Compression-based unsupervised clustering of spectral signatures

Cerra, Daniele; Bieniarz, Jakub; Avbelj, Janja; Reinartz, Peter; Mueller, R.

doi:10.1109/whispers.2011.6080862

Cited by 9 publications

(8 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We consider a dataset used in [2] to test a proposed compression-based similarity measure for time series, the Normalized Compression Distance (NCD). The authors in [2] compared NCD with Euclidean Distance, Spectral Angle, Spectral Correlation, and Spectral Information Divergence on a set of 41 spectra from three different classes of rock, and showed that their proposed distance measure gives the best clustering results in this domain.…”

Section: Rock Categorizationmentioning

confidence: 99%

“…The authors in [2] compared NCD with Euclidean Distance, Spectral Angle, Spectral Correlation, and Spectral Information Divergence on a set of 41 spectra from three different classes of rock, and showed that their proposed distance measure gives the best clustering results in this domain.…”

Section: Rock Categorizationmentioning

confidence: 99%

“…We present the result of hierarchical clustering using the u-shapelets in Figure 9 (left). To help the reader compare our clustering result with [2], we have included a screen capture of their best clustering result.…”

Section: Rock Categorizationmentioning

confidence: 99%

“…Cerra et al in [2] be isolated by 'cutting' the tree at convenient points." If we cut the tree of Figure 9 (right) at the points where we placed the pink lines, we find that the eleven nodes encircled are not clustered correctly.…”

Section: Rock Categorizationmentioning

confidence: 99%

See 3 more Smart Citations

Clustering Time Series Using Unsupervised-Shapelets

Zakaria

Mueen

Keogh

2012

2012 IEEE 12th International Conference on Data Mining

168

103

View full text Add to dashboard Cite

Abstract-Time series clustering has become an increasingly important research topic over the past decade. Most existing methods for time series clustering rely on distances calculated from the entire raw data using the Euclidean distance or Dynamic Time Warping distance as the distance measure. However, the presence of significant noise, dropouts, or extraneous data can greatly limit the accuracy of clustering in this domain. Moreover, for most real world problems, we cannot expect objects from the same class to be equal in length. As a consequence, most work on time series clustering only considers the clustering of individual time series "behaviors," e.g., individual heart beats or individual gait cycles, and contrives the time series in some way to make them all equal in length. However, contriving the data in such a way is often a harder problem than the clustering itself.In this work, we show that by using only some local patterns and deliberately ignoring the rest of the data, we can mitigate the above problems and cluster time series of different lengths, i.e., cluster one heartbeat with multiple heartbeats. To achieve this we exploit and extend a recently introduced concept in time series data mining called shapelets. Unlike existing work, our work demonstrates for the first time the unintuitive fact that shapelets can be learned from unlabeled time series. We show, with extensive empirical evaluation in diverse domains, that our method is more accurate than existing methods. Moreover, in addition to accurate clustering results, we show that our work also has the potential to give insights into the domains to which it is applied.

show abstract

Section: Rock Categorizationmentioning

confidence: 99%

Section: Rock Categorizationmentioning

confidence: 99%

Section: Rock Categorizationmentioning

confidence: 99%

Section: Rock Categorizationmentioning

confidence: 99%

See 2 more Smart Citations

Clustering Time Series Using Unsupervised-Shapelets

Zakaria

Mueen

Keogh

2012

2012 IEEE 12th International Conference on Data Mining

168

103

View full text Add to dashboard Cite

show abstract

“…Such distance can be applied to diverse data types [11], and one of its main advantages is its parameter-free approach, as the NCD depends only on the compressor adopted, with performance comparisons for general compression algorithms showing this dependance to be loose [12]. The NCD and its variants have been applied to spectral signatures collected from hyperspectral sensors [13,14], SAR images [15], multispectral data and satellite image time series [8,16].…”

Section: Preliminariesmentioning

confidence: 99%

Expanding the Algorithmic Information Theory Frame for Applications to Earth Observation

Cerra

Datcu

2013

Entropy

View full text Add to dashboard Cite

Recent years have witnessed an increased interest towards compression-based methods and their applications to remote sensing, as these have a data-driven and parameter-free approach and can be thus succesfully employed in several applications, especially in image information mining. This paper expands the algorithmic information theory frame, on which these methods are based. On the one hand, algorithms originally defined in the pattern matching domain are reformulated, allowing a better understanding of the available compression-based tools for remote sensing applications. On the other hand, the use of existing compression algorithms is proposed to store satellite images with added semantic value.

show abstract