Nonlocal graph theory based transductive learning for hyperspectral image classification

Huang, Baoxiang; Ge, Linyao; Chen, Ge; Radenkovic, Milena; Wang, Xiaopeng; Duan, Jinming; Pan, Zhenkuan

doi:10.1016/j.patcog.2021.107967

Cited by 33 publications

(13 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jasper dataset has four endmembers' contributions corresponding to tree, soil, water, and road. A graph-based architecture is proposed in [33] to classify the endmember contributions and, the authors compare the classification performance with ML techniques such as SVM, KNN, LDAKNN, PCAKNN, KPCAKNN, LDASVM, PCASVM KPCASVM, Convolution Neural Network (CNN), the abbreviation Linear Discriminant Analysis (LDA), Kernel PCA (KPCA) in the machine learning methods means the preprocessing step before the classification techniques. The accuracy measurements for Jasper image for classification into the endmembers contribution by class is as follows: Soil Class obtained 100% accuracy results using the PCAKNN method and SVM 99.905%, for water, and TLM-2 classifier obtained 98.959%, Finally, for tree class, TLM-2 obtained 97.622%.…”

Section: Discussion Of Ensemble Model Results For Jasper Hsimentioning

confidence: 99%

“…On the other hand, for two classes using as a label sub-set of non-belong (NB) and belong (B), we obtained 91.67%, 100.00%, and 100.00% for the three classes of trees, water, and soil, respectively. We improve the results compared to the method proposed in [33] for our two sub-labels approach. The best scaling method for Jasper dataset was the min-max-scaling.…”

Section: Discussion Of Ensemble Model Results For Jasper Hsimentioning

confidence: 99%

See 1 more Smart Citation

Hyperspectral Image Labeling and Classification Using an Ensemble Semi-Supervised Machine Learning Approach

Manian

Alfaro-Mejía

Tokars

2022

Sensors

View full text Add to dashboard Cite

Hyperspectral remote sensing has tremendous potential for monitoring land cover and water bodies from the rich spatial and spectral information contained in the images. It is a time and resource consuming task to obtain groundtruth data for these images by field sampling. A semi-supervised method for labeling and classification of hyperspectral images is presented. The unsupervised stage consists of image enhancement by feature extraction, followed by clustering for labeling and generating the groundtruth image. The supervised stage for classification consists of a preprocessing stage involving normalization, computation of principal components, and feature extraction. An ensemble of machine learning models takes the extracted features and groundtruth data from the unsupervised stage as input and a decision block then combines the output of the machines to label the image based on majority voting. The ensemble of machine learning methods includes support vector machines, gradient boosting, Gaussian classifier, and linear perceptron. Overall, the gradient boosting method gives the best performance for supervised classification of hyperspectral images. The presented ensemble method is useful for generating labeled data for hyperspectral images that do not have groundtruth information. It gives an overall accuracy of 93.74% for the Jasper hyperspectral image, 100% accuracy for the HSI2 Lake Erie images, and 99.92% for the classification of cyanobacteria or harmful algal blooms and surface scum. The method distinguishes well between blue green algae and surface scum. The full pipeline ensemble method for classifying Lake Erie images in a cloud server runs 24 times faster than a workstation.

show abstract

Section: Discussion Of Ensemble Model Results For Jasper Hsimentioning

confidence: 99%

Section: Discussion Of Ensemble Model Results For Jasper Hsimentioning

confidence: 99%

Hyperspectral Image Labeling and Classification Using an Ensemble Semi-Supervised Machine Learning Approach

Manian

Alfaro-Mejía

Tokars

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…For heterogeneous areas in HSI, the fixed small region may come from different categories, so only using a fixed size square window will not be feasible [43]. The nonlocal neighborhood idea [44] facilitates us to adopt many diverse regions to integrate the spectral and spatial information, the detailed pipeline of nonlocal diverse regions is shown in Figure 2. Now, spectral-spatial classification techniques via the neural network can use the spatial correlation of adjacent pixels [45].…”

Section: A Nonlocal Spectral Sequence Datamentioning

confidence: 99%

A Spectral Sequence-Based Nonlocal Long Short-Term Memory Network for Hyperspectral Image Classification

Huang

Wang

Shang

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

Efficient classification for hyperspectral image (HSI), which assigns each pixel of the image into a specific category, has been a critical research topic in the HSI analysis area. Under the supervised classification settings, the deep learning approaches are very useful for label prediction. However, most deep learning modeling methods can't get the utmost out of spectral information, which is critically important for object interpretation. Consequently, a spectral sequence-based nonlocal long short-term memory (LSTM) network for HSI classification is proposed in this paper. To boost the dominant role of spectral information, the long short-term memory network for sequential data processing is employed. Furthermore, the nonlocal diverse regions are exploited to learn contextual features for stronger discriminative ability. Meanwhile, the attention mechanism is adopted to increase the classification performance. Experiments on Salinas, Indian Pines (IndianP), Pavia University (PaviaU), and Houston University (HoustonU) datasets are implemented. Benefiting from the idea of processing spectral bands as sequence data and nonlocal diverse regions, spectral-spatial information is fully fused to achieve accurate pixel-level classification. It can be demonstrated that the proposed method can achieve better quantitative and qualitative classification results compared with the other state-of-the-art classification methods.

show abstract

“…Therefore, SSL is gaining researchers' interest. 41 Self-training, 42 Active Learning (AL), 43 graphbased learning 44 and transductive learning 45 are some of the semi-supervised learning methodologies which are used to increase learning performance. The primary goal of all investigations is to increase the training sample size.…”

Section: Semi-supervised Approaches For Classificationmentioning

confidence: 99%

“…The transductive methodology necessitates fewer resources because it is primarily concerned with lowering inference error for a given set of unlabelled data, rather than attempting to enhance the overall quality of the acquired hypothesis. Hence, Huang et al 45 proposed a strategy named transductive learning to overcome the limited labelling training sample and high dimensionality difficulties by extending non-local graph theory to the classification label space. Furthermore, Appice et al 53 used the S2CoTraC algorithm to implement a transductive learning technique based on a co-training schema.…”

Section: Semi-supervised Approaches For Classificationmentioning

confidence: 99%

An outlook: machine learning in hyperspectral image classification and dimensionality reduction techniques

Reddy

Harikiran

2022

J. Spectral Imaging

View full text Add to dashboard Cite

Hyperspectral imaging is used in a wide range of applications. When used in remote sensing, satellites and aircraft are employed to collect the images, which are used in agriculture, environmental monitoring, urban planning and defence. The exact classification of ground features in the images is a significant research issue and is currently receiving greater attention. Moreover, these images have a large spectral dimensionality, which adds computational complexity and affects classification precision. To handle these issues, dimensionality reduction is an essential step that improves the performance of classifiers. In the classification process, several strategies have produced good classification results. Of these, machine learning techniques are the most powerful approaches. As a result, this paper reviews three different types of hyperspectral image machine learning classification methods: cluster analysis, supervised and semi-supervised classification. Moreover, this paper shows the effectiveness of all these techniques for hyperspectral image classification and dimensionality reduction. Furthermore, this review will assist as a reference for future research to improve the classification and dimensionality reduction approaches.

show abstract

Nonlocal graph theory based transductive learning for hyperspectral image classification

Cited by 33 publications

References 38 publications

Hyperspectral Image Labeling and Classification Using an Ensemble Semi-Supervised Machine Learning Approach

Hyperspectral Image Labeling and Classification Using an Ensemble Semi-Supervised Machine Learning Approach

A Spectral Sequence-Based Nonlocal Long Short-Term Memory Network for Hyperspectral Image Classification

An outlook: machine learning in hyperspectral image classification and dimensionality reduction techniques

Contact Info

Product

Resources

About