Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification.
Saved in:
| Title: | Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification. |
|---|---|
| Authors: | Li, Guandong1 (AUTHOR) leeguandon@gmail.com, Ye, Mengxia2 (AUTHOR) |
| Source: | Arabian Journal for Science & Engineering (Springer Science & Business Media B.V. ). May2026, Vol. 51 Issue 9, p12183-12196. 14p. |
| Subject Terms: | *Splines, *Signal convolution, *Optimization algorithms, *Artificial neural networks, *Image recognition (Computer vision) |
| Abstract: | Deep neural networks face several challenges in hyperspectral image classification, including high-dimensional data, sparse distribution of ground objects, and spectral redundancy, which often lead to classification overfitting and limited generalization capability. To more efficiently adapt to ground object distributions while extracting image features without introducing excessive parameters and skipping redundant information, this paper proposes KANet based on an improved 3D-DenseNet model, consisting of 3D KAN Conv and an adaptive grid update mechanism. By introducing learnable univariate B-spline functions on network edges, specifically by flattening three-dimensional neighborhoods into vectors and applying B-spline-parameterized nonlinear activation functions to replace the fixed linear weights of traditional 3D convolutional kernels, we precisely capture complex spectral-spatial nonlinear relationships in hyperspectral data. Simultaneously, through a dynamic grid adjustment mechanism, we adaptively update the grid point positions of B-splines based on the statistical characteristics of input data, optimizing the resolution of spline functions to match the non-uniform distribution of spectral features, significantly improving the model's accuracy in high-dimensional data modeling and parameter efficiency, effectively alleviating the curse of dimensionality. This characteristic demonstrates superior neural scaling laws compared to traditional convolutional neural networks and reduces overfitting risks in small-sample and high-noise scenarios. KANet enhances model representation capability through a 3D dynamic expert convolution system without increasing network depth or width. The proposed method demonstrates superior performance on IN, UP, and KSC datasets, outperforming mainstream hyperspectral image classification approaches. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: enr DbLabel: Energy & Power Source An: 193807426 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Li%2C+Guandong%22">Li, Guandong</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> leeguandon@gmail.com</i><br /><searchLink fieldCode="AR" term="%22Ye%2C+Mengxia%22">Ye, Mengxia</searchLink><relatesTo>2</relatesTo> (AUTHOR) – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Arabian+Journal+for+Science+%26+Engineering+%28Springer+Science+%26+Business+Media+B%2EV%2E+%29%22">Arabian Journal for Science & Engineering (Springer Science & Business Media B.V. )</searchLink>. May2026, Vol. 51 Issue 9, p12183-12196. 14p. – Name: Subject Label: Subject Terms Group: Su Data: *<searchLink fieldCode="DE" term="%22Splines%22">Splines</searchLink><br />*<searchLink fieldCode="DE" term="%22Signal+convolution%22">Signal convolution</searchLink><br />*<searchLink fieldCode="DE" term="%22Optimization+algorithms%22">Optimization algorithms</searchLink><br />*<searchLink fieldCode="DE" term="%22Artificial+neural+networks%22">Artificial neural networks</searchLink><br />*<searchLink fieldCode="DE" term="%22Image+recognition+%28Computer+vision%29%22">Image recognition (Computer vision)</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Deep neural networks face several challenges in hyperspectral image classification, including high-dimensional data, sparse distribution of ground objects, and spectral redundancy, which often lead to classification overfitting and limited generalization capability. To more efficiently adapt to ground object distributions while extracting image features without introducing excessive parameters and skipping redundant information, this paper proposes KANet based on an improved 3D-DenseNet model, consisting of 3D KAN Conv and an adaptive grid update mechanism. By introducing learnable univariate B-spline functions on network edges, specifically by flattening three-dimensional neighborhoods into vectors and applying B-spline-parameterized nonlinear activation functions to replace the fixed linear weights of traditional 3D convolutional kernels, we precisely capture complex spectral-spatial nonlinear relationships in hyperspectral data. Simultaneously, through a dynamic grid adjustment mechanism, we adaptively update the grid point positions of B-splines based on the statistical characteristics of input data, optimizing the resolution of spline functions to match the non-uniform distribution of spectral features, significantly improving the model's accuracy in high-dimensional data modeling and parameter efficiency, effectively alleviating the curse of dimensionality. This characteristic demonstrates superior neural scaling laws compared to traditional convolutional neural networks and reduces overfitting risks in small-sample and high-noise scenarios. KANet enhances model representation capability through a 3D dynamic expert convolution system without increasing network depth or width. The proposed method demonstrates superior performance on IN, UP, and KSC datasets, outperforming mainstream hyperspectral image classification approaches. [ABSTRACT FROM AUTHOR] |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=enr&AN=193807426 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s13369-025-11001-3 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 14 StartPage: 12183 Subjects: – SubjectFull: Splines Type: general – SubjectFull: Signal convolution Type: general – SubjectFull: Optimization algorithms Type: general – SubjectFull: Artificial neural networks Type: general – SubjectFull: Image recognition (Computer vision) Type: general Titles: – TitleFull: Dynamic 3D KAN Convolution with Adaptive Grid Optimization for Hyperspectral Image Classification. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Li, Guandong – PersonEntity: Name: NameFull: Ye, Mengxia IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 05 Text: May2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 2193567X Numbering: – Type: volume Value: 51 – Type: issue Value: 9 Titles: – TitleFull: Arabian Journal for Science & Engineering (Springer Science & Business Media B.V. ) Type: main |
| ResultId | 1 |
