An attention mechanism assigns different weights to different features to help a model select the features most valuable for accurate classification. However, the traditional attention mechanism algorithm often allocates weights in a one-way fashion, which can result in a loss of feature information. To obtain better hyperspectral data classification results, a novel cross-attention mechanism and graph convolution integration algorithm are proposed in this letter. First, principal component analysis is used to reduce the dimensionality of hyperspectral images to obtain low-dimensional features that are more expressive. Second, the model uses a cross (horizontal and vertical directions) attention algorithm to allocate weights jointly based on its two strategies; then, it adopts a graph convolution algorithm to generate the directional relationships between the features. Finally, the generated deep features and the relationship between the deep features are used to complete the prediction of hyperspectral data. Experiments on three well-known hyperspectral data sets--Indian Pines, the University of Pavia, and Salinas--show that the proposed algorithm achieves better performances than do other well-known algorithms using different methods of training set division.

Remote Sensing Image Classification Based on a Cross-Attention Mechanism and Graph Convolution

In many statistical hyperspectral unmixing approaches, the unmixing task is essentially an optimization problem given a defined linear or nonlinear spectral mixture model. However, most of the model inference algorithms require a time-consuming iterative procedure. On the other hand, neural networks have been recently used to estimate abundances given some training samples, or directly estimate endmembers and abundances simultaneously in an unsupervised setting. However, their disadvantages are clear: lack of interpretability and reliance on the large training set. Model-inspired neural networks are constructed by the problem model and its corresponding inference algorithm. It incorporates the prior knowledge of physical model and algorithm into network architecture, combining the advantages of model-based and learning-based methods. This article deeply unfolds the linear mixture model and the corresponding iterative shrinkage-thresholding algorithm (ISTA) to build two unmixing network architectures. The first assumes that the set of endmembers are known, and the deep unfolded ISTA model is only for abundance estimation; and the second is used for blind unmixing to estimate both endmembers and abundances at the same time. The networks can be trained by supervised and unsupervised schemes, respectively, with a small-size training set, and then, unmixing becomes a feedforward process, which is very fast since no iteration is required. The experimental results show their competitive performance compared with the state-of-the-art unmixing approaches.

Spectral Mixture Model Inspired Network Architectures for Hyperspectral Unmixing

Although hyperspectral remote sensing images have rich spectral features, for small samples of remote sensing images, feature selection, feature mining, and feature integration are very important. A single model is difficult to apply to multiple tasks such as feature selection, feature mining, and feature integration during training, resulting in poor classification results for small sample classification of hyperspectral images. To improve the classification of small samples, a sequential joint deep learning algorithm is proposed in this paper. (In this algorithm, the deep features of multiscale convolution under an attention mechanism are integrated by using Bidirectional Long Short-Term Memory(Bi-LSTM) and AML.) First, we used principal component analysis (PCA) to reduce the dimensionality of the hyperspectral data and retain their key features. Second, the model uses an integrated attention mechanism to distribute the probability weight of the key input feature. Third, the model uses multiscale convolution to mine features after the distribution weight to obtain deep features. Fourth, the model uses bidirectional long short-term memory (Bi-LSTM) to integrate the convolution results at different scales. Finally, the softmax classifier is used to complete the classification of multiclass hyperspectral remote sensing images. Experiments were carried out on three public hyperspectral data sets, and the results proved that our proposed AML algorithm is effective, thus demonstrating powerful performance in the prediction of hyperspectral images (HSIs) of small samples.

Small Sample Classification of Hyperspectral Remote Sensing Images Based on Sequential Joint Deeping Learning Model

With the increasing maturity of optics and photonics, hyperspectral technology has also greatly advanced. Hyperspectral images composed of hundreds of adjacent bands and containing useful information can be easily obtained. However, unlike ordinary remote sensing images, each sample in hyperspectral remote sensing images has high-dimensional features and contains rich spatial and spectral information, which greatly increases the difficulty of feature selection and mining, increases the computational complexity, and limits the recognition accuracy of the model. Therefore, in this letter, a novel hybrid dilated-convolution-guided feature filtering and enhancement strategy (HDCFE-Net) model is proposed to classify hyperspectral images. Dilated convolution can reduce the spatial feature loss without reducing the receptive field and can obtain distant features. It can also be combined with the traditional convolution without losing its original information. We propose a feature filtering and enhancement strategy that eliminates redundant features and reduces computational complexity. The core concept is to set a threshold feature value, like the rounding method, to filter and enhance features. Experiments on three well-known hyperspectral datasets—Indian Pines (IPs), Pavia University (PU), and Salinas—show that in less than 1% (IPs: 5%) of the training samples, the overall accuracy (OA) of our method is 77%, 89%, and 91%, respectively, which is superior to several well-known methods. The experiments demonstrated the effectiveness and superiority of HDCFE-Net. 

Hybrid Dilated Convolution Guided Feature Filtering and Enhancement Strategy for Hyperspectral Image Classification

An attention mechanism assigns different weights to different features to help a model select the features most valuable for accurate classification. However, the traditional attention mechanism algorithm often allocates weights in a one-way fashion, which can result in a loss of feature information. To obtain better hyperspectral data classification results, a novel cross-attention mechanism and graph convolution integration algorithm are proposed in this letter. First, principal component analysis is used to reduce the dimensionality of hyperspectral images to obtain low-dimensional features that are more expressive. Second, the model uses a cross (horizontal and vertical directions) attention algorithm to allocate weights jointly based on its two strategies; then, it adopts a graph convolution algorithm to generate the directional relationships between the features. Finally, the generated deep features and the relationship between the deep features are used to complete the prediction of hyperspectral data. Experiments on three well-known hyperspectral data sets—Indian Pines, the University of Pavia, and Salinas—show that the proposed algorithm achieves better performances than do other well-known algorithms using different methods of training set division. 

Deep learning has been widely used for hyperspectral pixel classification due to its ability to generate deep feature representation. However, how to construct an efficient and powerful network suitable for hyperspectral data is still under exploration. In this paper, a novel neural network model is designed for taking full advantage of the spectral-spatial structure of hyperspectral data. First, we extract pixel-based intrinsic features from rich yet redundant spectral bands by a subnetwork with the supervised pre-training scheme. Second, in order to utilize the local spatial correlation among pixels, we share the previous subnetwork as a spectral feature extractor for each pixel in a patch of the image, after which the spectral features of all pixels in a patch are combined and fed into the subsequent classification subnetwork. Finally, the whole network is further fine-tuned to improve its classification performance. Especially, the spectral-spatial factorization scheme is applied in our model architecture, making the network size and the number of parameters great less than the existing spectral-spatial deep networks for hyperspectral image classification. Compared with other state-of-the-art deep learning methods, experiments on the hyperspectral data sets show that our method achieves 0.18%-7.6%, 0.1%-3.58%, and 0.21%-3.09% improvement on overall accuracy (OA), average accuracy (AA), and kappa, respectively, while having smaller network size and fewer parameters.

/pdf/deep-neural-network-based-hyperspectral-pixel-classification-2gg4kr3mmz.pdf

Deep Neural Network Based Hyperspectral Pixel Classification With Factorized Spectral-Spatial Feature Representation

Deep learning has been widely used for hyperspectral pixel classification due to its ability of generating deep feature representation. However, how to construct an efficient and powerful network suitable for hyperspectral data is still under exploration. In this paper, a novel neural network model is designed for taking full advantage of the spectral-spatial structure of hyperspectral data. Firstly, we extract pixel-based intrinsic features from rich yet redundant spectral bands by a subnetwork with supervised pre-training scheme. Secondly, in order to utilize the local spatial correlation among pixels, we share the previous subnetwork as a spectral feature extractor for each pixel in a patch of image, after which the spectral features of all pixels in a patch are combined and feeded into the subsequent classification subnetwork. Finally, the whole network is further fine-tuned to improve its classification performance. Specially, the spectral-spatial factorization scheme is applied in our model architecture, making the network size and the number of parameters great less than the existing spectral-spatial deep networks for hyperspectral image classification. Experiments on the hyperspectral data sets show that, compared with some state-of-art deep learning methods, our method achieves better classification results while having smaller network size and less parameters.

pdf/deep-neural-network-based-hyperspectral-pixel-classification-2b3vu5rm22.pdf

Tremendous efficient optimization methods have been proposed for strongly convex objectives optimization in modern machine learning. For non-strongly convex objectives, a popular approach is to apply a reduction from non-strongly convex to a strongly convex case via regularization techniques. Reduction on objectives with adaptive decrease on regularization tightens the optimal convergence of algorithms to be independent on logarithm factor. However, the initialization of parameter of regularization has a great impact on the performance of the reduction. In this paper, we propose an aggressive reduction to reduce the complexity of optimization for non-strongly convex objectives, and our reduction eliminates the impact of the initialization of parameter on the convergent performances of algorithms. Aggressive reduction not only adaptively decreases the regularization parameter, but also modifies regularization term as the distance between current point and the approximate minimizer. Our aggressive reduction can also shave off the non-optimal logarithm term theoretically, and make the convergent performance of algorithm more compact practically. Experimental results on logistic regression and image deblurring confirm this success in practice.

An aggressive reduction on the complexity of optimization for non-strongly convex objectives

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Siyu Chen

Author Tools

Chat about Author

Papers

Deep Neural Network Based Hyperspectral Pixel Classification With Factorized Spectral-Spatial Feature Representation

Deep Neural Network Based Hyperspectral Pixel Classification With Factorized Spectral-Spatial Feature Representation

An aggressive reduction on the complexity of optimization for non-strongly convex objectives