging performance; Then T IMG With T TAR The relationship changes to: Through the space-based detection link, we get: When the center of the target coincides with the center of the detector, the received target energy is highest: The geometric dispersion of the target in the image plane is smaller than the geometric scale of the detection pixel, that is, Ω Obj ≤Ω Pro , the above formula is simplified to: The target signal-to-noise ratio is characterized as: Among them, SNR Obj Indicates the signal-to-noise ratio of the detected target, SNR Set represents the background signal-to-noise ratio, σ Set It represents the mean square error of the scene radiation energy at the detection scale; The calculation formula of the signal-to-noise ratio in the target detectability characterization model is obtained as follows: The scene signal-to-noise ratio is but in, represents the average signal-to-noise ratio of the target in the image, σ Set Represents the mean square error of the scene radiation energy at the detection scale, SNR Obj represents the signal-to-noise ratio of the detected target, G Pro represents the detection gain coefficient, CoE represents the imaging performance of the optical system, and the energy concentration can be used, SNR Set Represents the background signal-to-noise ratio, SCR Obj Indicates the signal-to-noise ratio of the detected target; The preferred detection spectrum is screened from the preliminary detection spectrum using the target detectability characterization model.

2. A method for screening target detection spectrum according to claim 1, characterized in that: The target signal-to-noise ratio calculation formula is: Among them, SCR Obj represents the target signal-to-noise ratio, Ω Obj represents the target area, T Obj represents the target physical energy, σ Set It represents the mean square error of the scene radiation energy at the detection scale.

3. The method for screening a target detection spectrum according to claim 1, characterized in that: The signal-to-clutter ratio projection diagrams of the determined target corresponding to different backgrounds include two cases of 0km and 10km.

4. The method for screening a target detection spectrum according to claim 1, characterized in that: The method of screening the preferred detection spectrum segments from the preliminary detection spectrum segments by using the target detectability characterization model comprises: Selecting a working spectrum band from the primary detection spectrum band, and determining the central wavelength of the working spectrum band by the peak wavelength of the signal-to-noise ratio; The central wavelength of the working spectrum is matched with the actual performance of the detector in terms of spectral scale and geometric scale to obtain the optimal detection spectrum.

5. A device for screening a target detection spectrum, characterized in that: include: The scale acquisition module is used to obtain the background fluctuation scale of the target, quantify the background fluctuation scale and establish a clutter quantitative characterization model; The clutter quantification and establishment of a clutter quantitative characterization model comprises: The SWmetric measurement method is used to quantify the clutter of the background fluctuation scale, and the clutter distribution in different regions is statistically analyzed by the regionalization method. The clutter distribution calculation formula is: Among them, C is the clutter distribution, N is the number of regions where the background is divided, and σ i represents the variance of the ith region; The target signal-to-clutter ratio is determined through clutter distribution to establish a target detectability model, and a clutter quantitative characterization model is established by maximizing the difference between the target detectability model and the background; A detection spectrum segment preliminary selection module is used to determine the signal-to-clutter ratio projection diagram of the target corresponding to different backgrounds through a clutter quantitative characterization model, and select a spectrum segment including a peak signal-to-clutter ratio as a preliminary detection spectrum segment; The model building module is used to combine the clutter quantitative characterization model with the space-based detection link to obtain the target detectability characterization model; The combining of the clutter quantitative characterization model with the space-based detection link includes: Calculate the target signal-to-noise ratio: Among them, SNR IMG represents the target signal-to-noise ratio, T IMG is the energy component of the target in the image; During imaging, T IMG With T TAR The relationship is: Where τ represents the transmittance of the optical system, t inte represents the integration time, η represents the detector photoelectric conversion efficiency, Ω Pro represents the photosensitive element area of ​​the detector, and PSF represents the point spread function of the optical system; During the detection process, G Pro =t inte τη Among them, G Pro represents the detection gain coefficient, τ represents the transmittance of the optical system, t inte represents the integration time, η represents the detector photoelectric conversion efficiency, and CoE is used to represent the energy concentration of the optical system imaging performance; Then T IMG With T TAR The relationship changes to: Through the space-based detection link, we get: When the center of the target coincides with the center of the detector, the received target energy is highest: The geometric dispersion of the target in the image plane is smaller than the geometric scale of the detection pixel, that is, Ω Obj ≤Ω Pro , the above formula is simplified to: The target signal-to-noise ratio is characterized as: Among them, SNR Obj Indicates the signal-to-noise ratio of the detected target, SNR Set represents the background signal-to-noise ratio, σ Set It represents the mean square error of the scene radiation energy at the detection scale; The calculation formula of the signal-to-noise ratio in the target detectability characterization model is obtained as follows: The scene signal-to-noise ratio is but in, represents the average signal-to-noise ratio of the target in the image, σ Set Represents the mean square error of the scene radiation energy at the detection scale, SNR Obj represents the signal-to-noise ratio of the detected target, G Pro represents the detection gain coefficient, CoE represents the imaging performance of the optical system, and the energy concentration can be used, SNR Set Represents the background signal-to-noise ratio, SCR Obj Indicates the signal-to-noise ratio of the detected target; The detection spectrum segment optimization module is used to screen out the preferred detection spectrum segment from the preliminary detection spectrum segments through the target detectability characterization model.

6. A device, characterized in that: include: A memory and a processor, wherein readable instructions are stored in the memory, and when the readable instructions are executed by the processor, one or more processors execute the method for screening target detection spectrum according to any one of claims 1-4.

7. A storage medium, characterized in that: It comprises one or more non-volatile storage media storing readable instructions, and when the readable instructions are executed by one or more processors, the one or more processors execute the method for screening the target detection spectrum according to any one of claims 1-4.

Images