Infrared weak and small target detection method and device based on asymmetric penalty sparsity

Abstract

The invention discloses an infrared weak and small target detection method and device based on asymmetric penalty sparseness. Comprising the following steps: 1) constructing a to-be-detected image sequence into a three-dimensional space-time tensor # imgabs0 #; 2) carrying out principal component analysis on the # imgabs 1 # to obtain a principal component tensor # imgabs 2 #; 3) constructing an asymmetric penalty sparseness estimation function # imgabs3 # of the target; (4) constructing a tensor nuclear norm # imgabs4 # based on a Laplace operator; 5) establishing an infrared weak and small target detection model based on asymmetric penalty sparseness; the invention discloses an infrared weak and small target detection method based on asymmetric penalty sparseness, and the method comprises the steps of (1) establishing a model in the step (2), (3) establishing the model in the step (3), (4) utilizing an ADMM-based algorithm to solve the model established in the step (5) to obtain a target detection result tensor # imgabs5 #, and (7) converting the # imgabs6 # into a target detection result sequence to achieve infrared weak and small target detection.According to the infrared weak and small target detection method based on asymmetric penalty sparseness, the comprehensive detection performance of infrared weak and small targets can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to an infrared dim small target detection method and device based on asymmetric penalty sparsity. Background Art

[0002] At present, infrared search and tracking (IRST) systems have been widely used in the fields of national defense and public security, such as mine detection, low-altitude safety, missile tracking, and drone defense. As one of the core technologies of the IRST system, infrared weak target detection plays a vital role in achieving real-time warning and precise tracking. However, due to the characteristics of long-range imaging, the targets in infrared images usually show the characteristics of "small size and weak signal", lacking clear shape, texture details and other structured information. At the same time, the complex and changeable imaging environment often causes strong background interference and noise in infrared images, such as dense clouds, ocean clutter, thermal radiation, etc. These factors make infrared weak target detection a challenging research topic.

[0003] In complex scenes, the key to achieving high-precision infrared dim target detection lies in: (1) background modeling and representation, (2) target sparsity estimation, and (3) robustness and real-time performance. At present, infrared dim target detection methods mainly include traditional detection methods and deep learning-based detection methods. Traditional detection methods design detection algorithms by utilizing the prior characteristics of infrared images and the statistical characteristics of targets and backgrounds, including filtering-based methods, background modeling-based methods, low-rank sparse decomposition-based methods, and local contrast measurement-based methods. Deep learning-based methods rely on the learning ability of large-scale data and can extract complex feature information, including convolutional neural network (CNN)-based methods and transformer-based methods. However, existing methods cannot achieve optimal performance in terms of background suppression and target detection capabilities at the same time. Therefore, developing infrared dim target detection methods with higher robustness and generalization capabilities remains an important research direction. Summary of the invention

[0004] In view of the deficiencies in the prior art, the object of the present invention is to provide an infrared dim small target detection method and device based on asymmetric penalty sparsity.

[0005] The present invention constructs the infrared image sequence to be detected as a three-dimensional space-time tensor, makes full use of the space-time information of the infrared image sequence, and carries out infrared weak small target detection based on low-rank sparse decomposition. Conduct principal component analysis, use singular value decomposition to determine the number of principal components to be retained, and obtain the principal component tensor , it can achieve the initial suppression of clutter noise, retain most of the background components and target components, and lay the data foundation for the subsequent low-rank sparse decomposition. Furthermore, an asymmetric penalty sparsity estimation function of the target is constructed to improve the accuracy of estimating the sparsity of the target; a tensor nuclear norm based on the Laplace operator is constructed to improve the accuracy of estimating the low-rank background. Thus, an infrared dim target detection model based on asymmetric penalty sparsity is established to achieve infrared dim target detection. Through comprehensive verification of the embodiments, it is proved that the infrared dim target detection method based on asymmetric penalty sparsity has excellent effects in improving detection performance and background suppression, and provides new ideas and methods for research and application in related fields.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] On the one hand, the present invention discloses an infrared dim small target detection method based on asymmetric penalty sparsity, comprising the following steps:

[0008] Step 1): Stack the thermal infrared image sequence to be detected in the time dimension to construct a three-dimensional space-time tensor ;

[0009] Step 2): For the three-dimensional space-time tensor Conduct principal component analysis, use singular value decomposition to determine the number of principal components to be retained, and obtain the principal component tensor ;

[0010] Step 3): Construct an asymmetric penalty sparsity estimation function for the target , estimate the sparsity of the target;

[0011] Step 4): Perform a fast Fourier transform on the background tensor contained in the principal component tensor, and calculate the numerical sum of the Laplace function of the singular values ​​of all the positive slices in the obtained tensor to construct the tensor nuclear norm based on the Laplace operator , estimating the low-rank properties of the background;

[0012] Step 5): Combine the target asymmetric penalty sparsity estimation function and the tensor nuclear norm based on the Laplace operator to establish the principal component tensor Infrared dim small target detection model based on asymmetric penalty sparsity;

[0013] Step 6): Use the ADMM-based optimization solution algorithm to solve the infrared weak target detection model based on asymmetric penalty sparsity constructed in step 5) to obtain the target detection result tensor ;

[0014] Step 7): Transform the target detection result tensor It is converted into a target detection result sequence T as the infrared weak target detection result to realize infrared weak target detection.

[0015] The present invention also discloses an infrared dim small target detection device based on spatiotemporal background reconstruction for implementing the method, which comprises:

[0016] A space-time tensor construction module is used to construct the infrared image sequence to be detected into a three-dimensional space-time tensor;

[0017] Principal component analysis module, used to obtain the principal component tensor corresponding to the three-dimensional space-time tensor ;

[0018] The first design module is used to construct the asymmetric penalty sparsity estimation function of the target , estimate the sparsity of the target component;

[0019] The second design module is used to perform fast Fourier transform on the background tensor contained in the principal component tensor, and calculate the numerical sum of the Laplace function of the singular values ​​of all the frontal slices in the obtained tensor, construct the tensor nuclear norm based on the Laplace operator, and estimate the low rank of the background component;

[0020] A model building module based on asymmetric penalty sparsity is used to build an infrared dim small target detection model based on asymmetric penalty sparsity;

[0021] Model optimization solution module, used to solve and establish an infrared dim small target detection model based on asymmetric penalty sparsity;

[0022] The target detection result output module is used to output the target detection result image corresponding to each frame of infrared image in the infrared image sequence.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1) The present invention constructs a spatiotemporal tensor to make full use of the spatiotemporal information of the infrared image sequence, and conducts principal component analysis, uses singular value decomposition to determine the number of principal components to be retained, and obtains the principal component tensor , achieving preliminary suppression of clutter noise in the original data signal, retaining most of the background components and target components, and laying the data foundation for subsequent low-rank sparse decomposition.

[0025] 2) The present invention designs a sparsity estimation step based on an asymmetric penalty function, which effectively alleviates the problem of underestimation of signals due to uniform penalty imposed on signals of different intensities in the traditional sparsity estimation process. The designed method can more flexibly process target signals with different intensities, ensure that signal components are not over-penalized, and effectively reduce information loss; at the same time, it can prevent smaller signals from being ignored. Therefore, the designed sparsity estimation method based on an asymmetric penalty function provides an effective solution for improving the accuracy of sparse target estimation in infrared small target detection. The infrared weak small target detection model based on asymmetric penalty sparsity established by the present invention can achieve satisfactory target detection capability, background suppression capability and comprehensive detection performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of constructing a three-dimensional space-time tensor in the present invention;

[0027] Figure 2 It is a structural schematic diagram of an infrared dim small target detection device based on asymmetric penalty sparsity in the present invention;

[0028] Figure 3 This is a sample frame image of a thermal infrared image sequence used for experimental testing;

[0029] Figure 4 The thermal infrared small target detection result obtained by the method proposed by the present invention is an example frame of the thermal infrared image sequence;

[0030] Figure 5 Figure 1 is an original image of an example frame of a thermal infrared image and its thermal infrared weak target detection result diagram after detection by the method of the present invention, AGPCNet, AMFUNet, GSWLCM, MDWCM, IPCE and FGLR-MCP. The green box indicates target loss, the red box indicates the target area, and the orange dotted box indicates clutter interference. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in detail below in conjunction with specific embodiments. Specific embodiments are described below to simplify the present invention. However, it should be recognized that the present invention is not limited to the illustrated embodiments, and various modifications of the present invention are possible without departing from the basic principles, and these equivalent forms also fall within the scope defined by the claims attached to the present application.

[0032] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] The basic steps of the infrared small target detection method based on frequency domain data spherification of the present invention in this embodiment mainly include:

[0034] Step 1: Stack the original thermal infrared image sequence to be detected in the time dimension to construct a three-dimensional space-time tensor ;

[0035] Specifically, Figure 1 As shown in the figure, the original thermal infrared image sequence to be detected is stacked in the time dimension to form a three-dimensional space-time tensor , The size is , express The width of express Height, express Thickness;

[0036] Specifically in the embodiment, the constructed three-dimensional space-time tensor The size is .

[0037] Step 2: 3D space-time tensor Conduct principal component analysis, use singular value decomposition to determine the number of principal components to be retained, and obtain the principal component tensor ;

[0038] Specifically, the three-dimensional space-time tensor Expand to a two-dimensional matrix , represents the spread operator, The size is , width is , Gao Wei ;

[0039] The number of principal components is determined using singular value decomposition. ,in, and represents the unitary matrix obtained by singular value decomposition, express The singular value matrix of ; express No. singular values, defined Represents the contrast coefficient, from the index list From small to large , when first satisfied When , determine the number of principal components For this time ;

[0040] Pair Matrix Transpose Conduct principal component analysis and Dimensionality reduction principal components, we get

[0041] (1)

[0042] in, For the front The coefficient matrix of the principal components, is the score matrix;

[0043] The principal component reconstruction process is:

[0044] (2)

[0045] in, Indicates The score vector corresponding to the principal components is Indicates The coefficient vector corresponding to the principal components is The first column vector, represents the reconstructed principal component matrix; Reconstructed into the original three-dimensional space-time tensor size, the principal component tensor is obtained ,have ;

[0046] Specifically in the embodiment, Take it as 0.2.

[0047] Step 3: Construct an asymmetric penalty sparsity estimation function for the target , estimate the sparsity of the target;

[0048] Specifically, the asymmetric penalty function Defined as

[0049] (3)

[0050] in, represents the asymmetry control factor, A smoothing factor that represents the smoothness of the control function and the intensity of the penalty near zero;

[0051] According to formula (3), the asymmetric penalty sparsity estimation function of the target is constructed for

[0052] (4)

[0053] in, represents the target tensor;

[0054] Specifically in the embodiment, Take it as 0.001, Take it as 4.

[0055] Step 4: Construct the tensor nuclear norm based on the Laplace operator , estimating the low-rank properties of the background;

[0056] Specifically, the background tensor Perform fast Fourier transform to get the tensor ,calculate The numerical sum of the Laplace functions of the singular values ​​of all positive slices is defined as the tensor kernel norm based on the Laplace operator , the calculation formula is:

[0057] (5)

[0058] in, Represents the background tensor The tensor obtained after fast Fourier transformation, express The i-th front slice The kth singular value of is the control parameter of the Laplace operator, which is a positive constant;

[0059] Specifically in the embodiment, Take it as 0.001.

[0060] Step 5: Combine the target asymmetric penalty sparsity estimation function and the tensor nuclear norm based on the Laplace operator to establish the principal component tensor Infrared dim small target detection model based on asymmetric penalty sparsity;

[0061] Specifically, the asymmetric penalty sparsity estimation function of the target in step 3) and the tensor nuclear norm based on the Laplace operator in step 4) are combined to convert the principal component tensor Modeled as a background tensor , target tensor and the noise tensor The linear combination of The low-rank sparse decomposition process of

[0062] (6)

[0063] The sparsity reweighting strategy is introduced, and the sparsity reweighted tensor is calculated as ,in, Represents a positive number to prevent the denominator from being zero; thus, the principal component tensor is obtained Infrared dim small target detection model based on asymmetric penalty sparsity:

[0064] (7)

[0065] in, and represents the weight coefficient, represents the square of the Frobenius norm, Represents the Hadamard product.

[0066] Step 6: Use the ADMM-based optimization solution algorithm to solve the infrared weak target detection model based on asymmetric penalty sparsity constructed in step 5) to obtain the target detection result tensor ;

[0067] Specifically, the alternating direction method of multipliers (ADMM) algorithm is used to solve the model (7) in step 5), introducing auxiliary variables , and the equivalent model is

[0068] (8)

[0069] The corresponding augmented Lagrangian function for:

[0070] (9)

[0071] in, is the penalty coefficient, is the Lagrange multiplier;

[0072] Decompose the problem of solving equation (9) into several sub-problems, and update the variables alternately according to equations (10) to (17) until the iteration ends;

[0073] Auxiliary variables The iteration formula is:

[0074] (10)

[0075] in, represents the number of iterations, , and are all unitary matrices, represents the singular value matrix, represents the singular value contraction operator, and , Each singular value in ,have ;

[0076] Background tensor The iteration formula is:

[0077] (11)

[0078] Target Tensor The iteration formula is

[0079] (12)

[0080] Among them, the intermediate parameters , is a soft threshold shrinkage operator, and , is the soft threshold;

[0081] Sparse reweighted tensor The iteration formula is

[0082] (13)

[0083] Noise Tensor The iteration formula is

[0084] (14)

[0085] Lagrange multipliers The iteration formula is

[0086] (15)

[0087] (16)

[0088] Penalty coefficient The iteration formula is

[0089] (17)

[0090] in, represents the update coefficient, express The maximum value of

[0091] During the iteration process, the iteration stops when any of the following conditions is met: (1) relative error ; (2) During two adjacent iterations No change;

[0092] When the iteration stops, the final target tensor is obtained ;

[0093] Specifically in the embodiment, , , , , , , , , represents the unit tensor, , Represents a three-dimensional zero tensor.

[0094] Step 7: The final target tensor It is converted into a target detection result sequence T as the infrared weak target detection result to realize infrared weak target detection.

[0095] Corresponding to the aforementioned embodiment of an infrared dim small target detection method based on asymmetric penalty sparsity, the present invention further provides an embodiment of an infrared dim small target detection device based on spatiotemporal background reconstruction.

[0096] Figure 2 FIG. 1 is a block diagram of an infrared dim small target detection device based on asymmetric penalty sparsity according to an exemplary embodiment. Figure 2 As shown, the device comprises:

[0097] A space-time tensor construction module is used to construct the infrared image sequence to be detected into a three-dimensional space-time tensor;

[0098] Principal component analysis module, used to obtain the principal component tensor corresponding to the three-dimensional space-time tensor ;

[0099] The first design module is used to construct the asymmetric penalty sparsity estimation function of the target , estimate the sparsity of the target component;

[0100] The second design module is used to perform fast Fourier transform on the background tensor contained in the principal component tensor, and calculate the numerical sum of the Laplace function of the singular values ​​of all the frontal slices in the obtained tensor, construct the tensor nuclear norm based on the Laplace operator, and estimate the low rank of the background component;

[0101] A model building module based on asymmetric penalty sparsity is used to build an infrared dim small target detection model based on asymmetric penalty sparsity;

[0102] Model optimization solution module, used to solve and establish an infrared dim small target detection model based on asymmetric penalty sparsity;

[0103] The target detection result output module is used to output the target detection result image corresponding to each frame of infrared image in the infrared image sequence.

[0104] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0105] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic. The various modules in the device are a kind of logical function division. There may be other division methods in actual implementation. For example, multiple modules can be combined or integrated into another unit. Another point is that the connection between the modules shown or discussed can be a communication connection through some interfaces, which can be electrical or other forms. Some or all of the modules can be selected according to actual needs to achieve the purpose of the present application. Ordinary technicians in this field can understand and implement it without paying creative labor. The following takes the disclosed real thermal infrared image sequence as an example to illustrate the specific implementation method to reflect the technical effect of the present invention. The specific steps in the embodiment will not be repeated.

[0106] Example

[0107] The accompanying drawings shown in the embodiments of the present invention can make the purpose, technical solutions and advantages of the present invention more clearly described. It should be noted that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. All equivalent replacements, improvements, etc. made within the method ideas and principles provided by the present invention should be included in the protection scope of the present invention.

[0108] In this embodiment, the effectiveness of the infrared weak target detection method based on asymmetric penalty sparsity will be verified through a public infrared image sequence. The public infrared image sequence contains 150 consecutive infrared image frames, and the resolution of each frame is 256×256. The sequence contains a moving drone target, which has a significant grayscale contrast with the complex background; the data set is based on the ground and contains randomly distributed highlights, texture structures and other linear highlights; in addition, there are strong clutter and non-Gaussian noise components in the image. Figure 3 shows an example frame of an infrared image sequence, Figure 4 This is the infrared small target detection result obtained by this method. From the detection results, it can be intuitively observed that this method can significantly enhance the target, while effectively suppressing background interference and noise components, showing excellent target detection capabilities. In order to more comprehensively and objectively evaluate the performance of the infrared small target detection method based on asymmetric penalty sparsity, the detection performance of the algorithm is evaluated from a quantitative perspective through the 3D-ROC evaluation index system, as follows:

[0109] Table 1-3D-ROC evaluation index system

[0110]

[0111] In order to more objectively verify the effectiveness of the method of the present invention, six thermal infrared dim target detection methods are selected for comparison with the method of the present invention, including AGPCNet, AMFUNet, GSWLCM, MDWCM, IPCE and FGLR-MCP. Table 2 shows the indicators of the 3D-ROC evaluation system for infrared dim target detection results using the comparison methods and the method of the present invention. The bold values ​​and underlined values ​​represent the corresponding optimal performance and suboptimal performance, respectively.

[0112] AGPCNet comes from Zhang T, Cao S, Pu T, et al. AGPCNet: Attention-guidedpyramid context networks for infrared small target detection. arXiv 2021[J].arXiv preprint arXiv:2111.03580.

[0113] AMFUNet comes from Chung WY, Lee IH, Park C G. Lightweight infrared smalltarget detection network using full-scale skip connection U-Net[J]. IEEEGeoscience and Remote Sensing Letters, 2023, 20: 1-5.

[0114] GSWLCM comes from Qiu Z, Ma Y, Fan F, et al. Global sparsity-weighted localcontrast measure for infrared small target detection[J]. IEEE Geoscience andRemote Sensing Letters, 2022, 19: 1-5.

[0115] MDWCM comes from Lu R, Yang X, Li W, et al. Robust infrared small target detection via multidirectional derivative-based weighted contrast measure[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 19: 1-5.

[0116] IPCE comes from Zhang C, He Y, Tang Q, et al. Infrared small target detectionvia interpatch correlation enhancement and joint local visual saliency prior[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 60: 1-14.

[0117] FGLR-MCP comes from Liu T, Liu Y, Yang J, et al. Graph Laplacianregularization for fast infrared small target detection[J]. PatternRecognition, 2025, 158: 111077.

[0118] Figure 5 The original image of the infrared image example frame and the thermal infrared weak target detection result diagram after the method of the present invention, AGPCNet, AMFUNet, GSWLCM, MDWCM, IPCE and FGLR-MCP detection. Figure 5It can be seen from the target detection result images that AGPCNet, MDWCM, IPCE and FGLR-MCP all have the problem of target loss, and there are more clutter in the detection results of GSWLCM and AMFUNet. In contrast, the infrared weak target detection method based on asymmetric penalty sparsity proposed in the present invention can well suppress background, noise and other components, and enhance the significance of weak targets. In addition, according to the indicator results based on the 3D-ROC evaluation system shown in Table 2, the compared methods cannot achieve target enhancement and background suppression well, resulting in insufficient comprehensive performance; in contrast, the target detection ability, background suppression ability and comprehensive ability of the method proposed in the present invention can basically achieve satisfactory results. Combining the above qualitative and quantitative analysis, the infrared weak target detection method based on asymmetric penalty sparsity proposed in the present invention has superior target detection ability, background suppression ability and comprehensive effectiveness.

[0119] Table 2 - Quantitative indicators of the detection results of thermal infrared image example sequences using AGPCNet, AMFUNet, GSWLCM, MDWCM, IPCE, FGLR-MCP and the method of the present invention

[0120]

[0121] The accompanying drawings shown in the embodiments of the present invention can make the purpose, technical solutions and advantages of the present invention more clearly described. It should be noted that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. All equivalent replacements, improvements, etc. made within the method ideas and principles provided by the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting infrared dim small targets based on asymmetric penalty sparsity, characterized in that: The steps include: Step 1): Stack the thermal infrared image sequence to be detected in the time dimension to construct a three-dimensional space-time tensor ; Step 2): For the three-dimensional space-time tensor Conduct principal component analysis, use singular value decomposition to determine the number of principal components to be retained, and obtain the principal component tensor ; Step 3): Construct an asymmetric penalty sparsity estimation function for the target , estimate the sparsity of the target; Step 4): Perform a fast Fourier transform on the background tensor contained in the principal component tensor, and calculate the numerical sum of the Laplace function of the singular values ​​of all the positive slices in the obtained tensor to construct the tensor nuclear norm based on the Laplace operator , estimating the low-rank properties of the background; Step 5): Combine the target asymmetric penalty sparsity estimation function and the tensor nuclear norm based on the Laplace operator to establish the principal component tensor Infrared dim small target detection model based on asymmetric penalty sparsity; Step 6): Use the ADMM-based optimization solution algorithm to solve the infrared weak target detection model based on asymmetric penalty sparsity constructed in step 5) to obtain the target detection result tensor ; Step 7): Transform the target detection result tensor It is converted into a target detection result sequence T as the infrared weak target detection result to realize infrared weak target detection.

2. The infrared small target detection method based on asymmetric penalty sparsity according to claim 1 is characterized in that: The step 1) is: stacking the original thermal infrared image sequence to be detected in the time dimension direction to construct a three-dimensional space-time tensor , The size is , express The width of express Height, express Thickness.

3. The infrared small target detection method based on asymmetric penalty sparsity according to claim 2 is characterized in that: The step 2) is specifically as follows: The three-dimensional space-time tensor Expand to a two-dimensional matrix , represents the spread operator, The size is ; The number of principal components is determined using singular value decomposition. ,in, and represents the unitary matrix obtained by singular value decomposition, express The singular value matrix of ; express No. singular values, defined Represents the contrast coefficient, from the index list From small to large , when first satisfied When , determine the number of principal components For this time ; Pair Matrix Transpose Conduct principal component analysis and Dimensionality reduction principal components, we get （1） in, For the front The coefficient matrix of the principal components, is the score matrix; The principal component reconstruction process is: （2） in, Indicates The score vector corresponding to the principal components is Indicates The coefficient vector corresponding to the principal components is The first column vector, represents the reconstructed principal component matrix; Reconstructed into the original three-dimensional space-time tensor size, the principal component tensor is obtained ,have .

4. The infrared small target detection method based on asymmetric penalty sparsity according to claim 1 is characterized in that: The step 3) is specifically as follows: Asymmetric penalty function Defined as: （3） in, represents the asymmetry control factor, A smoothing factor that represents the smoothness of the control function and the intensity of the penalty near zero; According to formula (3), the asymmetric penalty sparsity estimation function of the target is constructed for （4） in, Represents the target tensor.

5. The infrared small target detection method based on asymmetric penalty sparsity according to claim 3 is characterized in that: The step 4) is specifically as follows: The background tensor Perform fast Fourier transform to get the tensor ,calculate The numerical sum of the Laplace functions of the singular values ​​of all positive slices is defined as the tensor kernel norm based on the Laplace operator , the calculation formula is: （5） in, Represents the background tensor The tensor obtained after fast Fourier transformation, express The i-th front slice The kth singular value of is the control parameter of the Laplace operator.

6. The infrared small target detection method based on asymmetric penalty sparsity according to claim 5 is characterized in that: The step 5) is specifically as follows: Combining the asymmetric penalty sparsity estimation function of the target in step 3) and the tensor nuclear norm based on the Laplace operator in step 4), the principal component tensor Modeled as a background tensor , target tensor and the noise tensor The linear combination of The low-rank sparse decomposition process of is modeled as: （6） The sparsity reweighting strategy is introduced, and the sparsity reweighted tensor is calculated as ,in, Represents a positive number to prevent the denominator from being zero; thus the principal component tensor is obtained Infrared dim small target detection model based on asymmetric penalty sparsity: （7） in, and represents the weight coefficient, represents the square of the Frobenius norm, Represents the Hadamard product.

7. The infrared small target detection method based on asymmetric penalty sparsity according to claim 6 is characterized in that: The step 6) is specifically as follows: The ADMM algorithm is used to solve the model (7) in step 5), introducing auxiliary variables , the equivalent model is: （8） The augmented Lagrangian function corresponding to the equivalent model for: （9） in, is the penalty coefficient, is the Lagrange multiplier; Decompose the problem of solving equation (9) into several sub-problems, and update the variables alternately according to equations (10) to (17) until the iteration ends; Auxiliary variables The iteration formula is: （10） in, represents the number of iterations, , and are all unitary matrices, represents the singular value matrix, represents the singular value contraction operator, and , Each singular value in ,have ; Background tensor The iteration formula is: （11） Target Tensor The iteration formula is （12） Among them, the intermediate parameters for , is a soft threshold shrinkage operator, and , is the soft threshold; Sparse reweighted tensors The iteration formula is （13） Noise Tensor The iteration formula is （14） Lagrange multipliers The iteration formula is （15） （16） Penalty coefficient The iteration formula is （17） in, represents the update coefficient, express The maximum value of During the iteration process, the iteration stop condition is: relative error , or in two adjacent iterations No change; When the iteration stops, the final target tensor is obtained .

8. The infrared small target detection method based on asymmetric penalty sparsity according to claim 1 is characterized in that: The step 7) is specifically as follows: The final target tensor Reconstructed into infrared target detection result map, the small target detection result map sequence T is obtained to realize thermal infrared small target detection.

9. An infrared dim small target detection device based on asymmetric penalty sparsity for implementing the method of claim 1, characterized in that: include: A space-time tensor construction module is used to construct the infrared image sequence to be detected into a three-dimensional space-time tensor; Principal component analysis module, used to obtain the principal component tensor corresponding to the three-dimensional space-time tensor ; The first design module is used to construct the asymmetric penalty sparsity estimation function of the target , estimate the sparsity of the target component; The second design module is used to perform fast Fourier transform on the background tensor contained in the principal component tensor, and calculate the numerical sum of the Laplace function of the singular values ​​of all the frontal slices in the obtained tensor, construct the tensor nuclear norm based on the Laplace operator, and estimate the low rank of the background component; A model building module based on asymmetric penalty sparsity is used to build an infrared dim small target detection model based on asymmetric penalty sparsity; Model optimization solution module, used to solve and establish an infrared dim small target detection model based on asymmetric penalty sparsity; The target detection result output module is used to output the target detection result image corresponding to each frame of infrared image in the infrared image sequence.

Images