scholarly journals Deep Fully Convolutional Embedding Networks for Hyperspectral Images Dimensionality Reduction

2021 ◽  
Vol 13 (4) ◽  
pp. 706
Author(s):  
Na Li ◽  
Deyun Zhou ◽  
Jiao Shi ◽  
Mingyang Zhang ◽  
Tao Wu ◽  
...  
Keyword(s):  
Dimensionality Reduction ◽  
Learning Task ◽  
Classification Performance ◽  
Hyperspectral Images ◽  
Data Reconstruction ◽  
Specific Task ◽  
Spectral Bands ◽  
Class Labels ◽  
Limited Training Samples ◽  
Low Dimensional

Due to the superior spatial–spectral extraction capability of the convolutional neural network (CNN), CNN shows great potential in dimensionality reduction (DR) of hyperspectral images (HSIs). However, most CNN-based methods are supervised while the class labels of HSIs are limited and difficult to obtain. While a few unsupervised CNN-based methods have been proposed recently, they always focus on data reconstruction and are lacking in the exploration of discriminability which is usually the primary goal of DR. To address these issues, we propose a deep fully convolutional embedding network (DFCEN), which not only considers data reconstruction but also introduces the specific learning task of enhancing feature discriminability. DFCEN has an end-to-end symmetric network structure that is the key for unsupervised learning. Moreover, a novel objective function containing two terms—the reconstruction term and the embedding term of a specific task—is established to supervise the learning of DFCEN towards improving the completeness and discriminability of low-dimensional data. In particular, the specific task is designed to explore and preserve relationships among samples in HSIs. Besides, due to the limited training samples, inherent complexity and the presence of noise in HSIs, a preprocessing where a few noise spectral bands are removed is adopted to improve the effectiveness of unsupervised DFCEN. Experimental results on three well-known hyperspectral datasets and two classifiers illustrate that the low dimensional features of DFCEN are highly separable and DFCEN has promising classification performance compared with other DR methods.

scholarly journals Classification of Hyperspectral Images with CNN in Agricultural Lands

2021 ◽  
Vol 3 (1) ◽  
pp. 6
Author(s):  
Eren Can Seyrek ◽  
Murat Uysal
Keyword(s):  
Classification Performance ◽  
Hyperspectral Images ◽  
Superior Performance ◽  
Support Vector ◽  
World Population ◽  
Agricultural Lands ◽  
Leading Role ◽  
Spectral Bands ◽  
Land Cover Maps

Hyperspectral images (HSI) offer detailed spectral reflectance information about sensed objects through provision of information on hundreds of narrow spectral bands. HSI have a leading role in a broad range of applications, such as in forestry, agriculture, geology, and environmental sciences. The monitoring and management of agricultural lands is of great importance for meeting the nutritional and other needs of a rapidly and continuously increasing world population. In relation to this, classification of HSI is an effective way for creating land use and land cover maps quickly and accurately. In recent years, classification of HSI using convolutional neural networks (CNN), which is a sub-field of deep learning, has become a very popular research topic and several CNN architectures have been developed by researchers. The aim of this study was to investigate the classification performance of CNN model on agricultural HSI scenes. For this purpose, a 3D-2D CNN framework and a well-known support vector machine (SVM) model were compared using the Indian Pines and Salinas Scene datasets that contain crop and mixed vegetation classes. As a result of this study, it was confirmed that use of 3D-2D CNN offers superior performance for classifying agricultural HSI datasets.

scholarly journals TASK-DEPENDENT BAND-SELECTION OF HYPERSPECTRAL IMAGES BY PROJECTION-BASED RANDOM FORESTS

2016 ◽  
Vol III-7 ◽  
pp. 263-270
Author(s):  
R. Hänsch ◽  
O. Hellwich
Keyword(s):  
Land Cover ◽  
Random Forests ◽  
Classification Performance ◽  
Hyperspectral Images ◽  
Band Selection ◽  
Spectral Correlation ◽  
Spectral Bands ◽  
The Given ◽  
Selection Of

The automatic classification of land cover types from hyperspectral images is a challenging problem due to (among others) the large amount of spectral bands and their high spatial and spectral correlation. The extraction of meaningful features, that enables a subsequent classifier to distinguish between different land cover classes, is often limited to a subset of all available data dimensions which is found by band selection techniques or other methods of dimensionality reduction. This work applies Projection-Based Random Forests to hyperspectral images, which not only overcome the need of an explicit feature extraction, but also provide mechanisms to automatically select spectral bands that contain original (i.e. non-redundant) as well as highly meaningful information for the given classification task. The proposed method is applied to four challenging hyperspectral datasets and it is shown that the effective number of spectral bands can be considerably limited without loosing too much of classification performance, e.g. a loss of 1 % accuracy if roughly 13 % of all available bands are used.

scholarly journals Dimensionality Reduction by Supervised Neighbor Embedding Using Laplacian Search

2014 ◽  
Vol 2014 ◽  
pp. 1-14 ◽  
Author(s):  
Jianwei Zheng ◽  
Hangke Zhang ◽  
Carlo Cattani ◽  
Wanliang Wang
Keyword(s):  
Dimensionality Reduction ◽  
System Analysis ◽  
Classification Performance ◽  
Living System ◽  
Linear Projection ◽  
Reduction Methods ◽  
Gradient Based ◽  
Class Labels ◽  
Global And Local ◽  
Training Efficiency

Dimensionality reduction is an important issue for numerous applications including biomedical images analysis and living system analysis. Neighbor embedding, those representing the global and local structure as well as dealing with multiple manifolds, such as the elastic embedding techniques, can go beyond traditional dimensionality reduction methods and find better optima. Nevertheless, existing neighbor embedding algorithms can not be directly applied in classification as suffering from several problems: (1) high computational complexity, (2) nonparametric mappings, and (3) lack of class labels information. We propose a supervised neighbor embedding called discriminative elastic embedding (DEE) which integrates linear projection matrix and class labels into the final objective function. In addition, we present the Laplacian search direction for fast convergence. DEE is evaluated in three aspects: embedding visualization, training efficiency, and classification performance. Experimental results on several benchmark databases present that the proposed DEE exhibits a supervised dimensionality reduction approach which not only has strong pattern revealing capability, but also brings computational advantages over standard gradient based methods.

scholarly journals TASK-DEPENDENT BAND-SELECTION OF HYPERSPECTRAL IMAGES BY PROJECTION-BASED RANDOM FORESTS

2016 ◽  
Vol III-7 ◽  
pp. 263-270
Author(s):  
R. Hänsch ◽  
O. Hellwich
Keyword(s):  
Land Cover ◽  
Random Forests ◽  
Classification Performance ◽  
Hyperspectral Images ◽  
Band Selection ◽  
Spectral Correlation ◽  
Spectral Bands ◽  
The Given ◽  
Selection Of

The automatic classification of land cover types from hyperspectral images is a challenging problem due to (among others) the large amount of spectral bands and their high spatial and spectral correlation. The extraction of meaningful features, that enables a subsequent classifier to distinguish between different land cover classes, is often limited to a subset of all available data dimensions which is found by band selection techniques or other methods of dimensionality reduction. This work applies Projection-Based Random Forests to hyperspectral images, which not only overcome the need of an explicit feature extraction, but also provide mechanisms to automatically select spectral bands that contain original (i.e. non-redundant) as well as highly meaningful information for the given classification task. The proposed method is applied to four challenging hyperspectral datasets and it is shown that the effective number of spectral bands can be considerably limited without loosing too much of classification performance, e.g. a loss of 1 % accuracy if roughly 13 % of all available bands are used.

scholarly journals Sub-Graph Regularization on Kernel Regression for Robust Semi-Supervised Dimensionality Reduction

Entropy ◽  
2019 ◽  
Vol 21 (11) ◽  
pp. 1125
Author(s):  
Jiao Liu ◽  
Mingbo Zhao ◽  
Weijian Kong
Keyword(s):  
Dimensionality Reduction ◽  
Supervised Learning ◽  
Kernel Regression ◽  
Classification Performance ◽  
Unlabeled Data ◽  
Graph Regularization ◽  
Linear Discriminant ◽  
Benchmark Datasets ◽  
Reduction Methods ◽  
Class Labels

Dimensionality reduction has always been a major problem for handling huge dimensionality datasets. Due to the utilization of labeled data, supervised dimensionality reduction methods such as Linear Discriminant Analysis tend achieve better classification performance compared with unsupervised methods. However, supervised methods need sufficient labeled data in order to achieve satisfying results. Therefore, semi-supervised learning (SSL) methods can be a practical selection rather than utilizing labeled data. In this paper, we develop a novel SSL method by extending anchor graph regularization (AGR) for dimensionality reduction. In detail, the AGR is an accelerating semi-supervised learning method to propagate the class labels to unlabeled data. However, it cannot handle new incoming samples. We thereby improve AGR by adding kernel regression on the basic objective function of AGR. Therefore, the proposed method can not only estimate the class labels of unlabeled data but also achieve dimensionality reduction. Extensive simulations on several benchmark datasets are conducted, and the simulation results verify the effectiveness for the proposed work.

scholarly journals Supervised dimensionality reduction for big data

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Joshua T. Vogelstein ◽  
Eric W. Bridgeford ◽  
Minh Tang ◽  
Da Zheng ◽  
Christopher Douville ◽  
...  
Keyword(s):  
Dimensionality Reduction ◽  
Data Science ◽  
Real Data ◽  
Low Rank ◽  
Conditional Moment ◽  
Desktop Computer ◽  
Reduction Techniques ◽  
Reduction Methods ◽  
The Individual ◽  
Low Dimensional

AbstractTo solve key biomedical problems, experimentalists now routinely measure millions or billions of features (dimensions) per sample, with the hope that data science techniques will be able to build accurate data-driven inferences. Because sample sizes are typically orders of magnitude smaller than the dimensionality of these data, valid inferences require finding a low-dimensional representation that preserves the discriminating information (e.g., whether the individual suffers from a particular disease). There is a lack of interpretable supervised dimensionality reduction methods that scale to millions of dimensions with strong statistical theoretical guarantees. We introduce an approach to extending principal components analysis by incorporating class-conditional moment estimates into the low-dimensional projection. The simplest version, Linear Optimal Low-rank projection, incorporates the class-conditional means. We prove, and substantiate with both synthetic and real data benchmarks, that Linear Optimal Low-Rank Projection and its generalizations lead to improved data representations for subsequent classification, while maintaining computational efficiency and scalability. Using multiple brain imaging datasets consisting of more than 150 million features, and several genomics datasets with more than 500,000 features, Linear Optimal Low-Rank Projection outperforms other scalable linear dimensionality reduction techniques in terms of accuracy, while only requiring a few minutes on a standard desktop computer.

scholarly journals Generalization-Based Acquisition of Training Data for Motor Primitive Learning by Neural Networks

2021 ◽  
Vol 11 (3) ◽  
pp. 1013
Author(s):  
Zvezdan Lončarević ◽  
Rok Pahič ◽  
Aleš Ude ◽  
Andrej Gams
Keyword(s):  
Neural Networks ◽  
Dimensionality Reduction ◽  
Gaussian Process Regression ◽  
Search Space ◽  
Robot Learning ◽  
Training Data ◽  
Practical Applications ◽  
Latent Space ◽  
Real Robot ◽  
Low Dimensional

Autonomous robot learning in unstructured environments often faces the problem that the dimensionality of the search space is too large for practical applications. Dimensionality reduction techniques have been developed to address this problem and describe motor skills in low-dimensional latent spaces. Most of these techniques require the availability of a sufficiently large database of example task executions to compute the latent space. However, the generation of many example task executions on a real robot is tedious, and prone to errors and equipment failures. The main result of this paper is a new approach for efficient database gathering by performing a small number of task executions with a real robot and applying statistical generalization, e.g., Gaussian process regression, to generate more data. We have shown in our experiments that the data generated this way can be used for dimensionality reduction with autoencoder neural networks. The resulting latent spaces can be exploited to implement robot learning more efficiently. The proposed approach has been evaluated on the problem of robotic throwing at a target. Simulation and real-world results with a humanoid robot TALOS are provided. They confirm the effectiveness of generalization-based database acquisition and the efficiency of learning in a low-dimensional latent space.

Improved interactive color visualization approach for hyperspectral images

2021 ◽  
pp. 147387162110481
Author(s):  
Haijun Yu ◽  
Shengyang Li
Keyword(s):  
Real Time ◽  
Hyperspectral Images ◽  
High Dimensional ◽  
Interactive Control ◽  
Output Image ◽  
Dr Method ◽  
The Rich ◽  
Low Dimensional ◽  
Color Visualization ◽  
Fusion Coefficient

Hyperspectral images (HSIs) have become increasingly prominent as they can maintain the subtle spectral differences of the imaged objects. Designing approaches and tools for analyzing HSIs presents a unique set of challenges due to their high-dimensional characteristics. An improved color visualization approach is proposed in this article to achieve communication between users and HSIs in the field of remote sensing. Under the real-time interactive control and color visualization, this approach can help users intuitively obtain the rich information hidden in original HSIs. Using the dimensionality reduction (DR) method based on band selection, high-dimensional HSIs are reduced to low-dimensional images. Through drop-down boxes, users can freely specify images that participate in the combination of RGB channels of the output image. Users can then interactively and independently set the fusion coefficient of each image within an interface based on concentric circles. At the same time, the output image will be calculated and visualized in real time, and the information it reflects will also be different. In this approach, channel combination and fusion coefficient setting are two independent processes, which allows users to interact more flexibly according to their needs. Furthermore, this approach is also applicable for interactive visualization of other types of multi-layer data.

Sign in / Sign up

Export Citation Format

Share Document