scispace - formally typeset
Journal ArticleDOI: 10.1109/MGRS.2021.3064051

Interpretable Hyperspectral AI: When Non-Convex Modeling meets Hyperspectral Remote Sensing

02 Mar 2021-arXiv: Computer Vision and Pattern Recognition (IEEE - Institute of Electrical and Electronics Engineers)-
Abstract: Hyperspectral imaging, also known as image spectrometry, is a landmark technique in geoscience and remote sensing (RS). In the past decade, enormous efforts have been made to process and analyze these hyperspectral (HS) products mainly by means of seasoned experts. However, with the ever-growing volume of data, the bulk of costs in manpower and material resources poses new challenges on reducing the burden of manual labor and improving efficiency. For this reason, it is, therefore, urgent to develop more intelligent and automatic approaches for various HS RS applications. Machine learning (ML) tools with convex optimization have successfully undertaken the tasks of numerous artificial intelligence (AI)-related applications. However, their ability in handling complex practical problems remains limited, particularly for HS data, due to the effects of various spectral variabilities in the process of HS imaging and the complexity and redundancy of higher dimensional HS signals. Compared to the convex models, non-convex modeling, which is capable of characterizing more complex real scenes and providing the model interpretability technically and theoretically, has been proven to be a feasible solution to reduce the gap between challenging HS vision tasks and currently advanced intelligent data processing models.

... read more

Topics: Hyperspectral imaging (53%)
Citations
  More

8 results found


Open accessJournal ArticleDOI: 10.1109/JSTARS.2021.3103176
Abstract: Recently, hyperspectral image classification based on deep learning has achieved considerable attention. Many convolutional neural network classification methods have emerged and exhibited superior classification performance. However, most methods focus on extracting features by using fixed convolution kernels and layer-wise representation, resulting in feature extraction singleness. Additionally, the feature fusion process is rough and simple. Numerous methods get accustomed to fusing different levels of features by stacking modules hierarchically, which ignore the combination of shallow and deep spectral-spatial features. In order to overcome the preceding issues, a novel multiscale dual-branch feature fusion and attention network is proposed. Specifically, we design a multiscale feature extraction (MSFE) module to extract spatial-spectral features at a granular level and expand the range of receptive fields, thereby enhancing the MSFE ability. Subsequently, we develop a dual-branch feature fusion interactive module that integrates the residual connection's feature reuse property and the dense connection's feature exploration capability, obtaining more discriminative features in both spatial and spectral branches. Additionally, we introduce a novel shuffle attention mechanism that allows for adaptive weighting of spatial and spectral features, further improving classification performance. Experimental results on three benchmark datasets demonstrate that our model outperforms other state-of-the-art methods while incurring the lower computational cost.

... read more

1 Citations


Open accessJournal ArticleDOI: 10.1109/JSTARS.2021.3097178
Shuaiqi Liu1, Siyu Miao1, Jian Su2, Bing Li3  +2 moreInstitutions (4)
Abstract: To reconstruct images with high spatial resolution and high spectral resolution, one of the most common methods is to fuse a low-resolution hyperspectral image (HSI) with a high-resolution (HR) multispectral image (MSI) of the same scene. Deep learning has been widely applied in the field of HSI-MSI fusion, which is limited with hardware. In order to break the limits, we construct an unsupervised multiattention-guided network named UMAG-Net without training data to better accomplish HSI-MSI fusion. UMAG-Net first extracts deep multiscale features of MSI by using a multiattention encoding network. Then, a loss function containing a pair of HSI and MSI is used to iteratively update parameters of UMAG-Net and learn prior knowledge of the fused image. Finally, a multiscale feature-guided network is constructed to generate an HR-HSI. The experimental results show the visual and quantitative superiority of the proposed method compared to other methods.

... read more

Topics: Image fusion (59%), Multispectral image (55%), Hyperspectral imaging (54%) ... read more

1 Citations


Open accessJournal ArticleDOI: 10.1016/J.JAG.2021.102582
Yanheng Wang1, Yanheng Wang2, Lianru Gao1, Danfeng Hong1  +5 moreInstitutions (2)
Abstract: Traditional change detection (CD) algorithms cannot meet the requirements of today’s high resolution remote sensing images (HR). Recently, deep learning-based CD has become a popular research topic. However, there are not many annotated samples for training deep learning (DL) models. Patch-based algorithm has become an important research direction in CD in response to the lack of training datasets, but the optimal patch size is relatively small and difficult to determine, which limits the use of spatial information and the extension of deep network. In this paper, we develop a feature-regularized mask DeepLab (FRM-DeepLab) for HRCD. First, a mask-based framework (MaskNet) that uses a few annotated samples to update model parameters is introduced. Based on MaskNet, we design a Mask-DeepLab to make full use of HR. Last, the deep features of unlabeled areas are extracted by an autoencoder as auxiliary information, and those features are concatenated in the middle-level features extracted by Mask-DeepLab to alleviate the influences of overfitting caused by small-scale samples. The algorithm is verified on three HRCD datasets. The visualization and quantitative analysis of the experiment results figure that this algorithm can implement significant performance improvement.

... read more

Topics: Autoencoder (56%), Overfitting (54%), Image segmentation (53%) ... read more

Journal ArticleDOI: 10.1016/J.SIGPRO.2021.108214
01 Nov 2021-Signal Processing
Abstract: Hyperspectral unmixing is the process of separating the signatures of different pure materials in a mixed pixel. For different reasons including the intrinsic variability of materials and variations in data collecting conditions, different forms of spectral signatures can be applied to a particular material. This is referred to as spectral variability. Scaling factors and bundle dictionary are two concepts that address illumination variations and intrinsic variabilities of materials, respectively. In this paper, we propose the linear mixing model with scaled bundle dictionary (LMM-SBD) method which combines both the scaling factors and bundle dictionary to benefit from the advantages of both approaches. Moreover, we use different spatial neighbors to account for the spatial coherence of the neighboring pixels and force their corresponding abundances to have a similar sparsity pattern by adding two mixed norms to the optimization problem. The produced problem is solved using an alternating direction method of multipliers approach. The proposed method is tested on a simulated data and three real hyperspectral data and the results verify that the proposed method is able to overcome the spectral variability of materials, exploit the spatial coherence of the neighboring pixels and yield a successful unmixing procedure compared with several state-of-the-art methods.

... read more

Topics: Hyperspectral imaging (56%), Spectral signature (51%), Bundle (50%)

Open accessJournal ArticleDOI: 10.1016/J.JAG.2021.102461
Xiaobin Zhao1, Zengfu Hou1, Xin Wu1, Wei Li1  +2 moreInstitutions (2)
Abstract: Traditional hyperspectral target detection methods use spectral domain information for target recognition. Although it can effectively retain intrinsic characteristics of substances, targets in homogeneous regions still cannot be effectively recognized. By projecting the spectral domain features on the transform domain to increase the separability of background and target, fractional domain-based revised constrained energy minimization detector is proposed. Firstly, the fractional Fourier transform is adopted to project the original spectral information into the fractional domain for improving the separability of background and target. Then, a newly revised constrained energy minimization detector is performed, where sliding double window strategy is used to make the best of the local spatial statistical characteristics of testing pixel. In order to make the best of inner window information, the mean value of Pearson correlation coefficient is measured between prior target pixel and testing pixel associated with its four neighborhood pixels. Extensive experiments for four real hyperspectral scenes indicate that the performance of the proposed algorithm is excellent when compared with other related detectors.

... read more

Topics: Fractional Fourier transform (58%), Hyperspectral imaging (55%), Pixel (53%) ... read more

References
  More

160 results found


Journal ArticleDOI: 10.1109/TIP.2003.819861
Abstract: Objective methods for assessing perceptual image quality traditionally attempted to quantify the visibility of errors (differences) between a distorted image and a reference image using a variety of known properties of the human visual system. Under the assumption that human visual perception is highly adapted for extracting structural information from a scene, we introduce an alternative complementary framework for quality assessment based on the degradation of structural information. As a specific example of this concept, we develop a structural similarity index and demonstrate its promise through a set of intuitive examples, as well as comparison to both subjective ratings and state-of-the-art objective methods on a database of images compressed with JPEG and JPEG2000. A MATLAB implementation of the proposed algorithm is available online at http://www.cns.nyu.edu//spl sim/lcv/ssim/.

... read more

Topics: Image quality (61%), Subjective video quality (56%), Human visual system model (56%) ... read more

30,333 Citations


Open accessBook
Stephen Boyd1, Neal Parikh1, Eric Chu1, Borja Peleato1  +1 moreInstitutions (2)
23 May 2011-
Abstract: Many problems of recent interest in statistics and machine learning can be posed in the framework of convex optimization. Due to the explosion in size and complexity of modern datasets, it is increasingly important to be able to solve problems with a very large number of features or training examples. As a result, both the decentralized collection or storage of these datasets as well as accompanying distributed solution methods are either necessary or at least highly desirable. In this review, we argue that the alternating direction method of multipliers is well suited to distributed convex optimization, and in particular to large-scale problems arising in statistics, machine learning, and related areas. The method was developed in the 1970s, with roots in the 1950s, and is equivalent or closely related to many other algorithms, such as dual decomposition, the method of multipliers, Douglas–Rachford splitting, Spingarn's method of partial inverses, Dykstra's alternating projections, Bregman iterative algorithms for l1 problems, proximal methods, and others. After briefly surveying the theory and history of the algorithm, we discuss applications to a wide variety of statistical and machine learning problems of recent interest, including the lasso, sparse logistic regression, basis pursuit, covariance selection, support vector machines, and many others. We also discuss general distributed optimization, extensions to the nonconvex setting, and efficient implementation, including some details on distributed MPI and Hadoop MapReduce implementations.

... read more

Topics: Online machine learning (59%), Statistical learning theory (57%), Convex optimization (56%) ... read more

14,958 Citations


Journal ArticleDOI: 10.1126/SCIENCE.290.5500.2323
Sam T. Roweis1, Lawrence K. Saul2Institutions (2)
22 Dec 2000-Science
Abstract: Many areas of science depend on exploratory data analysis and visualization. The need to analyze large amounts of multivariate data raises the fundamental problem of dimensionality reduction: how to discover compact representations of high-dimensional data. Here, we introduce locally linear embedding (LLE), an unsupervised learning algorithm that computes low-dimensional, neighborhood-preserving embeddings of high-dimensional inputs. Unlike clustering methods for local dimensionality reduction, LLE maps its inputs into a single global coordinate system of lower dimensionality, and its optimizations do not involve local minima. By exploiting the local symmetries of linear reconstructions, LLE is able to learn the global structure of nonlinear manifolds, such as those generated by images of faces or documents of text.

... read more

13,822 Citations


Journal ArticleDOI: 10.1109/TSP.2006.881199
Abstract: In recent years there has been a growing interest in the study of sparse representation of signals. Using an overcomplete dictionary that contains prototype signal-atoms, signals are described by sparse linear combinations of these atoms. Applications that use sparse representation are many and include compression, regularization in inverse problems, feature extraction, and more. Recent activity in this field has concentrated mainly on the study of pursuit algorithms that decompose signals with respect to a given dictionary. Designing dictionaries to better fit the above model can be done by either selecting one from a prespecified set of linear transforms or adapting the dictionary to a set of training signals. Both of these techniques have been considered, but this topic is largely still open. In this paper we propose a novel algorithm for adapting dictionaries in order to achieve sparse signal representations. Given a set of training signals, we seek the dictionary that leads to the best representation for each member in this set, under strict sparsity constraints. We present a new method-the K-SVD algorithm-generalizing the K-means clustering process. K-SVD is an iterative method that alternates between sparse coding of the examples based on the current dictionary and a process of updating the dictionary atoms to better fit the data. The update of the dictionary columns is combined with an update of the sparse representations, thereby accelerating convergence. The K-SVD algorithm is flexible and can work with any pursuit method (e.g., basis pursuit, FOCUSS, or matching pursuit). We analyze this algorithm and demonstrate its results both on synthetic tests and in applications on real image data

... read more

Topics: K-SVD (76%), Sparse approximation (66%), Matching pursuit (61%) ... read more

8,149 Citations


Open accessJournal ArticleDOI: 10.1162/089976603321780317
Mikhail Belkin1, Partha Niyogi1Institutions (1)
01 Jun 2003-Neural Computation
Abstract: One of the central problems in machine learning and pattern recognition is to develop appropriate representations for complex data. We consider the problem of constructing a representation for data lying on a low-dimensional manifold embedded in a high-dimensional space. Drawing on the correspondence between the graph Laplacian, the Laplace Beltrami operator on the manifold, and the connections to the heat equation, we propose a geometrically motivated algorithm for representing the high-dimensional data. The algorithm provides a computationally efficient approach to nonlinear dimensionality reduction that has locality-preserving properties and a natural connection to clustering. Some potential applications and illustrative examples are discussed.

... read more

Topics: Manifold alignment (63%), Diffusion map (62%), Spectral clustering (62%) ... read more

6,475 Citations