Unsupervised undersampling non-visible imaging method based on virtual scanning

Abstract

The application discloses a virtual scanning-based unsupervised undersampling non-visual imaging method and relates to the field of computational imaging. The virtual scanning-based unsupervised undersampling non-visual imaging method comprises the following steps: S1, constructing a simulation transient dataset by using a public transient renderer and ShapeNet 3D object data, and further processing to generate an undersampling non-visual transient dataset; S2, designing a denoiser network with an encoder-decoder structure composed of partial convolution and a reconstruction network combined with physical prior; S3, designing a framework and building a model: combining the undersampling non-visual reconstruction simulation transient dataset and the network framework described in S1-S2 to design a non-visual reconstruction scheme; S4, training the constructed network framework and the virtual scanning reconstruction network model in stages by using a deep learning Pytorch framework. The virtual scanning-based unsupervised undersampling non-visual imaging method is used to realize high-quality reconstruction effect from transient data collected under any irregular scanning pattern.

Description

Technical Field

[0001] The present invention relates to the field of computational imaging technology, and in particular to an unsupervised under-sampling non-viewing-area imaging method based on virtual scanning. Background Art

[0002] Non-line-of-sight (NLOS) imaging aims to reconstruct hidden scenes beyond the detector's line of sight. This emerging field has attracted widespread interest from researchers in various fields, including robotic vision, autonomous driving, rescue operations, remote sensing, and medical imaging. In a typical active confocal NLOS imaging system, both the laser source and the detector are focused on the same point on a relay surface. The laser pulse is reflected by the surface to illuminate the hidden scene. The detector captures the photons that bounce back from the scene to the relay surface, which is called a photon histogram. In order to improve the reconstruction resolution and quality, the relay surface is often scanned at multiple points. The captured photon time volume becomes a transient measurement (with dimensions of H×W×T). Through carefully designed algorithms, information about the hidden scene can be recovered from the transient measurement.

[0003] Existing work often focuses on improving reconstruction quality or scanning efficiency. Although significant breakthroughs have been achieved, these methods also face significant limitations, which seriously hinder their practical application in non-line-of-sight (NLOS). These methods require the imaging environment to contain large-area relay surfaces, enabling the imaging system to perform dense and regular scanning. However, in some real-world applications, only irregular relay surfaces such as lattices or fences may exist, which makes these methods' assumptions about the imaging environment invalid. Recovering hidden scenes using undersampled transient measurements obtained from irregular scanning is a serious ill-posed problem, resulting in artifacts and even structural loops in the reconstructed volume. This leads to the challenging task of achieving high-quality NLOS imaging from irregular undersampled transient measurements. To address this issue, CC-SOCR introduces a manually designed strong regularization term within a feature optimization framework. However, this method requires lengthy numerical iterations, which severely limits its application efficiency.

[0004] Recent deep learning-based non-line-of-sight imaging methods have achieved high-quality single-stage fast inference imaging, but they require a large amount of paired data as training supervision, and this data needs to be accurately aligned with the true albedo volume. In fact, such real data pairs are difficult to obtain and expensive. Therefore, in order to eliminate the heavy reliance of deep learning-based methods on paired data, it is very valuable to explore non-line-of-sight imaging under unsupervised learning.

[0005] In order to solve the above problems, the method of the present invention proposes an unsupervised under-sampled non-field of view imaging method based on virtual scanning, and test experiments on simulated and real under-sampled transient data show the optimal reconstruction results. Summary of the Invention

[0006] The purpose of the present invention is to propose an unsupervised under-sampling non-line-of-sight imaging method based on virtual scanning to solve the problems raised in the background technology and achieve higher quality non-line-of-sight reconstruction.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] The unsupervised under-sampling non-viewing-area imaging method based on virtual scanning specifically includes the following steps:

[0009] S1. Constructing an undersampled non-line-of-sight reconstruction simulation transient dataset: Using a public transient renderer and ShapeNet3D object data to construct a simulation transient dataset, adding noise and temporal jitter characteristics based on the non-line-of-sight imaging single-photon detector imaging model, and generating an undersampled non-line-of-sight transient dataset based on the set scanning pattern;

[0010] S2. Build a network framework: Design a denoiser network with an encoder-decoder structure consisting of partial convolutions and a reconstruction network combined with physical priors. Connect the denoiser network and the reconstruction network to reconstruct hidden scenes from irregular undersampled transients.

[0011] S3. Design framework and build model: Combine the undersampled NLOS reconstruction simulation transient dataset and network framework described in S1-S2 to design a NLOS reconstruction scheme. The scheme specifically includes the following contents:

[0012] ① Undersampled transient denoising: The captured irregular undersampled transient measurements are filled into the full sampling grid by zero-filling method, and the zero-filled transient and the mask are input into the denoiser network to obtain the initial denoised transient, which is then multiplied by the mask to remove the artifacts introduced by the network and obtain the denoised undersampled transient;

[0013] ② Hidden scene reconstruction: The denoised undersampled transient is initially reconstructed using the traditional LCT algorithm to obtain the value domain information of the true reconstruction result. This information is then input into the reconstruction network to obtain the zero domain information of the true reconstruction result. The value domain information is combined with the zero domain information to obtain the albedo reconstruction volume of the hidden scene.

[0014] ③ Design a virtual scanning training strategy: Based on the zero-value domain decomposition principle, an unsupervised virtual scanning training strategy and its loss function are designed to enable the neural network to learn the ability to recover zero-domain information;

[0015] S4. Training the unsupervised non-line-of-sight reconstruction framework: Use the deep learning Pytorch framework to train the constructed network framework and virtual scanning reconstruction network model in stages, and repeatedly traverse the undersampled non-line-of-sight reconstruction simulation transient dataset constructed in S1 until the model converges.

[0016] Preferably, the S1 specifically includes the following contents:

[0017] S1.1. Select and organize ShapeNet 3D object data;

[0018] S1.2. Modeling single-photon detector imaging under non-line-of-sight imaging conditions;

[0019] S1.3. Use the publicly available non-line-of-sight imaging transient measurement renderer to simulate ideal transient measurements based on ShapeNet 3D object data.

[0020] S1.4. Post-process the simulated ideal transient measurement based on the single-photon detector imaging model to make it close to the fully sampled transient measurement of the real acquisition;

[0021] S1.5. Extract undersampled data from the simulated full-sampled transient measurement according to the selected scanning pattern for training and construct a data set.

[0022] Preferably, the S2 specifically includes the following contents:

[0023] S2.1. A non-line-of-sight single-photon detector imaging model is given. The model is an imaging model that considers low-photon time-resolution detectors. The specific function is expressed as:

[0024]

[0025] where u represents the noise-free measurement, represents the noisy measurement, b represents the dark photons and background noise;

[0026] The system model of non-line-of-sight imaging is simplified to a linear inverse problem form, namely:

[0027] u=Hρ

[0028] Where H represents the forward degradation operator; ρ represents the three-dimensional albedo volume of the hidden scene;

[0029] S2.2. According to the imaging model given in S2.1, we derive an unsupervised unbiased estimator based on Stein's Unbiased Risk Estimation and an unsupervised denoising loss function. The specific function is expressed as:

[0030]

[0031] in, represents the unsupervised denoising loss function based on Stein's Unbiased Risk Estimation; i∈{0,…,I}, g∈{0,…,G}, I, G represent the number of transients and the number of sampling operators respectively; s represents the number of scanning points; t represents the time dimension, F φ represents the denoiser network; ε is a positive number; e is a binary vector;

[0032] S2.3. Use partial convolution to build a network model based on the encoder-decoder structure, where the partial convolution operator is expressed as:

[0033]

[0034] Where M represents the mask; x represents the input; W represents the convolution kernel parameter; b represents the bias, and x' represents the output;

[0035] S2.4. Construct a Unet-like network that combines physical priors and attention mechanism. Use the traditional LCT algorithm to perform domain conversion and initial reconstruction on the denoised transient to obtain value domain information. Then use the reconstruction network to restore the zero domain information. The restored zero domain information and value domain information are added together to experimentally reconstruct the complete hidden scene.

[0036] Preferably, the solution ③ specifically includes the following contents:

[0037] 3.1) Select a fully sampled transient and a sampling matrix, extract data from the fully sampled transient according to the sampling matrix to form undersampled data;

[0038] 3.2) Fill the undersampled data to the fully sampled grid using zero padding, input it and the mask into the denoiser network to obtain the initial denoised transient, and then multiply it with the mask to remove the artifacts introduced by the network to obtain the denoised undersampled transient;

[0039] 3.3) The denoised undersampled transient is initially reconstructed using the traditional LCT algorithm to obtain the range information of the true reconstruction result. The range information of the true reconstruction result is input into the reconstruction network. The output of the reconstruction network is combined with the range information to obtain the albedo reconstruction volume of the hidden scene.

[0040] 3.4) Observe the albedo reconstruction volume through the fully sampled forward operator, and then select a sampling matrix different from that in step 3.1) and multiply it with the new fully sampled observation to obtain a new undersampled transient;

[0041] 3.5) Repeat step 3.3) with the newly obtained undersampled transient in step 3.4) to obtain a new albedo reconstruction volume;

[0042] 3.6) The reconstruction network in step 3.3 is supervised by a measurement consistency loss function and a virtual scan loss function, where the measurement consistency loss function is expressed as:

[0043]

[0044] Among them, L MC represents the measurement consistency loss function; Indicates that the sampling matrix M is used g The obtained undersampled transient; H g Representation and sampling matrix M g The corresponding forward operator; f θ Represents the traditional algorithm LCT and the reconstruction network F θ The reconstruction modules composed of

[0045] The virtual scanning loss function is expressed as:

[0046]

[0047] Among them, L VS is the virtual scanning loss function; H k Representation and sampling matrix M k The corresponding forward operator.

[0048] Preferably, the S4 includes the following contents:

[0049] S4.1. Build an optimizer and gradient update strategy based on the Pytorch framework, and train the network using the undersampled non-field-of-view reconstruction simulation transient dataset constructed in S1.

[0050] S4.2. Use the unsupervised denoising loss function based on Stein's Unbiased Risk Estimation in S2 to train the denoiser network.

[0051] S4.3. Freeze the denoiser network weights and train the reconstruction network proposed in S2.4 using the virtual scanning strategy proposed in Scheme ③.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] (1) We propose an unsupervised non-line-of-sight imaging framework that can learn implicit priors from arbitrary undersampled transient measurements, overcoming the reliance on paired data that is difficult to obtain and align.

[0054] (2) This paper introduces a virtual scanning strategy that enables the network to learn in both value space and null space, thereby achieving high-quality reconstruction and extending non-line-of-sight imaging to realistic scenarios with irregular relay surfaces.

[0055] (3) We introduce a SURE-based denoiser, an unsupervised physics-guided module that incorporates a physical model of low-photon time-resolution detectors to enhance the robustness of our method to noise.

[0056] (4) We evaluate our method on simulated data, publicly available data, and captured real data, demonstrating superior reconstruction quality and inference speed significantly faster than current state-of-the-art methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 The unsupervised under-sampling non-viewing-area imaging system model based on virtual scanning constructed in Example 1 of the present invention;

[0058] Figure 2 This is an overall framework diagram of the unsupervised under-sampling non-viewing-area imaging method based on virtual scanning proposed in Example 1 of the present invention;

[0059] Figure 3-4 This is a diagram showing the effect of testing the method of the present invention and the traditional reconstruction method on a data set proposed in Example 2 of the present invention. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0061] Example 1:

[0062] See also Figure 2 The present invention proposes an unsupervised under-sampling non-viewing-area imaging method based on virtual scanning, comprising the following steps:

[0063] S1. Constructing an undersampled non-line-of-sight reconstruction simulation transient dataset: Using a public transient renderer and ShapeNet3D object data to construct a simulation transient dataset, adding noise and temporal jitter characteristics based on the non-line-of-sight imaging single-photon detector imaging model, and generating an undersampled non-line-of-sight transient dataset based on the set scanning pattern;

[0064] S2. Build a network framework: Design a denoiser network F with an encoder-decoder structure consisting of partial convolutions φ , a reconstruction network F that combines physical priors θ ,Connecting the two networks enables the reconstruction of hidden scenes from ,irregular undersampled transients;

[0065] S3. Design scheme and build model: Design a non-line-of-sight reconstruction scheme based on the undersampled non-line-of-sight reconstruction simulation transient dataset and network framework described in S1-S2;

[0066] S4. Training the unsupervised non-viewing reconstruction framework: Using the deep learning Pytorch framework, the unsupervised denoiser and virtual scan reconstruction network model constructed in S1 and S2 are trained in stages, repeatedly traversing the simulated transient dataset constructed in S4 until the model converges.

[0067] The above method further includes the following contents:

[0068] 3.1 Denoiser

[0069] SURE-based denoiser. The SURE-based denoiser consists of two parts: the SURE loss function used to supervise the denoiser training and the network module that constitutes the denoiser itself.

[0070] The denoiser can be trained using end-to-end supervised training of noise-free and noisy data pairs. The mean square error loss (MSE) is generally used for supervised training:

[0071]

[0072] In the unsupervised case, it is impossible to obtain a noise-free transient dataset {u j} is used as supervision. By using the SURE-based loss function through approximate derivation, an unbiased estimate of the above-mentioned supervision loss can be obtained, thereby enabling denoiser training in the absence of supervision.

[0073] Existing SURE-based work has considered Gaussian noise models, Poisson noise models, and Gaussian-Poisson mixed noise models, but lacks algorithms suitable for non-line-of-sight imaging. To this end, the noise model of non-line-of-sight single-photon detector imaging is considered.

[0074]

[0075] where u represents the noise-free measurement, represents the noisy measurement, and b represents the dark photons and background noise.

[0076] Substituting the above noise model into the derivation process, we can obtain the following unsupervised SURE denoising loss suitable for non-line-of-sight imaging:

[0077]

[0078] in, represents the unsupervised denoising loss function based on Stein's Unbiased Risk Estimation; i∈{0,…,I}, g∈{0,…,G}, I, G represent the number of transients and the number of sampling operators respectively; s represents the number of scanning points; t represents the time dimension, F φ represents the denoiser network; ε is a positive number; e is a binary vector;

[0079] In order to fully exploit the spatial correlation of undersampled transients, this method restores the undersampled transient data to a fully sampled grid through zero padding to reconstruct its spatial correlation. At the same time, considering that ordinary convolutional layers regard all voxel values ​​as valid, ordinary convolutional layers will introduce zero padding information, resulting in errors.

[0080] Therefore, this method constructs an encoder-decoder network consisting of partial convolutions, which adjust the part involved in the calculation according to the input mask, expressed as:

[0081]

[0082] Among them, M represents the mask; x represents the input; W represents the convolution kernel parameter; b represents the bias, and x' represents the output.

[0083] After passing through the denoising network, this method multiplies the mask by the initial denoising result to remove artifacts introduced by the network in non-scanned locations.

[0084] 3.2 Virtual Scanning Training Strategy

[0085] The non-line-of-sight imaging forward model can be approximately simplified to a linear model:

[0086] u=Hρ

[0087] where u represents the noise-free measurement, H is the forward degradation operator, and ρ is the 3D albedo volume of the hidden scene.

[0088] According to the zero-field decomposition theorem, any albedo volume ρ can be decomposed into:

[0089]

[0090] in, is the pseudo-inverse operator of H, and the first term ρ1 on the right is called the range information of ρ (D r (ρ)), the second term ρ2 is the zero-domain information of ρ (D n (ρ)).

[0091] Standard unsupervised training generally uses measurement consistency loss to supervise network training, which is expressed as:

[0092]

[0093] Among them, H is the forward degeneration operator, f θ To reconstruct the network, u is the transient measurement. However, this loss function can only enable the network to restore the target's value range information, and cannot accurately restore the zero domain information. Specifically, this method selects the LCT algorithm as the pseudo-inverse operator, and the output can be uniformly expressed as D r (ρ)+v n , where D r (ρ) is the range information recovered by the traditional algorithm LCT, v n The zero-field information recovered for the network. and H v n = 0, there are multiple output results that satisfy the measurement consistency loss constraint, namely:

[0094]

[0095] in, And Hv n = 0 and. Therefore, the measurement consistency loss function makes f θ Uniquely recover the true result.

[0096] To this end, this method proposes a virtual scanning strategy to enable the network to learn the ability to restore the null space. Specifically, the network is first reconstructed from Reconstruct 3D albedo volume ρ (1) Then select and obtain Different forward degradation operators H k , in order to simplify the operation, H k Decomposed into H r ⊙M k , through H k ρ (1) Implement virtual scanning to obtain new transient data u kr At this point, the same virtual scanning network is used to obtain a new three-dimensional resolution volume ρ (2) The virtual scanning loss requirement ρ (1) =ρ (2) Under this loss constraint, according to the zero-field value decomposition theorem, the network needs to recover ρ (1) Null space information D n (ρ (1) ), in order to perfectly reconstruct ρ (1) ,Right now:

[0097] ρ (1) =D r (ρ (1) )+D n (ρ (1) )

[0098] During training, by traversing different forward degradation operators to enhance the network's robustness to different scanning patterns, the network can acquire the ability to recover null space information, thereby achieving high-quality 3D reconstruction.

[0099] 3.3 Training Model

[0100] The method of the present invention is implemented using PyTorch and uses the Adam optimizer with a weight decay of 10 -8 In the first stage, the SURE-based denoising model F φ The training is performed for 40 cycles with a batch size of 4. The initial learning rate is set to 1×10 -3 , and reduce it by a factor of 10 in the 30th cycle. Subsequently, in the second stage, the virtual scan reconstructs the network F θ The training was performed for 20 epochs with a batch size of 2 and an initial learning rate of 5×10 -4 , reducing it by a factor of 10 in the 10th epoch. In each epoch, we randomly selected 40 complete simulated transient signals for each relay surface and extracted them to generate irregular undersampled transients for training. All models were trained on two NVIDIA 3090 GPUs, taking a total of nearly 40 hours. For the loss function, the hyperparameters ε, β, and b were set to 0.1, 0.001, and 4, respectively.

[0101] Example 2:

[0102] Based on Example 1, but with the following differences:

[0103] The method of the present invention is compared with three traditional direct reconstruction algorithms (LCT, FK, RSD), two learning-based algorithms (unsupervised NeTF, supervised USM) and one iterative algorithm (CC-SOCR). Since LCT, FK, RSD and USM are designed for dense regular scanning, the present invention uses a zero-filled version of irregular undersampled transients as input. NeTF and CC-SOCR are able to directly input irregular undersampled transients. For a fair comparison, the present invention also applies a SURE-based denoiser to all compared methods for pre-denoising. As an exception, the present invention does not apply a pre-denoising step to CC-SOCR because there is no obvious improvement in the results of CC-SOCR reconstruction. The reason is that the powerful CC-SOCR regularization inherently performs denoising to some extent.

[0104] We first tested our method on publicly available real datasets and then on captured data. Without proper regularization, direct reconstruction methods such as LCT and RSD exhibit severe aliasing artifacts due to undersampling. Due to its sensitivity to depth, FK cannot recover the image. Figure 3The distal structure shown. NeTF tends to generate blurred shapes due to its difficulty in utilizing the limited information in the IUT. For irregular undersampling, USM is almost overwhelmed by aliasing artifacts, but can recover acceptable results in regular undersampling (e.g. Figure 4 ). CC-SOCR is able to recover the main object structure in all test cases, but loses detailed structure at farther depths (such as "Man Deer", "Letter", "Deer" and "Teaser"), and performs poorly on the "Dragon" object with smooth material or the letter "SU" object with anti-reflective material. The method of the present invention achieves the best quality across a variety of objects and sampling modes. In short, despite being trained only on a simple alphanumeric dataset and blinds-like relay surfaces, the method produces stable results on real-world data with a variety of attributes and relay surfaces. It significantly outperforms the range space solver LCT, and the virtual scan reconstruction network proposed by the method successfully removes aliasing artifacts while preserving the scene structure. The excellent generalization ability highlights the effectiveness of the proposed method in learning out-of-domain spatial information.

[0105] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An unsupervised under-sampling non-viewing-area imaging method based on virtual scanning, characterized in that: The specific steps include: S1. Constructing an undersampled NLOS reconstruction simulation transient dataset: Using a publicly available transient renderer and ShapeNet 3D object data, we construct a simulation transient dataset. We add noise and temporal jitter features based on the NLOS imaging single-photon detector imaging model, and generate an undersampled NLOS transient dataset based on a set scanning pattern. S2. Build a network framework: Design a denoiser network with an encoder-decoder structure consisting of partial convolutions and a reconstruction network combined with physical priors. Connect the denoiser network and the reconstruction network to reconstruct hidden scenes from irregular undersampled transients. S3. Design framework and build model: Combine the undersampled NLOS reconstruction simulation transient dataset and network framework described in S1-S2 to design a NLOS reconstruction scheme. The scheme specifically includes the following contents: ① Undersampled transient denoising: The captured irregular undersampled transient measurements are filled into the full sampling grid by zero-filling method, and the zero-filled transient and the mask are input into the denoiser network to obtain the initial denoised transient, which is then multiplied by the mask to remove the artifacts introduced by the network and obtain the denoised undersampled transient; ② Hidden scene reconstruction: The denoised undersampled transient is initially reconstructed using the traditional LCT algorithm to obtain the value domain information of the true reconstruction result. This information is then input into the reconstruction network to obtain the zero domain information of the true reconstruction result. The value domain information is combined with the zero domain information to obtain the albedo reconstruction volume of the hidden scene. ③ Design a virtual scanning training strategy: Based on the zero-value domain decomposition principle, an unsupervised virtual scanning training strategy and its loss function are designed to enable the neural network to learn the ability to recover zero-domain information; S4. Training the unsupervised non-line-of-sight reconstruction framework: Use the deep learning Pytorch framework to train the constructed network framework and virtual scanning reconstruction network model in stages, and repeatedly traverse the undersampled non-line-of-sight reconstruction simulation transient dataset constructed in S1 until the model converges.

2. The unsupervised under-sampling non-viewing-area imaging method based on virtual scanning according to claim 1, characterized in that: The S1 specifically includes the following contents: S1.

1. Select and organize ShapeNet 3D object data; S1.

2. Modeling single-photon detector imaging under non-line-of-sight imaging conditions; S1.

3. Use the publicly available non-line-of-sight imaging transient measurement renderer to simulate ideal transient measurements based on ShapeNet 3D object data. S1.

4. Post-process the simulated ideal transient measurement based on the single-photon detector imaging model to make it close to the fully sampled transient measurement of the real acquisition; S1.

5. Extract undersampled data from the simulated full-sampled transient measurement according to the selected scanning pattern for training and construct a data set.

3. The unsupervised under-sampling non-viewing-area imaging method based on virtual scanning according to claim 2, characterized in that: The S2 specifically includes the following contents: S2.

1. A non-line-of-sight single-photon detector imaging model is given. The model is an imaging model that considers low-photon time-resolution detectors. The specific function is expressed as: where u represents the noise-free measurement, represents the noisy measurement, b represents the dark photons and background noise; The system model of non-line-of-sight imaging is simplified to a linear inverse problem form, namely: u=Hρ Where H represents the forward degradation operator; ρ represents the three-dimensional albedo volume of the hidden scene; S2.

2. According to the imaging model given in S2.1, we derive an unsupervised unbiased estimator based on Stein's Unbiased Risk Estimation and an unsupervised denoising loss function. The specific function is expressed as: in, represents the unsupervised denoising loss function based on Stein's Unbiased Risk Estimation; i∈{0,…,I}, g∈{0,…,G}, I, G represent the number of transients and the number of sampling operators respectively; s represents the number of scanning points; t represents the time dimension, F φ represents the denoiser network; ε is a positive number; e is a binary vector; S2.

3. Use partial convolution to build a network model based on the encoder-decoder structure, where the partial convolution operator is expressed as: Where M represents the mask; x represents the input; W represents the convolution kernel parameter; b represents the bias, and x' represents the output; S2.

4. Construct a Unet-like network that combines physical priors and attention mechanism. Use the traditional LCT algorithm to perform domain conversion and initial reconstruction on the denoised transient to obtain value domain information. Then use the reconstruction network to restore the zero domain information. The restored zero domain information and value domain information are added together to experimentally reconstruct the complete hidden scene.

4. The unsupervised under-sampling non-viewing-area imaging method based on virtual scanning according to claim 3, characterized in that: The above-mentioned plan ③ specifically includes the following contents: 3.1) Select a fully sampled transient and a sampling matrix, extract data from the fully sampled transient according to the sampling matrix to form undersampled data; 3.2) Fill the undersampled data to the fully sampled grid using zero padding, input it and the mask into the denoiser network to obtain the initial denoised transient, and then multiply it with the mask to remove the artifacts introduced by the network to obtain the denoised undersampled transient; 3.3) The denoised undersampled transient is initially reconstructed using the traditional LCT algorithm to obtain the range information of the true reconstruction result. The range information of the true reconstruction result is input into the reconstruction network. The output of the reconstruction network is combined with the range information to obtain the albedo reconstruction volume of the hidden scene. 3.4) Observe the albedo reconstruction volume through the fully sampled forward operator, and then select a sampling matrix different from that in step 3.1) and multiply it with the new fully sampled observation to obtain a new undersampled transient; 3.5) Repeat step 3.3) with the newly obtained undersampled transient in step 3.4) to obtain a new albedo reconstruction volume; 3.6) The reconstruction network in step 3.3 is supervised by a measurement consistency loss function and a virtual scan loss function, where the measurement consistency loss function is expressed as: Among them, L MC represents the measurement consistency loss function; Indicates that the sampling matrix M is used g The obtained undersampled transient; H g Representation and sampling matrix M g The corresponding forward operator; f θ Represents the traditional algorithm LCT and the reconstruction network F θ The reconstruction modules composed of The virtual scanning loss function is expressed as: Among them, L VS is the virtual scanning loss function; H k Representation and sampling matrix M k The corresponding forward operator.

5. The unsupervised under-sampling non-viewing-area imaging method based on virtual scanning according to claim 4, characterized in that: The S4 includes the following: S4.

1. Build an optimizer and gradient update strategy based on the Pytorch framework, and train the network using the undersampled non-field-of-view reconstruction simulation transient dataset constructed in S1. S4.

2. Use the unsupervised denoising loss function based on Stein's Unbiased Risk Estimation in S2 to train the denoiser network. S4.

3. Freeze the denoiser network weights and train the reconstruction network proposed in S2.4 using the virtual scanning strategy proposed in Scheme ③.

Images