Hyperspectral image fusion method and system based on superband aggregation and Dirichlet

Abstract

The present invention relates to the technical field of hyperspectral image processing, and provides a hyperspectral image fusion method and system based on superband aggregation and Dirichlet. The method comprises acquiring a LRHS image through a remote sensing optical imaging system; fusing the LRHS image with a panchromatic image through a spatial spectral data distribution fusion module to obtain a PANHS image; performing band aggregation on the LRHS through an aggregation algorithm, calculating a compressed correlation graph based on superbands, and extracting current state features based on a recurrent memory network; obtaining memory state features through a Dirichlet autoencoder module; obtaining fusion features through a recurrent memory network; integrating the fusion features into the PANHS image to obtain integrated features; and obtaining a high-spatial-resolution hyperspectral image through a residual block. The present invention significantly improves the spatial resolution of hyperspectral images while maintaining spectral consistency, providing technical support for the refined analysis and practical application of hyperspectral images.

Description

Technical Field

[0001] The present invention relates to the technical field of hyperspectral image processing, and in particular to a hyperspectral image fusion method and system based on superband aggregation and Dirichlet. Background Art

[0002] Hyperspectral (HS) images, obtained by sampling hundreds of continuous, narrow spectral bands using a spectral imaging system, provide rich spectral information and are commonly used to study the spectral characteristics of materials. However, due to the physical design limitations of optical imaging systems, a trade-off between spectral and spatial resolution is required. Single-spectral imaging systems are unable to produce high-spatial-resolution hyperspectral (HRHS) images, often resulting in low-spatial-resolution hyperspectral (LRHS) images. This limits the application of HS images in fields such as mineral exploration, ecosystem monitoring, and agricultural monitoring. Panchromatic (PAN) imaging systems, which only output single-band images, offer high spatial resolution but low spectral resolution. Hyperspectral pansharpening (HS pansharpening) aims to fully utilize the spectral and spatial information of HS images to generate HRHS images. Due to the high market demand for HRHS images, this technology has attracted widespread attention and has demonstrated promising performance in numerous downstream remote sensing imagery tasks, including military surveillance, environmental monitoring, and target recognition and classification.

[0003] Traditional hyperspectral pan-sharpening methods can be further divided into four categories: component substitution (CS)-based methods, multiresolution analysis (MRA)-based methods, Bayesian-based methods, and model-based methods. Although traditional methods can improve the spatial resolution of HS images, the quality of the fused image will be severely degraded due to inappropriate prior knowledge modeling, unknown sensor characteristics, mismatched prior assumptions, and the fact that the features manually constructed from the dictionary are inconsistent with reality. There are still significant deficiencies in hyperspectral image sequence modeling and fusion network interpretability. These problems restrict the promotion and performance improvement of hyperspectral image pan-sharpening technology in practical applications, and breakthroughs and innovations are urgently needed at the theoretical and technical levels. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the related art. To this end, the present invention provides a hyperspectral image fusion method and system based on superband aggregation and Dirichlet. By combining the superband aggregation method with a recurrent memory network, the present invention realizes sequential spatial spectral fusion modeling between PAN and LRHS images, achieves the physical interpretability of hyperspectral decomposition, significantly improves the spatial resolution of hyperspectral images, and maintains the spectral consistency of hyperspectral images, providing technical support for the refined analysis and practical application of hyperspectral images.

[0005] The present invention provides a hyperspectral image fusion method based on superband aggregation and Dirichlet, comprising:

[0006] S1: Acquire LRHS images through remote sensing optical imaging system;

[0007] S2: The LRHS image and the panchromatic image are fused through the spatial spectral data distribution fusion module to obtain the PANHS image;

[0008] S3: Aggregate the LRHS image bands using an aggregation algorithm to obtain superbands, calculate the LRHS image compression correlation map based on the superbands, and calculate the PANHS image compression correlation map based on the superbands;

[0009] S4: extracting the spatial spectral correlation of the LRHS image compression correlation graph according to the recurrent memory network to obtain the current state characteristics of the LRHS image; extracting the spatial spectral correlation of the PANHS image compression correlation graph according to the recurrent memory network to obtain the current state characteristics of the PANHS image;

[0010] S5: The spectral and spatial information of the current state features of the LRHS image and the current state features of the PANHS image are fused through the Dirichlet autoencoder module to obtain the memory state features; candidate features are extracted through the recurrent memory network, and the memory state features and candidate features are fused to obtain the fused features;

[0011] S6: Integrate the fused features into the PANHS image to obtain the integrated features;

[0012] S7: Repeat steps S3 to S6 to obtain multiple integrated features, and fuse the multiple integrated features through the residual block to obtain a hyperspectral image with high spatial resolution.

[0013] Furthermore, step S2 includes:

[0014] S21: obtaining an upsampled LRHS image by upsampling the LRHS image;

[0015] S22: The upsampled LRHS image and the panchromatic image are fused through the spatial spectral data distribution fusion module to obtain a PANHS image.

[0016] Furthermore, step S22 includes:

[0017] S221: Calculate the adaptive mean of the upsampled LRHS image and the adaptive variance of the upsampled LRHS image by parallel convolution.

[0018] S222: Assign the adaptive mean of the upsampled LRHS image and the adaptive variance of the upsampled LRHS image to the panchromatic image to obtain a PANHS image.

[0019] Furthermore, step S3 includes:

[0020] S31: Through the aggregation algorithm, the high correlation between the LRHS image waves is used to aggregate the bands and obtain the super bands;

[0021] S32: Calculate a correlation map of the LRHS image based on the superband; select the first k bands in the correlation map to form a sub-compression correlation map;

[0022] S33: performing channel selection on the LRHS image according to the secondary compression correlation map, selecting a band in each superband to obtain the LRHS compression correlation map, performing channel selection on the PANHS image according to the secondary compression correlation map to obtain the PANHS compression correlation map.

[0023] Furthermore, the aggregation algorithm is an improved K-mean classical aggregation algorithm, and only local iterative calculations are performed in each iteration.

[0024] Furthermore, in step S4, the calculation expression of the current state feature of the LRHS image is:

[0025]

[0026]

[0027]

[0028] in, For the Group LRHS forget gate state, is the Sigmoid activation function, For the Group LRHS compressed association graph, For the Set PANHS forget gate state, For the Group PANHS compressed association graph, For the Status characteristics of group LRHS, is the Hadamard product, For the Status characteristics of group LRHS, is the Tanh activation function, For the Group LRHS observation characteristics;

[0029] The calculation expression of the current state feature of the PANHS image is:

[0030]

[0031] in, For the Status characteristics of the PANHS group, For the Status characteristics of the PANHS group, For the Observational characteristics of the PANHS group.

[0032] Furthermore, in step S5, the calculation expression of the memory state feature is:

[0033]

[0034] in, For the Memory status characteristics of group PANHS, For the Memory state characteristics of group LRHS, is the Dirichlet autoencoder module, For the Status characteristics of group LRHS, For the Status characteristics of the PANHS group.

[0035] Furthermore, the calculation expression of the fusion feature is:

[0036]

[0037]

[0038]

[0039] in, For the Group reset matrix, is the Sigmoid activation function, For the Group PANHS compressed association graph, For the Group candidate features, is the Tanh activation function, For the Group fusion features, For the Memory status characteristics of group PANHS.

[0040] Furthermore, the recurrent memory network includes a forget gate layer, an observation feature layer, a reset gate layer and a candidate feature layer.

[0041] The present invention also provides a hyperspectral image fusion system based on superband aggregation and Dirichlet, which is used to execute the above-mentioned hyperspectral image fusion method based on superband aggregation and Dirichlet, comprising:

[0042] An acquisition module, configured to acquire LRHS images through a remote sensing optical imaging system;

[0043] a first fusion module, configured to fuse the LRHS image and the panchromatic image through a spatial spectral data distribution fusion module to obtain a PANHS image;

[0044] An aggregation module is used to aggregate the LRHS bands using an aggregation algorithm to obtain a superband, calculate a compression correlation map of the LRHS image based on the superband, and calculate a compression correlation map of the PANHS image based on the superband;

[0045] A feature extraction module is configured to extract the spatial spectral correlation of the LRHS image compression correlation graph based on a recurrent memory network to obtain the current state feature of the LRHS image, and to extract the spatial spectral correlation of the PANHS image compression correlation graph based on the recurrent memory network to obtain the current state feature of the PANHS image;

[0046] a second fusion module, which is configured to fuse the spectral and spatial information of the current state features of the LRHS image and the current state features of the PANHS image through the Dirichlet autoencoder module to obtain a memory state feature; extract candidate features through a recurrent memory network, and fuse the memory state feature and the candidate features to obtain a fused feature;

[0047] An integration module, wherein the integration module is used to integrate the fusion features into the PANHS image to obtain an integrated feature;

[0048] The third fusion module is used to obtain multiple integrated features, and fuse the multiple integrated features through the residual block to obtain a hyperspectral image with high spatial resolution.

[0049] The above one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects:

[0050] By combining the super-band aggregation method with the recurrent memory network, sequential spatial spectral fusion modeling between PAN and LRHS images is realized, the physical interpretability of hyperspectral decomposition is achieved, the spatial resolution of hyperspectral images is significantly improved, and the spectral consistency of hyperspectral images is maintained, providing technical support for the refined analysis and practical application of hyperspectral images.

[0051] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 The present invention provides a flow chart of a hyperspectral image fusion method based on superband aggregation and Dirichlet.

[0054] Figure 2 This is a schematic diagram of the network structure of a hyperspectral image fusion method based on superband aggregation and Dirichlet provided by the present invention.

[0055] Figure 3 It is a schematic diagram of the structure of the spatial spectrum data distribution fusion module provided by the present invention.

[0056] Figure 4 It is a structural diagram of the cyclic memory fusion module provided by the present invention.

[0057] Figure 5 This is a structural diagram of a hyperspectral image fusion system based on superband aggregation and Dirichlet provided by the present invention.

[0058] Figure 6 It is a schematic diagram of the comparative experimental results of the quantitative indicator results of the present invention on the Pavia center dataset.

[0059] Reference numerals:

[0060] 101. Acquisition module; 102. First fusion module; 103. Aggregation module; 104. Feature extraction module; 105. Second fusion module; 106. Integration module; 107. Third fusion module. DETAILED DESCRIPTION

[0061] To make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below. Obviously, the embodiments described are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0062] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0063] The following combination Figures 1 to 6 The hyperspectral image fusion method and system based on superband aggregation and Dirichlet of the present invention are described.

[0064] like Figure 1 As shown in FIG, a hyperspectral image fusion method based on superband aggregation and Dirichlet includes:

[0065] S1: Acquire LRHS images through remote sensing optical imaging system;

[0066] Acquire images from remote sensing optical imaging systems for applications such as mineral detection, ecosystem monitoring, and agricultural testing.

[0067] S2: The LRHS image and the panchromatic image are fused through the spatial spectral data distribution fusion module to obtain the PANHS image (panchromatic image with spectral information).

[0068] S21: obtaining an upsampled LRHS image by upsampling the LRHS image;

[0069] In some specific embodiments of the present invention, the LRHS image is subjected to 4-fold bicubic upsampling to obtain an upsampled LRHS image.

[0070] S22: The upsampled LRHS image and the panchromatic image are fused through the spatial spectral data distribution fusion module to obtain the PANHS image;

[0071] S221: Calculate the adaptive mean of the upsampled LRHS image and the adaptive variance of the upsampled LRHS image by parallel convolution.

[0072] S222: assigning the adaptive mean of the upsampled LRHS image and the adaptive variance of the upsampled LRHS image to the panchromatic image to obtain a PANHS image;

[0073] The calculation expression of PANHS image is:

[0074]

[0075] in, is a full-color image with spectral information, is a full-color image, is the variance calculation, is the LRHS image after 4 times bicubic upsampling, is a convolution operation with a convolution kernel of 3×3. Calculate the mean.

[0076] The structure of spatial spectral data distribution fusion module is as follows: Figure 3 shown.

[0077] S3: Aggregate the LRHS image bands using an aggregation algorithm to obtain superbands, calculate the LRHS image compression correlation map based on the superbands, and calculate the PANHS image compression correlation map based on the superbands;

[0078] S31: Through the aggregation algorithm, the high correlation between the LRHS image waves is used to aggregate the bands and obtain the super bands;

[0079] The aggregation algorithm is an improved K-mean classic aggregation algorithm, and only local iterative calculations are performed in each iteration;

[0080] S32: Calculate a correlation map of the LRHS image based on the superband; select the first k bands in the correlation map to form a sub-compression correlation map;

[0081] In some specific embodiments of the present invention, it is assumed that the hyperspectral image , where C is the number of channels, H is the image height, and W is the image width. is a real matrix, and the number of channels within the superband is , then the number of superbands , , assuming the channel For the One channel in a superband, belong In each iteration, each channel is calculated only with the surrounding adjacent superbands, which ensures high efficiency of calculation. In the iteration, the correlation graph of the LRHS image is calculated, and the calculation expression is:

[0082]

[0083] in, It is Rank The value of the column, for Channel No. Line images and superbands of Iteration No. The square of the norm distance between the columns of the image.

[0084] is the correlation graph of LRHS image The elements in .

[0085] S33: performing channel selection on the LRHS image according to the secondary compression correlation map, selecting a band in each superband to obtain the LRHS compression correlation map, performing channel selection on the PANHS image according to the secondary compression correlation map to obtain the PANHS compression correlation map.

[0086] S4: extracting the spatial spectral correlation of the LRHS image compression correlation graph according to the recurrent memory network to obtain the current state feature of the LRHS image; extracting the spatial spectral correlation of the PANHS image compression correlation graph according to the recurrent memory network to obtain the current state feature of the PANHS image;

[0087] The calculation expression of the current state feature of the LRHS image is:

[0088]

[0089]

[0090]

[0091] in, For the Group LRHS forget gate state, is the Sigmoid activation function, For the Group LRHS compressed association graph, For the Set PANHS forget gate state, For the Group PANHS compressed association graph, For the Status characteristics of group LRHS, is the Hadamard product, For the Status characteristics of group LRHS, is the Tanh activation function, For the Group LRHS observation characteristics;

[0092] The calculation expression of the current state feature of the PANHS image is:

[0093]

[0094] in, For the Status characteristics of the PANHS group, For the Status characteristics of the PANHS group, For the Observational characteristics of the PANHS group.

[0095] S5: The spectral and spatial information of the current state features of the LRHS image and the current state features of the PANHS image are fused through the Dirichlet autoencoder module to obtain the memory state features; candidate features are extracted through the recurrent memory network, and the memory state features and candidate features are fused to obtain the fused features;

[0096] The calculation expression of memory state characteristics is:

[0097]

[0098] in, For the Memory status characteristics of group PANHS, For the Memory state characteristics of group LRHS, is the Dirichlet autoencoder module, For the Status characteristics of group LRHS, For the Status characteristics of the PANHS group.

[0099] In the hyperspectral field, hyperspectral images can be unmixed and decomposed into the matrix product of the abundance matrix and the endmember matrix. The unmixing of hyperspectral images can be achieved by constructing an encoder-decoder structure based on the Dirichlet idea. The abundance matrix can be generated by the encoder, and then the decoder parameters are used as the endmember matrix. The abundance matrix output by the encoder passes through the decoder to obtain the unmixed hyperspectral image. The endmember matrix contains spectral information. The LRHS and HRHS images have the same endmember matrix, but the abundance matrix of LRHS contains less spatial information. Different encoders are used to decompose and calculate the abundance matrices of LRHS and PANHS. Since the endmember matrix of LRHS contains complete spectral information, the fusion of spatial and spectral information is achieved by sharing the decoders of the two. It contains the rich spatial state memory in PANHS and the complete spectral state memory in LRHS. The final information fusion is then completed by resetting the gate. The calculation expression is:

[0100]

[0101]

[0102]

[0103] in, For the Group reset matrix, is the Sigmoid activation function, For the Group PANHS compressed association graph, For the Group candidate features, is the Tanh activation function, For the Group fusion features, For the Memory status characteristics of group PANHS.

[0104] S6: According to the PANHS image compression correlation map, the fusion feature is integrated into the PANHS image with spectral information to obtain the integrated feature;

[0105] The position index in the PANHS image compression association graph is used to add the fusion feature to the PANHS image. The calculation expression of the integrated feature is:

[0106]

[0107] in, To integrate features, For feature integration, is the first group of fusion features, is the second group of fusion features, For the Group fusion features, A full-color image.

[0108] S7: Repeat steps S3 to S6 to obtain multiple integrated features, and fuse the multiple integrated features through the residual block to obtain a hyperspectral image with high spatial resolution.

[0109] The recurrent memory network consists of a forget gate layer, an observation feature layer, a reset gate layer, and a candidate feature layer.

[0110] The recurrent memory network and the Dirichlet autoencoder module form a recurrent memory fusion module. The recurrent memory fusion module is as follows: Figure 4 The overall network structure of the present invention is as shown. Figure 2 shown.

[0111] like Figure 5 As shown, a hyperspectral image fusion system based on superband aggregation and Dirichlet is used to perform a hyperspectral image fusion method based on superband aggregation and Dirichlet, including:

[0112] The acquisition module 101 is used to acquire LRHS images through a remote sensing optical imaging system;

[0113] The first fusion module 102 is used to fuse the LRHS image and the panchromatic image through the spatial spectral data distribution fusion module to obtain a PANHS image;

[0114] The aggregation module 103 is used to perform band aggregation on the LRHS image through an aggregation algorithm to obtain a super-band, calculate the LRHS image compression correlation map based on the super-band, and calculate the PANHS image compression correlation map based on the super-band;

[0115] The feature extraction module 104 is used to extract the spatial spectral correlation of the LRHS image compression correlation graph based on the recurrent memory network to obtain the current state feature of the LRHS image, and to extract the spatial spectral correlation of the PANHS image compression correlation graph based on the recurrent memory network to obtain the current state feature of the PANHS image;

[0116] The second fusion module 105 is used to fuse the spectral and spatial information of the current state features of the LRHS image and the current state features of the PANHS image through the Dirichlet autoencoder module to obtain a memory state feature; extract candidate features through a recurrent memory network, and fuse the memory state features and the candidate features to obtain a fused feature;

[0117] The integration module 106 is used to integrate the fusion features into the PANHS image to obtain the integrated features;

[0118] The third fusion module 107 is used to obtain multiple integrated features, and fuse the multiple integrated features through the residual block to obtain a hyperspectral image with high spatial resolution.

[0119] Through the collaborative work of the above modules, the super-band aggregation method is combined with the recurrent memory network to realize the sequential spatial spectral fusion modeling between PAN and LRHS images, achieve the physical interpretability of hyperspectral decomposition, significantly improve the spatial resolution of hyperspectral images, and maintain the spectral consistency of hyperspectral images, providing technical support for the refined analysis and practical application of hyperspectral images.

[0120] The present invention conducted quantitative index result comparison experiments and subjective comparison experiments with multiple internationally most advanced algorithms on the public data set Pavia center. Figure 6 To better visualize the differences between the fusion results, the mean absolute error (MAE) between the fusion results and the ground truth (GT) is calculated and displayed. A zoomed-in local area is shown in the lower right corner of the result, and the corresponding RGB image is shown in the lower left. These images show that the MAE graph for our method contains more dark blue areas, indicating the lowest fusion error. Figure 6 Among them, the international advanced algorithms are FPFNet (2023), HyperRefiner (2023), TreeSNet (2024), and DFCFN (2025). FPFNet (2023) is the 2023 version of the feature pyramid fusion network, HyperRefiner (2023) is the 2023 version of the refined hyperspectral pan-spectral network, TreeSNet (2024) is the 2024 version of the tree-structured neural network, and DFCFN (2025) is the 2025 version of the dual-stage feature correction fusion network. As shown in Table 1, it can be seen from the quantitative index results that the quantitative result indicators of the algorithm of the present invention are better than those of the international advanced algorithms in recent years.

[0121] Table 1 Comparative experiment results of quantitative indicators of Pavia center dataset

[0122]

[0123] Among them, CC is the correlation coefficient, SAM is the spectral angle mapping, RMSE is the root mean square error, ERGAS is the dimensionless global relative error, PSNR is the peak signal-to-noise ratio, "↑" means the higher the value, the better, and "↓" means the lower the value, the better.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A hyperspectral image fusion method based on superband aggregation and Dirichlet, characterized in that: include: S1: Acquire LRHS images through remote sensing optical imaging system; S2: The LRHS image and the panchromatic image are fused through the spatial spectral data distribution fusion module to obtain the PANHS image; S3: Aggregate the LRHS image bands using an aggregation algorithm to obtain superbands, calculate the LRHS image compression correlation map based on the superbands, and calculate the PANHS image compression correlation map based on the superbands; S31: Through the aggregation algorithm, the high correlation between the LRHS image bands is used to perform band aggregation to obtain super bands; S32: Calculate a correlation map of the LRHS image based on the superband; select the first k bands in the correlation map to form a sub-compression correlation map; S33: performing channel selection on the LRHS image according to the secondary compression correlation map, selecting a band in each superband to obtain the LRHS compression correlation map, and performing channel selection on the PANHS image according to the secondary compression correlation map to obtain the PANHS compression correlation map; S4: extracting the spatial spectral correlation of the LRHS image compression correlation graph based on the recurrent memory network to obtain the current state features of the LRHS image; The spatial-spectral correlation of the compressed correlation graph of the PANHS image is extracted based on the recurrent memory network to obtain the current state characteristics of the PANHS image. The calculation expression of the current state feature of the LRHS image is: in, For the Group LRHS forget gate state, is the Sigmoid activation function, For the Group LRHS compressed association graph, For the Set PANHS forget gate state, For the Group PANHS compressed association graph, For the Status characteristics of group LRHS, is the Hadamard product, For the Status characteristics of group LRHS, is the Tanh activation function, For the Group LRHS observation characteristics; The calculation expression of the current state feature of the PANHS image is: in, For the Status characteristics of the PANHS group, For the Status characteristics of the PANHS group, For the Group PANHS observational characteristics; The recurrent memory network includes a forget gate layer, an observation feature layer, a reset gate layer and a candidate feature layer; S5: The spectral and spatial information of the current state features of the LRHS image and the current state features of the PANHS image are fused through the Dirichlet autoencoder module to obtain the memory state features; candidate features are extracted through the recurrent memory network, and the memory state features and candidate features are fused to obtain the fused features; S6: Integrate the fused features into the PANHS image to obtain the integrated features; S7: Repeat steps S3 to S6 to obtain multiple integrated features, and fuse the multiple integrated features through the residual block to obtain a hyperspectral image with high spatial resolution.

2. The hyperspectral image fusion method based on superband aggregation and Dirichlet according to claim 1 is characterized in that: Step S2 includes: S21: obtaining an upsampled LRHS image by upsampling the LRHS image; S22: The upsampled LRHS image and the panchromatic image are fused through the spatial spectral data distribution fusion module to obtain a PANHS image.

3. The hyperspectral image fusion method based on superband aggregation and Dirichlet according to claim 2, characterized in that: Step S22 includes: S221: Calculate the adaptive mean of the upsampled LRHS image and the adaptive variance of the upsampled LRHS image by parallel convolution. S222: Assign the adaptive mean of the upsampled LRHS image and the adaptive variance of the upsampled LRHS image to the panchromatic image to obtain a PANHS image.

4. The hyperspectral image fusion method based on superband aggregation and Dirichlet according to claim 1, characterized in that: The aggregation algorithm is an improved K-mean classical aggregation algorithm, and only local iterative calculations are performed in each iteration.

5. The hyperspectral image fusion method based on superband aggregation and Dirichlet according to claim 1, characterized in that: In step S5, the calculation expression of the memory state feature is: in, For the Memory status characteristics of group PANHS, For the Memory state characteristics of group LRHS, is the Dirichlet autoencoder module, For the Status characteristics of group LRHS, For the Status characteristics of the PANHS group.

6. The hyperspectral image fusion method based on superband aggregation and Dirichlet according to claim 1, characterized in that: The calculation expression of the fusion feature is: in, For the Group reset matrix, is the Sigmoid activation function, For the Group PANHS compressed association graph, For the Group candidate features, is the Tanh activation function, For the Group fusion features, For the Memory status characteristics of group PANHS.

7. A hyperspectral image fusion system based on superband aggregation and Dirichlet, characterized by: The method for performing a hyperspectral image fusion method based on superband aggregation and Dirichlet according to any one of claims 1 to 6 comprises: An acquisition module, configured to acquire LRHS images through a remote sensing optical imaging system; a first fusion module, configured to fuse the LRHS image and the panchromatic image through a spatial spectral data distribution fusion module to obtain a PANHS image; An aggregation module is configured to perform band aggregation on the LRHS image using an aggregation algorithm to obtain a superband, calculate a compression correlation graph of the LRHS image based on the superband, and calculate a compression correlation graph of the PANHS image based on the superband; A feature extraction module is configured to extract the spatial spectral correlation of the LRHS image compression correlation graph based on a recurrent memory network to obtain the current state feature of the LRHS image, and to extract the spatial spectral correlation of the PANHS image compression correlation graph based on the recurrent memory network to obtain the current state feature of the PANHS image; a second fusion module, which is configured to fuse the spectral and spatial information of the current state features of the LRHS image and the current state features of the PANHS image through the Dirichlet autoencoder module to obtain a memory state feature; extract candidate features through a recurrent memory network, and fuse the memory state feature and the candidate features to obtain a fused feature; An integration module, wherein the integration module is used to integrate the fusion features into the PANHS image to obtain an integrated feature; The third fusion module is used to obtain multiple integrated features, and fuse the multiple integrated features through the residual block to obtain a hyperspectral image with high spatial resolution.

Images