Semi-Supervised Graph-Based Hyperspectral Image Classification

Camps‐Valls, Gustau; Bandos, Tatyana V.; Zhou, Dengyong

doi:10.1109/tgrs.2007.895416

Cited by 577 publications

(272 citation statements)

References 28 publications

Supporting

Mentioning

262

Contrasting

Unclassified

Order By: Relevance

“…With much less labeled samples, the proposed algorithm still produces better results than the compared competitors introduced in [10,19,20]. With just 10 labeled samples per class, the results obtained by the proposed supervised and semi-supervised algorithms are 77.98% and 80.15%, respectively.…”

Section: Real Aviris Imagementioning

confidence: 88%

“…With just 10 labeled samples per class, the results obtained by the proposed supervised and semi-supervised algorithms are 77.98% and 80.15%, respectively. Using 5 labeled samples per class, the best accuracy presented in [20] is 66.04%, whereas proposed semi-supervised approach yields 72.62%, which is 6.58% larger than 66.04% from [20].…”

Section: Real Aviris Imagementioning

confidence: 90%

“…Table 1 shows segmentation results as a function of the size of the initial labeled samples. These results are compared with those provided by LORSAL and by the state-of-the-art algorithms for semi-supervised hyperspectral classification and segmentation [10,19,20]. The proposed semi-supervised algorithm yields effective results, as shown in the segmentation map displayed in Fig.…”

Section: Real Aviris Imagementioning

confidence: 99%

See 2 more Smart Citations

Semi-supervised hyperspectral image classification based on a Markov random field and sparse multinomial logistic regression

Bioucas-Dias

2009

2009 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

This paper introduces a new semi-supervised classification and segmentation approach tailored to hyperspectral images. The posterior distributions of the classes are modeled by the multinomial logistic regression. The contextual information inherent to the spatial configuration of the image pixels is modeled by a Multi-Level Logistic (MLL) Markov-Gibbs random field. The multinomial logistic regressors, assumed to be random vectors with independent Laplacian components, are learned using the recently introduced LOR-SAL algorithm. The maximum a posteriori (MAP) segmentation is computed via the α-Expansion algorithm, a powerful graph cut based approach to integer optimization. The effectiveness of the proposed methodology is illustrated by classifying simulated and real data sets. Comparisons with state-of-art methods are also included.

show abstract

Section: Real Aviris Imagementioning

confidence: 88%

Section: Real Aviris Imagementioning

confidence: 90%

Section: Real Aviris Imagementioning

confidence: 99%

See 1 more Smart Citation

Semi-supervised hyperspectral image classification based on a Markov random field and sparse multinomial logistic regression

Bioucas-Dias

2009

2009 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

show abstract

“…The whole image containing 16 classes is considered. In first scenario, we followed the procedure presented in [12].Therefore, we present classification for 9 classes and 5 labeled samples only. Results for different dimensional reduction approaches are presented in Fig.…”

Section: Methodsmentioning

confidence: 99%

Morphological scale-space for hyperspectral images and dimensionality exploration using tensor modeling

Velasco-Forero

Angulo

2009

2009 First Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing

View full text Add to dashboard Cite

This paper proposes a framework to integrate spatial information into unsupervised feature extraction for hyperspectral images. In this approach a nonlinear scale-space representation using morphological levelings is formulated. In order to apply feature extraction, Tensor Principal Components are computed involving spatial and spectral information. The proposed method has shown significant gain over the conventional schemes used with real hyperspectral images. In addition, the proposed framework opens a wide field for future developments in which spatial information can be easily integrated into the feature extraction stage. Examples using real hyperspectral images with high spatial resolution showed excellent performance even with a low number of training samples.

show abstract

“…In [15], Huynh and Robles-Kelly employed a compact B-Spline basis to represent reflectance spectra. On the other hand, graph-based approaches usually aim at classifying rather than reconstructing hyperspectral images [4,6].…”

Section: Introductionmentioning

confidence: 99%

A Comparative Evaluation of Spectral Reflectance Representations for Spectrum Reconstruction, Interpolation and Classification

Huynh

Robles-Kelly

2013

2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops

View full text Add to dashboard Cite

Semi-Supervised Graph-Based Hyperspectral Image Classification

Cited by 577 publications

References 28 publications

Semi-supervised hyperspectral image classification based on a Markov random field and sparse multinomial logistic regression

Semi-supervised hyperspectral image classification based on a Markov random field and sparse multinomial logistic regression

Morphological scale-space for hyperspectral images and dimensionality exploration using tensor modeling

A Comparative Evaluation of Spectral Reflectance Representations for Spectrum Reconstruction, Interpolation and Classification

Contact Info

Product

Resources

About