Methods, apparatus, equipment and media for evaluating interference effects based on similarity evaluation

Abstract

This invention relates to the field of infrared image evaluation technology, and particularly to a method, apparatus, device, and medium for evaluating interference effects based on similarity assessment. The method includes: acquiring different infrared images; evaluating the degree of influence of particulate matter on the infrared images by calculating the similarity of infrared interference effects between different infrared images; wherein calculating the similarity of infrared interference effects includes: determining the interference caused by particulate matter in the infrared images; determining the overall interference similarity between two infrared images; determining the interference matrices of the two infrared images respectively; obtaining the interference distribution similarity between the two infrared images; and determining the final similarity of infrared interference effects based on the overall interference similarity and interference distribution similarity between the two infrared images. This invention enables rapid analysis and evaluation of the infrared interference effects caused by particulate matter swarms passing through detectors.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of infrared image evaluation technology, and in particular to a method, apparatus, device and medium for evaluating interference effects based on similarity evaluation. Background Technology

[0002] When an aircraft is in flight and the infrared detection system is working, some airborne or spaceborne particles often pass through the detector's field of view. These particles can cause infrared interference to the detector. This interference is closely related to factors such as the state and distribution of the particles themselves and their distance from the detector.

[0003] Currently, the impact of infrared interference caused by particulate matter on images is mostly assessed subjectively, and there is no reliable quantitative assessment method yet. Summary of the Invention

[0004] To address the difficulty of accurately assessing the impact of infrared interference caused by particulate matter using existing technologies, this invention provides a method, apparatus, electronic device, and storage medium for evaluating interference effects based on similarity assessment. This method can calculate the similarity of infrared interference effects between two infrared images, thereby enabling rapid analysis and assessment of the impact of infrared interference caused by particle groups passing through the detector.

[0005] In a first aspect, embodiments of the present invention provide a method for evaluating interference effects based on similarity assessment, including:

[0006] Acquire different infrared images;

[0007] The impact of particulate matter on infrared images is evaluated by calculating the similarity of infrared interference effects between different infrared images.

[0008] The similarity of infrared interference effects between two different infrared images is calculated using the following method:

[0009] Two infrared images were preprocessed to determine the interference caused by particulate matter in the infrared images;

[0010] The overall interference similarity between the two infrared images is determined based on the total number, total scale, and total brightness of interference in each image.

[0011] Based on the interference caused by particulate matter in infrared images and the corresponding statistical rules, interference matrices for two infrared images are determined respectively. The interference matrix includes the interference pattern, field scale, and brightness-related classification statistical vectors. The elements of the classification statistical vectors are the number of interferences counted in intervals according to the corresponding statistical rules.

[0012] Based on the interference matrix of the two infrared images, the similarity of the interference distribution between the two infrared images is obtained;

[0013] The final infrared interference effect similarity is determined based on the overall interference similarity and the interference distribution similarity between the two infrared images.

[0014] Optionally, determining interference caused by particulate matter in the infrared image includes:

[0015] The judgment is based on the pixel maximum and minimum values ​​of a single infrared image; if L min ≤0.3L cut +L cut And L max ≤0.3L min +L min If the interference quantity in the infrared image is 0, then L is considered to be 0. min L represents the minimum pixel value of an infrared image. max L represents the maximum pixel value of an infrared image. cut This represents the detectable threshold of the infrared detector; if L min >0.3L cut +L cut If L min ≤0.3L cut +L cut And L max >0.3L min +L min If so, proceed with the next steps;

[0016] Using ΔL = 0.3L min +L min As a threshold, the infrared image is binarized by setting the pixel values ​​of the infrared image that exceed the threshold value ΔL to 1 and the other pixel values ​​to 0, thus obtaining a binarized image.

[0017] Based on the obtained binarized image, the total number of interferences in the infrared image is determined by detecting the connectivity of the image.

[0018] The total scale of interference in the infrared image is obtained by counting the number of pixel points whose pixel values ​​exceed the threshold value ΔL.

[0019] The total brightness of the interference in the infrared image is obtained by statistically analyzing the pixel values ​​of pixels whose pixel values ​​exceed the threshold value ΔL.

[0020] Optionally, determining the overall interference similarity between two infrared images includes:

[0021] The total number, total scale, and total brightness of the interference in each of the two infrared images are dimensionless, resulting in two sets of corresponding dimensionless parameters, expressed as follows:

[0022]

[0023]

[0024] Where, N′ g,tol 、D′ g,tol and L′ g,tol N″ represents the total number, total scale, and total brightness of interference in the first infrared image, respectively. g,tol D″ g,tol and L″ g,tol Let [n′, d′, l′] represent the total number, total scale, and total brightness of interference in the second infrared image, respectively. p1 and [n″,d″,l″] p2 These represent the dimensionless parameter groups corresponding to the first and second infrared images, respectively.

[0025] Using the dimensionless parameter set [n′,d′,l′] corresponding to the first infrared image p1 Using this as a benchmark, if the dimensionless parameter set corresponding to the second infrared image is [n″,d″,l″], then... p2 If n″≥n′, d″≥d′, l″≥l′ or n″≤n′, d″≤d′, l″≤l′, then a positive deviation is determined; otherwise, a negative deviation is determined.

[0026] Based on the dimensionless parameter set [n′,d′,l′] p1 and [n″,d″,l″] p2 The overall interference similarity between two infrared images is calculated using the following expression:

[0027]

[0028] in, σ represents the distance between the parameter vectors corresponding to the two infrared images; σ represents the adjustment factor, which is 0.5 if there is a positive deviation and 0.8 if there is a negative deviation.

[0029] Optionally, the step of determining the interference matrices of the two infrared images based on the interference caused by particulate matter in the infrared images and the corresponding statistical rules includes:

[0030] Based on the interference caused by particulate matter in the infrared image, the proportion of each interference is determined to represent the pattern of interference; the proportion is determined by the percentage of the actual area of ​​the interference in the bounding rectangle of the interference.

[0031] Based on the corresponding statistical rules, the percentage is divided into various statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the style.

[0032] Based on the interference caused by particulate matter in the infrared image, the field of view scale of each interference is determined;

[0033] According to the corresponding statistical rules, the statistical intervals of the field of view scale are divided, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the field of view scale.

[0034] The brightness of each interference is determined based on the interference caused by particulate matter in the infrared image;

[0035] Based on the corresponding statistical rules, the brightness is divided into various statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the brightness.

[0036] Optionally, obtaining the interference distribution similarity between the two infrared images based on the interference matrix of the two infrared images includes:

[0037] The interference matrices of the two infrared images are normalized respectively to obtain two normalized matrices n. p1 and n p2 ;

[0038] Based on the two normalized matrices n p1 and n p2 Calculate the Euclidean distance d(n) p1 ,n p2 );

[0039] Based on the obtained Euclidean distance d(n) p1 ,n p2 To calculate the similarity of interference distribution between two infrared images, the expression is:

[0040]

[0041] Optionally, determining the final infrared interference effect similarity based on the overall interference similarity and the interference distribution similarity between the two infrared images includes:

[0042] The final expression for the similarity of the infrared interference effect is obtained through multiplication: η g =f(L)sim(n) p1 ,n p2 ), where f(L) represents the overall interference similarity between the two infrared images, sim(n p1 ,n p2The expression represents the similarity of the interference distribution between two infrared images.

[0043] Optionally, the interference effect assessment method further includes:

[0044] The qualitative evaluation result of the overall similarity is determined based on the overall interference similarity between the two infrared images;

[0045] If the overall interference similarity f(L) between two infrared images is ≥ 0.9, they are determined to be highly similar overall; if 0.7 ≤ f(L) < 0.9, they are determined to be similar overall; if 0.3 ≤ f(L) < 0.7, they are determined to be close overall; if 0.1 ≤ f(L) < 0.3, they are determined to be dissimilar overall; if f(L) < 0.1, they are determined to be highly dissimilar overall.

[0046] Secondly, embodiments of the present invention also provide an interference effect assessment device based on similarity evaluation, comprising:

[0047] The image acquisition module is used to acquire different infrared images;

[0048] The similarity calculation module is used to evaluate the degree of influence of particulate matter on infrared images by calculating the similarity of infrared interference effects between different infrared images;

[0049] The similarity calculation module calculates the similarity of infrared interference effects between two different infrared images in the following way:

[0050] Two infrared images were preprocessed to determine the interference caused by particulate matter in the infrared images;

[0051] The overall interference similarity between the two infrared images is determined based on the total number, total scale, and total brightness of interference in each image.

[0052] Based on the interference caused by particulate matter in infrared images and the corresponding statistical rules, interference matrices for two infrared images are determined respectively. The interference matrix includes the interference pattern, field scale, and brightness-related classification statistical vectors. The elements of the classification statistical vectors are the number of interferences counted in intervals according to the corresponding statistical rules.

[0053] Based on the interference matrix of the two infrared images, the similarity of the interference distribution between the two infrared images is obtained;

[0054] The final infrared interference effect similarity is determined based on the overall interference similarity and the interference distribution similarity between the two infrared images.

[0055] Thirdly, embodiments of the present invention also provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the interference effect evaluation method described in any embodiment of this specification.

[0056] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the interference effect evaluation method described in any embodiment of this specification.

[0057] This invention provides a method, apparatus, electronic device, and storage medium for evaluating interference effects based on similarity assessment. By calculating the similarity of infrared interference effects between different infrared images, this invention evaluates the degree of influence of particulate matter on infrared images, achieving a quantitative evaluation of the impact of infrared interference effects and filling a gap in the prior art. This invention also provides a method for calculating the similarity of infrared interference effects between two different infrared images. First, the overall interference similarity between the two infrared images is determined, then the interference distribution similarity is determined, and finally, the similarity of the infrared interference effects between the two infrared images is obtained. This invention enables rapid analysis and evaluation of the infrared interference effects caused by particulate matter swarms passing through detectors, solving the problem of the inability to quantitatively calculate the similarity of infrared interference caused by particulate matter. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart illustrating the steps of an interference effect assessment method based on similarity evaluation, provided in an embodiment of the present invention.

[0060] Figure 2 This is a flowchart illustrating the steps for calculating the similarity of infrared interference effects between two different infrared images according to an embodiment of the present invention;

[0061] Figures 3(a) and 3(b) show two images affected by infrared interference.

[0062] Figures 4(a), 4(b), and 4(c) show the classification statistics for the patterns, field scales, and brightness of the interference, respectively.

[0063] Figure 5 This is a hardware architecture diagram of an electronic device provided in an embodiment of the present invention;

[0064] Figure 6 This is a structural diagram of an interference effect evaluation device based on similarity evaluation provided in an embodiment of the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0066] Research and analysis of infrared interference caused by particulate matter in space or air on detectors reveals that when a swarm of particles passes through the detector, it creates a near-field interference effect at the detector's aperture, typically including different forms of interference such as solid circles, spots, and rings. The exact strength of the interference at different times and the similarity of the interference effects are currently impossible to quantitatively assess. Therefore, this invention provides a method, apparatus, electronic device, and storage medium for evaluating interference effects based on similarity assessment. By calculating the similarity of infrared interference effects between different infrared images, the degree of influence of particulate matter on infrared images can be evaluated. For example, infrared images of particles entering or leaving the detector's field of view can be used as reference images. The similarity between the infrared image of particles passing through the detector's field of view and the reference image can be used to evaluate the intensity of the influence of particles on infrared images at different times. Alternatively, an infrared image captured by one detector can be used as a reference image. The similarity between the infrared image captured by another detector and the reference image can be used to evaluate whether the interference effects caused by different particles are similar.

[0067] The following describes the specific implementation of the above concept.

[0068] Please refer to Figure 1 and Figure 2 This invention provides a method for evaluating interference effects based on similarity assessment, the method comprising:

[0069] Step 100: Acquire different infrared images;

[0070] Step 200: By calculating the similarity of infrared interference effects between different infrared images, the degree of influence of particulate matter on infrared images is evaluated.

[0071] The similarity of infrared interference effects between two different infrared images is calculated using the following method:

[0072] Step 200-0: Preprocess the two infrared images to determine the interference caused by particulate matter in the infrared images;

[0073] Step 200-2: Determine the overall interference similarity between the two infrared images based on the total number, total scale, and total brightness of interference in each image.

[0074] Step 200-4: Based on the interference caused by particulate matter in the infrared images and the corresponding statistical rules, determine the interference matrices of the two infrared images respectively;

[0075] The interference matrix includes a classification statistical vector corresponding to the interference pattern, field scale, and brightness. The elements of the classification statistical vector are the number of interferences counted in intervals according to the corresponding statistical rules.

[0076] Step 200-6: Based on the interference matrix of the two infrared images, obtain the interference distribution similarity between the two infrared images;

[0077] Step 200-8: Based on the overall interference similarity and the interference distribution similarity between the two infrared images, determine the final infrared interference effect similarity.

[0078] This invention provides a method for evaluating interference effects based on similarity assessment. It can calculate the similarity of infrared interference effects between two infrared images, thereby enabling rapid analysis and evaluation of the impact of particle swarms passing through the detector on infrared interference effects. The embodiments also provide a method for calculating the similarity of infrared interference effects. First, the images are preprocessed. Second, the total number, total scale, and total brightness of interference in each image are calculated to determine the overall interference similarity between the two images. Third, based on statistical criteria for patterns, statistical intervals for pattern classification are determined, and the number of interferences corresponding to each pattern type is calculated, establishing a classification statistical vector to represent the pattern classification results. Based on statistical criteria for field of view scale, statistical intervals for field of view scale classification are determined, and the number of interferences corresponding to each field of view scale type is calculated, establishing a classification statistical vector to represent the field of view scale classification results. Based on statistical criteria for brightness, statistical intervals for brightness classification are determined, and the number of interferences corresponding to each brightness type is calculated, establishing a classification statistical vector to represent the brightness classification results. Then, the similarity of interference distribution between the two images is calculated. Finally, a quantitative characterization of the similarity of infrared interference effects between the two images is obtained. This invention can solve the problem of being unable to quantitatively calculate the similarity of infrared interference caused by particulate matter. Based on the obtained infrared interference similarity, the degree of influence of particulate matter on infrared images can be quantitatively evaluated.

[0079] Since the focus of this invention is on quantitatively calculating the similarity of infrared interference caused by particulate matter, the following description is provided. Figure 2 The steps shown are for quantitatively calculating the similarity of infrared interference caused by particulate matter.

[0080] Optionally, for step 200-0, the two infrared images are preprocessed, including:

[0081] Check if the two infrared images are the same size; if they are different, use interpolation to make the two infrared images the same size.

[0082] The original two infrared images can be characterized as L1(i,j) 1≤i≤N x ′,1≤j≤N y ′,L2(i,j)1≤i≤N x ",1≤j≤N y ", if N x ≠N x "or N y ≠N y ", then take N x =max{N x ′,N x "}, N y =max{N y ′,N y "}, N x and N y Representing the dimensions of the infrared image in the X and Y directions, as a unified evaluation standard, based on N x and N y Interpolate one or two infrared images to obtain two infrared images of the same size, L1(i,j) 1≤i≤N x ,1≤j≤N y and L2(i,j)1≤i≤N x ,1≤j≤N y This is for subsequent processing.

[0083] Optionally, determining interference caused by particulate matter in the infrared image includes:

[0084] Based on a single infrared image (which can be represented as L(i,j) 1≤i≤N) x ,1≤j≤N y The maximum and minimum values ​​of pixels are determined by L; if L min ≤0.3L cut +L cut And L max ≤0.3L min +L min If the interference quantity in the infrared image is 0, then L is considered to be 0. min L represents the minimum number of pixels in a single infrared image. max L represents the maximum pixel value of a single infrared image. cut This represents the detectable threshold of the infrared detector. Below this value, the detected signal is equivalent to noise, making it impossible to distinguish the target and obtain a usable infrared image. If L...min >0.3L cut +L cut If L min ≤0.3L cut +L cut And L max >0.3L min +L min If so, proceed with the next steps;

[0085] Using ΔL = 0.3L min +L min As a threshold, the single infrared image is binarized by setting pixel values ​​exceeding the threshold ΔL to 1 and all other pixel values ​​to 0, resulting in a binarized image, expressed as l(i,j), 1≤i≤N. x ,1≤j≤N j ,in

[0086] Based on the obtained binarized image, the total number N of interferences in the infrared image is determined by detecting the connectivity of the image. g,tol ;

[0087] For pixel points in the infrared image whose pixel values ​​exceed a threshold ΔL (i.e., L(i,j)>ΔL), a statistical analysis is performed to obtain the total scale of interference in the infrared image, expressed as follows:

[0088] For pixels in an infrared image whose pixel values ​​exceed a threshold ΔL, pixel value statistics are performed to obtain the total brightness of interference in the infrared image, expressed as follows:

[0089] Using the above embodiments, interference caused by particulate matter in infrared images can be quickly determined, and the overall interference parameters [N′] of the two infrared images can be obtained. g,tol ,D′ g,tol ,L′ g,tol ] p1 and [N″ g,tol ,D″ g,tol ,L″ g,tol ] p2 If the maximum value in the image is within the threshold fluctuation range, the infrared image fluctuation is caused by detector signal fluctuation, i.e., L. min ≤0.3L cut +L cut And L max ≤0.3L min +L min If the minimum value in the image exceeds the fluctuation range of the threshold value, then it is considered that no infrared interference has occurred, and the amount of infrared interference in the image is 0; if the minimum value in the image exceeds the fluctuation range of the threshold value, i.e., L...min >0.3L cut +L cut If the image as a whole is affected by interference, and the interference is strong, then the image as a whole is a strong interference. These two cases are different from the cases where it is necessary to distinguish each interference for comprehensive evaluation, so infrared interference effect similarity calculation is not required.

[0090] Furthermore, a two-pass scanning method can be used to detect graph connectivity, thereby obtaining the total number of image interferences N. g,tol The specific steps are as follows:

[0091] A1: Traverse the grid in the binary image, i.e., the image points. If there are no elements in the top, left, top-left, and top-right grids, i.e., l(i,j) = 0, then assign a new label to the current grid and increment the label number by 1.

[0092] A2: When an element exists in the top, left, or top-left grid, assign that element's value to the current grid as a label.

[0093] A3: When there are multiple elements in the top, left, and top-left grids, take the lowest value as the label of the current grid.

[0094] A4, traverse each non-zero grid, and use the pre-built disjoint-set data structure to set the value of each grid to the value of the corresponding root grid.

[0095] The more detailed process of detecting graphic connectivity using the two-pass scanning method can be found in existing techniques and will not be elaborated further here. Using the above method, the number of independent interferences in the binarized image can be quickly determined, allowing for the statistical determination of the total number of interferences in a single image.

[0096] Optionally, for step 200-2, determining the overall interference similarity between the two infrared images includes:

[0097] The total number, total scale, and total brightness of the interference in each of the two infrared images are dimensionless, resulting in two sets of corresponding dimensionless parameters, expressed as follows:

[0098]

[0099]

[0100] Where, N′ g,tol 、D′ g,tol and L′ g,tol N″ represents the total number, total scale, and total brightness of interference in the first infrared image, respectively. g,tol D″ g,tol and L″ g,tolLet [n′, d′, l′] represent the total number, total scale, and total brightness of interference in the second infrared image, respectively. p1 and [n″,d″,l″] p2 These represent the dimensionless parameter groups corresponding to the first and second infrared images, respectively. Subscripts p1 and p2 are used to distinguish between the first and second infrared images. n′ and n″ represent the dimensionless parameters corresponding to the total number of interferences in the first and second infrared images, respectively. d′ and d″ represent the dimensionless parameters corresponding to the total scale of interferences in the first and second infrared images, respectively. l′ and l″ represent the dimensionless parameters corresponding to the total brightness of interferences in the first and second infrared images, respectively.

[0101] Using the dimensionless parameter set [n′,d′,l′] corresponding to the first infrared image p1 Using this as a benchmark, if the dimensionless parameter set corresponding to the second infrared image is [n″,d″,l″], then... p2 If n″≥n′, d″≥d′, l″≥l′ or n″≤n′, d″≤d′, l″≤l′, then the second infrared image is determined to have a positive deviation relative to the first infrared image; otherwise, it is determined to have a negative deviation.

[0102] Based on the dimensionless parameter set [n′,d′,l′] p1 and [n″,d″,l″] p2 The overall interference similarity between two infrared images is calculated using the following expression:

[0103]

[0104] in, σ represents the distance between the parameter vectors corresponding to the two infrared images; σ represents the adjustment factor, which is 0.5 if there is a positive deviation and 0.8 if there is a negative deviation.

[0105] Using the above embodiments, different adjustment factors can be used to calculate the overall interference similarity for cases of positive or negative deviation, and finally the corresponding overall interference similarity can be determined.

[0106] Furthermore, the method also includes: determining the overall similarity qualitative evaluation result based on the overall interference similarity between the two infrared images;

[0107] If the overall interference similarity f(L) between two infrared images is ≥ 0.9, the qualitative evaluation result of the overall similarity is determined to be highly similar overall; if 0.7 ≤ f(L) < 0.9, it is determined to be similar overall; if 0.3 ≤ f(L) < 0.7, it is determined to be close overall; if 0.1 ≤ f(L) < 0.3, it is determined to be dissimilar overall; if f(L) < 0.1, it is determined to be highly dissimilar overall.

[0108] Through the above embodiments, it is possible to achieve a qualitative characterization of the overall interference similarity between two infrared images, which can be used for rapid analysis and qualitative evaluation of the infrared interference effect caused by particle swarms passing through the detector.

[0109] Optionally, for step 200-4, the following may be further included:

[0110] Based on the interference caused by particulate matter in the infrared image, the proportion of each type of interference is determined to represent the pattern of interference; the proportion is determined by the percentage of the actual area of ​​the interference within the bounding rectangle of the interference, expressed as follows: N zone =(I max -I min (J) max -J min ), N cell I represents the actual area of ​​the interference. max and I min J represents the maximum and minimum values ​​of the interference's coordinates in the X direction of the infrared image. max and J min This represents the maximum and minimum values ​​of the interference's coordinates in the Y direction of the image;

[0111] Based on the corresponding statistical rules, the percentage is divided into various statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the style.

[0112] Based on the interference caused by particles in the infrared image, the field-of-view scale of each interference is determined, expressed as follows: θ x and θ y N represents the field of view angles in the X and Y directions of the infrared image. x and N y Indicates the dimensions of the infrared image in the X and Y directions;

[0113] According to the corresponding statistical rules, the statistical intervals of the field of view scale are divided, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the field of view scale.

[0114] Based on the interference caused by particulate matter in the infrared image, the brightness of each interference is determined by the following expression: L k This represents the pixel value of the k-th pixel in the infrared image where the interference occurs.

[0115] Based on the corresponding statistical rules, the brightness is divided into various statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the brightness.

[0116] Furthermore, in step 200-4, the statistical rules include dividing the two infrared images into N equally divided statistical intervals based on the magnitude of their corresponding maximum and minimum values. N is a positive integer, and its value can be selected from 8 to 12.

[0117] Using the statistical rules in the above embodiments, the classification interval range is determined by the corresponding maximum and minimum values, which can better adapt to different images. If N is too small, it may not be able to reflect the distribution characteristics of interference; if N is too large, it may lose the meaning of classification statistics.

[0118] In one specific implementation, N = 10, step 200-4 includes:

[0119] Based on the interference caused by particulate matter in infrared images, the proportion of each type of interference is determined to represent the interference pattern; δ can then be obtained. g,zone,p1 (k), 1≤k≤N g,tol,p1 δ g,zone,p2 (k), 1≤k≤N g,tol,p2 , where δ g,zone,p1 (k) and δ g,zone,p2 (k) represents the proportion of the k-th interference in the first and second infrared images, respectively, N g,tol,p1 and N g,tol,p2 These represent the total number of interferences in the first and second infrared images, respectively;

[0120] Based on the corresponding statistical rules, the statistical intervals of the proportion are divided, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the pattern. Among them, the statistical rules for interference patterns include dividing the statistical intervals into 10 equal parts based on the maximum and minimum values ​​of the interference proportion in the two infrared images; that is, taking the maximum proportion δ among all interferences in the two infrared images. max and minimum percentage δ min After determining the range, the proportion is divided into 10 equal parts to form statistical intervals, thus obtaining the proportion range of the m-th interval. Then, count the number of interferences in each interval to obtain N. form,p1 (i),1≤i≤10 and N form,p2 (i), 1≤i≤10, N form,p1 (i) and N form,p2 (i) represents the number of interferences obtained in the i-th statistical interval of the corresponding proportion of the first and second infrared images, which is also the i-th element of the corresponding classification statistical vector;

[0121] Based on the interference caused by particulate matter in the infrared image, the field-of-view scale of each interference is determined; D can then be obtained. g,p1 (k), 1≤k≤N g,tol,p1 D g,p2 (k), 1≤k≤N g,tol,p2 Dg,p1 (k) and D g,p2 (k) represents the field-of-view scale of the k-th interference in the first and second infrared images, respectively;

[0122] Based on the corresponding statistical rules, the field of view scale is divided into statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the field of view scale. The statistical rules for the interference field of view scale include dividing the field of view into 10 equally divided statistical intervals based on the maximum and minimum values ​​of the interference field of view scale in the two infrared images; that is, taking the largest field of view scale D among all interferences in the two infrared images. max and minimum field of view scale D min After determining the range, the field of view is divided into 10 equal parts, resulting in statistical intervals for each interval. The range of the field of view for the m-th interval is then calculated. Then, count the number of interferences in each interval to obtain N. scale,p1 (i),1≤i≤10 and N scale,p2 [i), 1≤i≤10, N scale,p1 (i) and N scale,p2 (i) represents the number of interferences obtained in the i-th statistical interval of the field of view scale corresponding to the first and second infrared images, which is also the i-th element of the corresponding classification statistical vector;

[0123] Based on the interference caused by particulate matter in the infrared image, the brightness of each interference is determined; L can then be obtained. g,p1 (k), 1≤k≤N g,tol,p1 L g,p2 (k), 1≤k≤N g,tol,p2 L g,p1 (k) and L g,p2 (k) represents the brightness of the k-th interference in the first and second infrared images, respectively;

[0124] Based on the corresponding statistical rules, the brightness is divided into various statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the brightness. The statistical rules for interference brightness include dividing the data into 10 equally divided statistical intervals based on the maximum and minimum values ​​of the interference brightness in the two infrared images; that is, taking the maximum brightness L among all interferences in the two infrared images. max and minimum brightness L min After determining the range, the brightness is divided into 10 equal parts to create statistical intervals, resulting in the interference brightness range for the m-th interval. Then, count the number of interferences in each interval to obtain N. energy,p1 (i),1≤i≤10 and N energy,p2 (i), 1≤i≤10, N energy,p1 (i) and N energy,p2(i) represents the number of interferences counted within the i-th statistical interval of the interference brightness corresponding to the first and second infrared images, which is also the i-th element of the corresponding classification statistical vector;

[0125] The final interference matrices of the two images are as follows: and

[0126] Figures 3(a) and 3(b) show two images affected by infrared interference. Figures 4(a), 4(b), and 4(c) show the classification statistics of the interference pattern, field of view scale, and brightness obtained from the two infrared images shown in Figures 3(a) and 3(b), respectively. For ease of display, in Figures 4(a), 4(b), and 4(c), Image1 represents the image shown in Figure 3(a), and Image2 represents the image shown in Figure 3(b).

[0127] Optionally, for step 200-6, the following may be further included:

[0128] The interference matrices of the two infrared images are normalized respectively to obtain two normalized matrices n. p1 and n p2 The expression is:

[0129]

[0130] Where, n form,p1 (i), n scale,p1 (i) and n energy,p1 (i) is matrix n p1 The elements, n form,p2 (i), n scale,p2 (i) and n energy,p2 (i) is matrix n p2 The elements; 1≤i≤N;

[0131] Based on the two normalized matrices n p1 and n p2 Calculate the Euclidean distance d(n) p1 ,n p2 When N = 10, the corresponding expression is:

[0132]

[0133] Based on the obtained Euclidean distance d(n) p1 ,n p2 To calculate the similarity of interference distribution between two infrared images, the expression is:

[0134]

[0135] Using the above embodiments, it is possible to quantitatively calculate the similarity of interference distribution between infrared images.

[0136] Optionally, for step 200-8, the following may be further included:

[0137] The final expression for the similarity of the infrared interference effect is obtained through multiplication: η g =f(L)sim(n) p1 ,n p2 ), where f(L) represents the overall interference similarity between the two infrared images, sim(n p1 ,n p2 The expression represents the similarity of the interference distribution between two infrared images.

[0138] The above embodiments provide a specific method for determining the similarity of infrared interference effects. In other embodiments, other methods can be used to couple the overall interference similarity with the interference distribution similarity. Quantitative description of interference similarity can guide the optimized design of detector anti-interference performance.

[0139] like Figure 5 , Figure 6 As shown, this embodiment of the invention provides a device for evaluating interference effects based on similarity assessment. The device embodiment can be implemented through software, hardware, or a combination of both. From a hardware perspective, as... Figure 5 The diagram shown is a hardware architecture diagram of an electronic device for evaluating interference effects based on similarity assessment, provided in an embodiment of the present invention. (Except for...) Figure 5 In addition to the processor, memory, network interface, and non-volatile memory shown, the electronic device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing packets. Taking software implementation as an example, such as... Figure 6 As shown, a device in a logical sense is formed by the CPU of its electronic device reading the corresponding computer program from non-volatile memory into memory and running it. This embodiment provides a similarity-based interference effect assessment device, comprising:

[0140] Image acquisition module 601 is used to acquire different infrared images;

[0141] The similarity calculation module 602 is used to evaluate the degree of influence of particulate matter on infrared images by calculating the similarity of infrared interference effects between different infrared images;

[0142] The similarity calculation module 602 calculates the similarity of infrared interference effects between two different infrared images in the following manner:

[0143] Two infrared images were preprocessed to determine the interference caused by particulate matter in the infrared images;

[0144] The overall interference similarity between the two infrared images is determined based on the total number, total scale, and total brightness of interference in each image.

[0145] Based on the interference caused by particulate matter in infrared images and the corresponding statistical rules, interference matrices for two infrared images are determined respectively. The interference matrix includes the interference pattern, field scale, and brightness-related classification statistical vectors. The elements of the classification statistical vectors are the number of interferences counted in intervals according to the corresponding statistical rules.

[0146] Based on the interference matrix of the two infrared images, the similarity of the interference distribution between the two infrared images is obtained;

[0147] The final infrared interference effect similarity is determined based on the overall interference similarity and the interference distribution similarity between the two infrared images.

[0148] In this embodiment of the invention, the image acquisition module 601 can be used to execute step 100 in the above method embodiment, and the similarity calculation module 602 can be used to execute step 200 in the above method embodiment.

[0149] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on an interference effect assessment device based on similarity evaluation. In other embodiments of the present invention, an interference effect assessment device based on similarity evaluation may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0150] The information interaction and execution process between the modules in the above-mentioned device are based on the same concept as the method embodiment of the present invention, and the specific details can be found in the description of the method embodiment of the present invention, and will not be repeated here.

[0151] This invention also provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements an interference effect evaluation method based on similarity evaluation according to any embodiment of this invention.

[0152] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform an interference effect evaluation method based on similarity assessment according to any embodiment of this invention.

[0153] Specifically, a system or apparatus equipped with a storage medium may be provided, on which software program code implementing the functions of any of the embodiments described above is stored, and the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage medium.

[0154] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.

[0155] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0156] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system or other components operating on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0157] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion module connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion module execute some and all of the actual operations, thereby realizing the function of any of the above embodiments.

[0158] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0159] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as ROM, RAM, magnetic disk, or optical disk.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating interference effects based on similarity assessment, characterized in that, include: Acquire different infrared images; The impact of particulate matter on infrared images is evaluated by calculating the similarity of infrared interference effects between different infrared images. The similarity of infrared interference effects between two different infrared images is calculated using the following method: Two infrared images were preprocessed to determine the interference caused by particulate matter in the infrared images; The overall interference similarity between the two infrared images is determined based on the total number, total scale, and total brightness of interference in each image. Based on the interference caused by particulate matter in infrared images and the corresponding statistical rules, interference matrices for two infrared images are determined respectively. The interference matrix includes the interference pattern, field scale, and brightness-corresponding classification statistical vectors. The elements of the classification statistical vectors are the number of interferences counted in intervals according to the corresponding statistical rules. Based on the interference matrix of the two infrared images, the similarity of the interference distribution between the two infrared images is obtained; The final infrared interference effect similarity is determined based on the overall interference similarity and the interference distribution similarity between the two infrared images.

2. The interference effect evaluation method according to claim 1, characterized in that, The determination of interference caused by particulate matter in the infrared image includes: The judgment is based on the pixel maximum and minimum values ​​of a single infrared image; if L min ≤0.3L cut +L cut And L max ≤0.3L min +L min If the interference quantity in the infrared image is 0, then L is considered to be 0. min L represents the minimum pixel value of an infrared image. max L represents the maximum pixel value of an infrared image. cut This represents the detectable threshold of the infrared detector; if L min >0.3L cut +L cut If L min ≤0.3L cut +L cut And L max >0.3L min +L min If so, proceed with the next steps; Using ΔL = 0.3L min +L min As a threshold, the infrared image is binarized by setting the pixel values ​​of the infrared image that exceed the threshold value ΔL to 1 and the other pixel values ​​to 0, thus obtaining a binarized image. Based on the obtained binarized image, the total number of interferences in the infrared image is determined by detecting the connectivity of the image. The total scale of interference in the infrared image is obtained by counting the number of pixel points whose pixel values ​​exceed the threshold value ΔL. The total brightness of the interference in the infrared image is obtained by statistically analyzing the pixel values ​​of pixels whose pixel values ​​exceed the threshold value ΔL.

3. The interference effect evaluation method according to claim 1, characterized in that, Determining the overall interference similarity between two infrared images includes: The total number, total scale, and total brightness of the interference in each of the two infrared images are dimensionless, resulting in two sets of corresponding dimensionless parameters, expressed as follows: Where, N′ g,tol 、D′ g,tol and L′ g,tol N″ represents the total number, total scale, and total brightness of interference in the first infrared image, respectively. g,tol D″ g,tol and L″ g,tol Let [n′, d′, l′] represent the total number, total scale, and total brightness of interference in the second infrared image, respectively. p1 and [n″,d″,l″] p2 These represent the dimensionless parameter groups corresponding to the first and second infrared images, respectively. Using the dimensionless parameter set [n′,d′,l′] corresponding to the first infrared image p1 Using this as a benchmark, if the dimensionless parameter set corresponding to the second infrared image is [n″,d″,l″], then... p2 If n″≥n′, d″≥d′, l″≥l′ or n″≤n′, d″≤d′, l″≤l′, then a positive deviation is determined; otherwise, a negative deviation is determined. Based on the dimensionless parameter set [n′,d′,l′] p1 and [n″,d″,l″] p2 The overall interference similarity between two infrared images is calculated using the following expression: in, σ represents the distance between the parameter vectors corresponding to the two infrared images; σ represents the adjustment factor, which is 0.5 if there is a positive deviation and 0.8 if there is a negative deviation.

4. The interference effect evaluation method according to claim 1, characterized in that, The interference matrices for the two infrared images are determined based on the interference caused by particulate matter in the infrared images and the corresponding statistical rules, including: Based on the interference caused by particulate matter in the infrared image, the proportion of each interference is determined to represent the pattern of interference; the proportion is determined by the percentage of the actual area of ​​the interference in the bounding rectangle of the interference. Based on the corresponding statistical rules, the percentage is divided into various statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the style. Based on the interference caused by particulate matter in the infrared image, the field of view scale of each interference is determined; According to the corresponding statistical rules, the statistical intervals of the field of view scale are divided, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the field of view scale. The brightness of each interference is determined based on the interference caused by particulate matter in the infrared image; Based on the corresponding statistical rules, the brightness is divided into various statistical intervals, and the number of interference statistics corresponding to each statistical interval is determined to obtain the classification statistical vector corresponding to the brightness.

5. The interference effect evaluation method according to claim 4, characterized in that, The interference matrix based on the two infrared images is used to obtain the interference distribution similarity between the two infrared images, including: The interference matrices of the two infrared images are normalized respectively to obtain two normalized matrices n. p1 and n p2 ; Based on the two normalized matrices n p1 and n p2 Calculate the Euclidean distance d(n) p1 ,n p2 ); Based on the obtained Euclidean distance d(n) p1 ,n p2 To calculate the similarity of interference distribution between two infrared images, the expression is:

6. The interference effect evaluation method according to claim 1, characterized in that, The determination of the final infrared interference effect similarity based on the overall interference similarity and the interference distribution similarity between two infrared images includes: The final expression for the similarity of the infrared interference effect is obtained through multiplication: η g =f(L)sim(n) p1 ,n p2 ), where f(L) represents the overall interference similarity between the two infrared images, sim(n p1 ,n p2 The expression represents the similarity of the interference distribution between two infrared images.

7. The interference effect evaluation method according to claim 3, characterized in that, Also includes: The qualitative evaluation result of the overall similarity is determined based on the overall interference similarity between the two infrared images; If the overall interference similarity f(L) between two infrared images is ≥ 0.9, they are determined to be highly similar overall; if 0.7 ≤ f(L) < 0.9, they are determined to be similar overall; if 0.3 ≤ f(L) < 0.7, they are determined to be close overall; if 0.1 ≤ f(L) < 0.3, they are determined to be dissimilar overall; if f(L) < 0.1, they are determined to be highly dissimilar overall.

8. A device for evaluating interference effects based on similarity assessment, characterized in that, include: The image acquisition module is used to acquire different infrared images; The similarity calculation module is used to evaluate the degree of influence of particulate matter on infrared images by calculating the similarity of infrared interference effects between different infrared images; The similarity calculation module calculates the similarity of infrared interference effects between two different infrared images in the following way: Two infrared images were preprocessed to determine the interference caused by particulate matter in the infrared images; The overall interference similarity between the two infrared images is determined based on the total number, total scale, and total brightness of interference in each image. Based on the interference caused by particulate matter in infrared images and the corresponding statistical rules, interference matrices for two infrared images are determined respectively. The interference matrix includes the interference pattern, field scale, and brightness-corresponding classification statistical vectors. The elements of the classification statistical vectors are the number of interferences counted in intervals according to the corresponding statistical rules. Based on the interference matrix of the two infrared images, the similarity of the interference distribution between the two infrared images is obtained; The final infrared interference effect similarity is determined based on the overall interference similarity and the interference distribution similarity between the two infrared images.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the interference effect evaluation method as described in any one of claims 1-7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed in a computer, the computer is instructed to perform the interference effect evaluation method according to any one of claims 1-7.

Images