Hyperspectral image cross-domain classification method and device suitable for complex ground scenes

Abstract

The present invention relates to a cross-domain classification method and device for hyperspectral images applicable to complex land object scenes, including: acquiring hyperspectral images, and independently extracting features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images; performing interaction and fusion operations on the features of each dimension; achieving matching of the feature spaces of the source domain and the target domain; training a cross-domain classification model for hyperspectral images, and jointly optimizing the model through a multi-task loss function; predicting the target domain data to obtain the final classification result. The beneficial effects of the present invention are: the present invention successfully addresses the challenges brought about by the difference in feature distribution under cross-temporal and cross-scene conditions by performing deep adaptive optimization at multiple levels of domain-level and class-level feature alignment. The algorithm not only improves the global consistency of the feature spaces of the source domain and the target domain, but also ensures the accurate alignment of features of different categories through a refined alignment strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image classification, and more specifically, to a hyperspectral image cross-domain classification method and device applicable to complex land object scenes. Background Art

[0002] In the field of remote sensing image classification, hyperspectral imagery, due to its high spatial and spectral resolution, provides rich ground-based information and is widely used in land cover classification, environmental monitoring, and resource management. However, hyperspectral images captured at different times or in different scene environments often exhibit significant differences in feature distribution due to various factors, such as shooting conditions, climate change, and seasonal differences. These differences are particularly pronounced in complex environments such as wetlands, due to the high heterogeneity of ground-based features and the complexity of their spectral signatures. This feature inconsistency significantly impacts the generalization ability of traditional classification models, resulting in poor performance when processing hyperspectral imagery across time or scenes.

[0003] Traditional hyperspectral image classification methods usually rely on source domain data for model training and assume that the trained model can be seamlessly applied to target domain data. However, due to the difference in feature distribution between the source domain and the target domain, this assumption is difficult to hold in practical applications. Especially in complex environments such as wetlands, due to the diversity of surface cover types and drastic changes in environmental conditions (such as light and humidity), this problem is more significant, resulting in a significant decrease in classification accuracy. To address these challenges, domain adaptation (DA) technology has been introduced to improve the performance of the model on target domain data. Domain adaptation aims to improve the generalization ability of the model by reducing the difference in feature distribution between the source domain and the target domain. This usually involves techniques such as feature map alignment, recalibration, and sample selection to enable the trained model to better adapt to the feature distribution of the target domain.

[0004] With the rapid development of deep learning technology, many advanced domain adaptation methods have been proposed and applied to hyperspectral image classification tasks. These methods, including generative adversarial networks (GANs), domain adversarial neural networks (DANNs), and adaptive feature selection techniques, aim to align the feature distributions of the source and target domains through different strategies. For example, GANs generate pseudo-source samples that mimic target domain samples through adversarial training, thereby improving the model's adaptability to the target domain. This method generates pseudo samples with a similar distribution to the target domain, enabling the classifier to better learn target domain features, thereby enhancing the model's generalization ability. Domain adversarial neural networks, on the other hand, directly align the feature distributions of the source and target domains by introducing a domain adversarial loss. This domain adversarial loss trains the network's feature extractor to make the source and target domain features as similar as possible in a shared space, thereby reducing the discrepancy between the domains. Adaptive feature selection techniques reduce the impact of feature discrepancies on classification by selecting the most useful features for the classification task. This method reduces feature dimensionality and improves classification performance by filtering out features that contribute less to the classification task. Although these techniques have made some progress in domain adaptation tasks in hyperspectral image classification, several challenges remain. Existing methods may have limited performance when dealing with complex environmental conditions and highly variable spectral features. For example, when dealing with complex environments such as wetlands, these methods may have difficulty coping with the high variability of spectral features. Although generative adversarial networks and domain adversarial neural networks can alleviate the difference in feature distribution between the source and target domains to a certain extent, their alignment effect may still be affected by actual environmental changes. In addition, although adaptive feature selection techniques can reduce feature differences, they usually require a lot of computing resources and time for feature screening, and their effectiveness is also limited by the quality of the selected features and the applicability of the screening method. Summary of the Invention

[0005] The purpose of the present invention is to address the deficiencies of the existing technology and propose a cross-domain classification method and device for hyperspectral images suitable for complex ground object scenes.

[0006] First, a cross-domain classification method for hyperspectral images suitable for complex ground scenes is provided, including:

[0007] S1. Acquire hyperspectral images and independently extract features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images;

[0008] S2, interact and fuse the features of each dimension;

[0009] S3. By optimizing the feature distribution mapping and combining the dual constraints of domain alignment and class alignment, the feature space of the source and target domains is matched.

[0010] S4. Use labeled source domain data and unlabeled target domain data to train a cross-domain hyperspectral image classification model, and jointly optimize the model through a multi-task loss function;

[0011] S5. Use the trained model to predict the target domain data and obtain the final classification results.

[0012] Preferably, in S1, the mathematical expression for independently extracting features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral image is:

[0013]

[0014] Where X∈R H×W×B×T is the input hyperspectral image, H, W and B represent the height, width and number of bands of the image respectively, T represents the time or scene dimension, H ′ 、W ′ 、B ′ and T ′ They represent the corresponding dimension sizes after feature extraction of hyperspectral images in different dimensions, L h 、L s 、L t are the number of feature extraction layers for spatial, spectral, and temporal / scene dimensions, represents the weight matrix of the lth layer corresponding to the spatial, spectral, and temporal / scene dimensions, are the bias terms corresponding to each layer, σ(·) is the activation function, * represents the convolution operation, F spatial 、F spectral and F temporal Represent the extracted spatial, spectral and temporal / scene features, respectively.

[0015] Preferably, in S2, the mathematical expression for performing interaction and fusion operations on features of each dimension is:

[0016]

[0017] Where i, j represents the index position of the feature, σ(·) represents the activation function used to perform nonlinear transformation on the interaction value of the input feature, and W att is the attention weight matrix, represents the dot product operation of the vector, Represents the multidimensional concatenation vector of the i-th feature, A ij represents the interaction value between the i-th and j-th features, measuring their association strength under the attention mechanism; F fusion It represents the features obtained by interactively fusing spatial, spectral, and temporal / scene features. The coefficients α, β, and γ are used to control the weights of features in each dimension, satisfying α + β + γ = 1.

[0018] Preferably, in S3, the mathematical expression for achieving the matching of the source domain and the target domain feature space is:

[0019] L domain_align =‖M(F s )-F T ‖ 2

[0020]

[0021] Among them, F s and F T are the source domain and target domain features, and represents the n-th category feature of the source domain and the target domain, N represents the number of categories, M(·) is the feature mapping function, ‖·‖ 2 represents the Euclidean norm, L domain_align is the domain alignment loss between the source and target domains, L class_align is the category alignment loss between the source and target domains.

[0022] Preferably, in S4, the source domain classification loss is calculated, and the model parameters are optimized by combining the source domain classification loss and feature alignment loss. The mathematical expression is:

[0023]

[0024] L=L s +λ1L domain_align +λ2L class_align

[0025] Among them, L s is the source domain classification loss calculated using the cross entropy loss function, y i,c is the true label of the i-th sample in category c, is the predicted probability, L is the final comprehensive optimization objective function, and λ1 and λ2 are adjustment parameters used to control the weights of different loss terms.

[0026] Preferably, in S5, the target domain data is input into the trained classification model, and its mathematical expression is:

[0027]

[0028] in, is the prediction result of the target domain data, W and b are model parameters.

[0029] In a second aspect, a hyperspectral image cross-domain classification device suitable for complex ground object scenes is provided, which is used to execute any of the methods described in the first aspect, including:

[0030] The acquisition module is used to acquire hyperspectral images and independently extract features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images;

[0031] Interactive fusion module, used to interact and fuse features of each dimension;

[0032] The matching module is used to achieve matching between the source and target domain feature spaces by optimizing the feature distribution mapping and combining the dual constraints of domain alignment and class alignment;

[0033] The training module is used to train a cross-domain hyperspectral image classification model using labeled source domain data and unlabeled target domain data, and jointly optimize the model through a multi-task loss function;

[0034] The prediction module is used to use the trained model to predict the target domain data and obtain the final classification results.

[0035] According to a third aspect, a computer storage medium is provided, wherein a computer program is stored in the computer storage medium; when the computer program is executed on a computer, the computer executes any one of the methods described in the first aspect.

[0036] In a fourth aspect, an electronic device is provided, including:

[0037] Memory, used to store computer programs;

[0038] A processor is used to execute the computer program to implement any method as described in the first aspect.

[0039] The beneficial effects of the present invention are:

[0040] 1. This paper innovatively designs a Multidimensional Feedback Feature Deep Mining Network (MRFDMNet), which enables parallel feature extraction from hyperspectral images in three dimensions: spatial, spectral, and temporal / scene. This significantly improves the efficient mining and representation of multidimensional information. Combining deep convolutional layers with a residual structure, this network captures the geometric characteristics of objects in the spatial dimension, extracts the unique spectral properties of materials in the spectral dimension, and captures the dynamic changes of objects in the temporal and scene dimensions. This enhances the accuracy and detail perception of hyperspectral imagery in complex environments.

[0041] 2. This invention overcomes the limitations of traditional feature fusion methods by introducing an innovative multidimensional interactive fusion strategy, significantly enhancing the synergy and information integration capabilities between features of different dimensions. This mechanism optimizes the interactive expression of spatial, spectral, and temporal / scene features, improves global feature consistency, and equips the model with stronger multidimensional information perception capabilities. In complex environment classification tasks, the model accuracy and robustness are significantly improved.

[0042] 3. The proposed Multi-Level Alignment Consensus Algorithm (MLACA) successfully addresses the challenges posed by differences in feature distribution across time and scene conditions by performing deep adaptive optimization at multiple levels of domain-level and class-level feature alignment. This algorithm not only improves the global consistency of the source and target domain feature spaces, but also ensures accurate alignment of features from different categories through a refined alignment strategy. This significantly enhances the model's generalization and robustness in complex environments, making it suitable for remote sensing image classification tasks in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 Flowchart of the cross-domain classification method of hyperspectral images applicable to complex ground object scenes provided by the present invention;

[0044] Figure 2 Schematic diagram of two sets of hyperspectral cross-temporal domain adaptation data for Hangzhou Bay and Yancheng provided by the present invention;

[0045] Figure 3 Schematic diagram of the Yellow River Estuary-Yancheng hyperspectral cross-scene domain adaptation data provided by the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further described below with reference to the following examples. The following examples are provided only to facilitate understanding of the present invention. It should be noted that, without departing from the principles of the present invention, it is possible for a person skilled in the art to make various modifications to the present invention, and such improvements and modifications fall within the scope of the claims of the present invention.

[0047] Example 1:

[0048] In order to solve the problems of the prior art, Example 1 of the present application provides a cross-domain classification method for hyperspectral images suitable for complex land object scenes, which aims to effectively solve the classification problems caused by differences in feature distribution of hyperspectral images under different time and scene environments. This technology aims to deal with the problem of inconsistency of hyperspectral image features caused by changes in the environment or conditions by improving feature extraction and domain alignment strategies. The experiment was carried out on a self-constructed wetland hyperspectral dataset. The results showed that the method showed excellent performance and stability in a variety of classification tasks, demonstrating its broad application prospects in the fields of environmental monitoring and resource management. This research result is not only of great significance to the development of hyperspectral remote sensing image classification technology, but also provides an effective solution for environmental change analysis and surface cover monitoring in complex environments.

[0049] Specifically, such as Figure 1 As shown, the method includes:

[0050] S1. Acquire hyperspectral images and independently extract features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images.

[0051] In S1, a parallel 3D feature extraction module is designed to fully exploit the multidimensional information of the data. The purpose of this step is to ensure that the model can fully capture the spatial structure, spectral characteristics, and temporal or scene changes of complex objects by performing multidimensional feature extraction on hyperspectral imagery, providing accurate and comprehensive input for subsequent feature fusion and classification.

[0052] The mathematical expression for independently extracting features from the spatial, spectral, and temporal / scene dimensions of hyperspectral images is:

[0053]

[0054] Where X∈R H×W×B×T is the input hyperspectral image, H, W and B represent the height, width and number of bands of the image respectively, T represents the time or scene dimension, H ′ 、W ′ 、B ′ and T ′ They represent the corresponding dimension sizes after feature extraction of hyperspectral images in different dimensions, L h 、L s 、L t are the number of feature extraction layers for spatial, spectral, and temporal / scene dimensions, represents the weight matrix of the lth layer corresponding to the spatial, spectral, and temporal / scene dimensions, are the bias terms corresponding to each layer, σ(·) is the activation function, * represents the convolution operation, F spatial 、F spectral and F temporal Represent the extracted spatial, spectral and temporal / scene features, respectively.

[0055] S2. Perform interaction and fusion operations on features of each dimension.

[0056] In S2, an innovative interactive feature fusion mechanism is introduced to interactively operate on features of each dimension to achieve efficient fusion and enhancement of information. The purpose of this step is to enhance the complementarity and synergy between information by interactively fusing features of each dimension, thereby enhancing the richness and robustness of feature representation and enabling the model to better cope with classification tasks in complex environments. The mathematical expression for the interactive and fusion operations on features of each dimension is:

[0057]

[0058] Among them, i, j represents the index position of the feature, Watt is the attention weight matrix, represents the dot product operation of the vector, represents the multidimensional concatenation vector of the i-th feature, σ(·) represents the activation function for nonlinear transformation of the input feature interaction value, A ij represents the interaction value between the i-th and j-th features, measuring their association strength under the attention mechanism; F fusion It represents the features obtained by interactively fusing spatial, spectral, and temporal / scene features. The coefficients α, β, and γ are used to control the weights of features in each dimension, satisfying α + β + γ = 1.

[0059] S3. Aiming at the differences in feature distribution between different times and scenes, an efficient adaptive alignment strategy is designed. By optimizing the feature distribution mapping and combining the dual constraints of domain alignment and class alignment, the precise matching of the feature space of the source and target domains is achieved.

[0060] This step aims to reduce the feature distribution deviation under different time and scene conditions by adaptively aligning the source domain and target domain features, ensuring that the feature space can maintain global consistency and intra-class aggregation across time and scenes, thereby improving the model's feature expression ability and classification accuracy in a changing environment.

[0061] In S3, the mathematical expression for achieving the matching of the source domain and target domain feature spaces is:

[0062] L domain_align =‖M(F s )-F T ‖ 2

[0063]

[0064] Among them, F s and F T are the source domain and target domain features, and represents the n-th category feature of the source domain and the target domain, N represents the number of categories, M(·) is the feature mapping function, ‖·‖ 2 represents the Euclidean norm, L domain_align is the domain alignment loss between the source and target domains, L class_align For the category alignment loss S4 between the source domain and the target domain, the hyperspectral image cross-domain classification model is trained using labeled source domain data and unlabeled target domain data, and the model is jointly optimized through a multi-task loss function.

[0065] S4. Use labeled source domain data and unlabeled target domain data to train a cross-domain classification model for hyperspectral images, and jointly optimize the model through a multi-task loss function.

[0066] The purpose of this step is to use labeled source domain data for supervised learning and unlabeled target domain data for unsupervised alignment. Combined with the multi-task loss function, the model parameters are optimized to achieve comprehensive improvement in feature alignment and classification performance between the source and target domains.

[0067] S5. Use the trained model to predict the target domain data and obtain the final classification results.

[0068] The purpose of this step is to accurately classify the target domain data through the trained model, verify the model's generalization ability across time and scenarios, and obtain the final high-precision classification results.

[0069] Example 2:

[0070] Based on Example 1, Example 2 of the present application provides a more specific cross-domain classification method for hyperspectral images applicable to complex ground object scenes, including:

[0071] S1. Acquire hyperspectral images and independently extract features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images.

[0072] S2. Perform interaction and fusion operations on features of each dimension.

[0073] S3. By optimizing the feature distribution mapping and combining the dual constraints of domain alignment and class alignment, the feature space matching of the source domain and the target domain is achieved.

[0074] S4. Use labeled source domain data and unlabeled target domain data to train a cross-domain classification model for hyperspectral images, and jointly optimize the model through a multi-task loss function.

[0075] In S4, the source domain classification loss is calculated, and the model parameters are optimized by combining the source domain classification loss and feature alignment loss. Its mathematical expression is:

[0076]

[0077] L=L s +λ1L domain_align +λ2L class_align

[0078] Among them, L s is the source domain classification loss calculated using the cross entropy loss function, y i,c is the true label of the i-th sample in category c, is the prediction probability, L is the final comprehensive optimization objective function, which combines the source domain classification loss, domain alignment loss and category alignment loss, λ1 and λ2 are adjustment parameters used to control the weights of different loss terms.

[0079] S5. Use the trained model to predict the target domain data and obtain the final classification results.

[0080] In S5, the target domain data is input into the trained classification model, and its mathematical expression is:

[0081]

[0082] in, is the prediction result of the target domain data, W and b are model parameters.

[0083] The classification results of the proposed method were compared with those of four other mainstream methods using a self-constructed hyperspectral dataset (images from the ZY1-02D satellite). The specific results are shown in Tables 1-3. The results show that on the Hangzhou Bay cross-temporal data dataset, the proposed method achieved an overall accuracy (OA) of 87.40% with a Kappa coefficient of 0.8324; on the Yancheng cross-temporal data dataset, the proposed method achieved an overall accuracy (OA) of 82.07% with a Kappa coefficient of 0.7662; and on the Yellow River Estuary-Yancheng cross-scene dataset, the proposed method achieved an overall accuracy (OA) of 67.82% with a Kappa coefficient of 0.6053. The results demonstrate that on these three sets of hyperspectral domain adaptation datasets, the proposed method significantly outperforms the other compared methods, demonstrating strong cross-temporal and cross-scene adaptability, further validating its effectiveness in complex remote sensing image classification tasks.

[0084] Table 1 Experimental results of different domain adaptation methods on Hangzhou Bay cross-temporal data (2021.01.29→2022.01.03)

[0085]

[0086] Table 2 Experimental results of different domain adaptation methods on Yancheng cross-temporal data (2022.02.24→2023.01.20)

[0087]

[0088] Table 3 Experimental results of different domain adaptation methods on the Yellow River Estuary-Yancheng cross-scene data (2020.06.28→2020.09.06)

[0089]

[0090]

[0091] It should be noted that the parts in this embodiment that are the same or similar to those in Example 1 can be referenced to each other and will not be described in detail in this application.

[0092] Example 3:

[0093] Based on Examples 1 and 2, Example 3 of the present application provides a hyperspectral image cross-domain classification device suitable for complex ground object scenes, including:

[0094] The acquisition module is used to acquire hyperspectral images and independently extract features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images;

[0095] Interactive fusion module, used to interact and fuse features of each dimension;

[0096] The matching module is used to achieve matching between the source and target domain feature spaces by optimizing the feature distribution mapping and combining the dual constraints of domain alignment and class alignment;

[0097] The training module is used to train a cross-domain hyperspectral image classification model using labeled source domain data and unlabeled target domain data, and jointly optimize the model through a multi-task loss function;

[0098] The prediction module is used to use the trained model to predict the target domain data and obtain the final classification results.

[0099] Specifically, the device provided in this embodiment is a device corresponding to the method provided in Examples 1 and 2. Therefore, the parts in this embodiment that are the same or similar to those in Examples 1 and 2 can be referenced to each other and will not be repeated in this application.

Claims

1. A cross-domain classification method for hyperspectral images suitable for complex ground scenes, characterized by: include: S1. Acquire hyperspectral images and independently extract features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images; S2. Perform interaction and fusion operations on the features of each dimension. In S2, the mathematical expression for performing interaction and fusion operations on the features of each dimension is: Among them, i, j represents the index position of the feature, W att is the attention weight matrix, represents the dot product operation of the vector, represents the multidimensional concatenation vector of the i-th feature, σ(·) represents the activation function for nonlinear transformation of the input feature interaction value, A ij represents the interaction value between the i-th and j-th features, measuring their association strength under the attention mechanism; F fusion It represents the features obtained by interactively fusing spatial, spectral, and temporal / scene features. The coefficients α, β, and γ are used to control the weights of features in each dimension, satisfying α+β+γ=1. S3. By optimizing the feature distribution mapping and combining the dual constraints of domain alignment and class alignment, the matching of the source domain and target domain feature spaces is achieved. In S3, the mathematical expression for achieving the matching of the source domain and target domain feature spaces is: L domain_align =‖M(F s )-F T ‖ 2 Among them, F s and F T are the source domain and target domain features, and represents the n-th category feature of the source domain and the target domain, N represents the number of categories, M(·) is the feature mapping function, ‖·‖ 2 represents the Euclidean norm, L domain_align is the domain alignment loss between the source and target domains, L class_align is the category alignment loss between the source domain and the target domain; S4. Use labeled source domain data and unlabeled target domain data to train a cross-domain hyperspectral image classification model, and jointly optimize the model through a multi-task loss function; S5. Use the trained model to predict the target domain data and obtain the final classification results.

2. The cross-domain classification method for hyperspectral images applicable to complex ground object scenes according to claim 1 is characterized in that: In S1, the mathematical expression for independently extracting features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral image is: Where X∈R H×W×B×T is the input hyperspectral image, H, W and B represent the height, width and number of bands of the image respectively, T represents the time or scene dimension, H ′ 、W ′ 、B ′ and T ′ They represent the corresponding dimension sizes after feature extraction of hyperspectral images in different dimensions, L h 、L s 、L t are the number of feature extraction layers for spatial, spectral, and temporal / scene dimensions, represents the weight matrix of the lth layer corresponding to the spatial, spectral, and temporal / scene dimensions, are the bias terms corresponding to each layer, σ(·) is the activation function, * represents the convolution operation, F spatial 、F spectral and F temporal Represent the extracted spatial, spectral and temporal / scene features, respectively.

3. The cross-domain classification method for hyperspectral images applicable to complex terrain scenes according to claim 2 is characterized in that: In S4, the source domain classification loss is calculated, and the model parameters are optimized by combining the source domain classification loss and feature alignment loss. Its mathematical expression is: L=L s +λ1L domain_align +λ2L class_align Among them, L s is the source domain classification loss calculated using the cross entropy loss function, y i,c is the true label of the i-th sample in category c, is the predicted probability, L is the final comprehensive optimization objective function, and λ1 and λ2 are adjustment parameters used to control the weights of different loss terms.

4. The cross-domain classification method for hyperspectral images applicable to complex ground object scenes according to claim 3 is characterized in that: In S5, the target domain data is input into the trained classification model, and its mathematical expression is: in, is the prediction result of the target domain data, W and b are model parameters.

5. A hyperspectral image cross-domain classification device suitable for complex ground scenes, characterized by: The method for executing any one of claims 1 to 4 comprises: The acquisition module is used to acquire hyperspectral images and independently extract features from the spatial, spectral, and temporal / scene dimensions of the hyperspectral images; The interactive fusion module is used to perform interactive and fusion operations on the features of each dimension. In the interactive fusion module, the mathematical expression for the interactive and fusion operations on the features of each dimension is: Among them, i, j represents the index position of the feature, W att is the attention weight matrix, represents the dot product operation of the vector, represents the multidimensional concatenation vector of the i-th feature, σ(·) represents the activation function for nonlinear transformation of the input feature interaction value, A ij represents the interaction value between the i-th and j-th features, measuring their association strength under the attention mechanism; F fusion It represents the features obtained by interactively fusing spatial, spectral, and temporal / scene features. The coefficients α, β, and γ are used to control the weights of features in each dimension, satisfying α+β+γ=1. The matching module is used to achieve matching of the source domain and target domain feature spaces by optimizing the feature distribution mapping and combining the dual constraints of domain alignment and class alignment. In the matching module, the mathematical expression for achieving matching of the source domain and target domain feature spaces is: L domain_align =‖M(F s )-F T ‖ 2 Among them, F s and F T are the source domain and target domain features, and represents the n-th category feature of the source domain and the target domain, N represents the number of categories, M(·) is the feature mapping function, ‖·‖ 2 represents the Euclidean norm, L domain_align is the domain alignment loss between the source and target domains, L class_align is the category alignment loss between the source domain and the target domain; A training module is used to train a cross-domain hyperspectral image classification model using labeled source domain data and unlabeled target domain data, and jointly optimize the model through a multi-task loss function; The prediction module is used to use the trained model to predict the target domain data and obtain the final classification results.

6. A computer storage medium, characterized in that The computer storage medium stores a computer program; when the computer program is run on a computer, the computer executes the method according to any one of claims 1 to 4.

7. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the method according to any one of claims 1 to 4.

Images