Special Issue "Advances on Clustering Algorithms for Image Processing"

A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Remote Sensing Image Processing".

Deadline for manuscript submissions: closed (30 September 2020).

Special Issue Editor

Prof. Dr. Pasi Fränti
E-Mail Website
Guest Editor
School of Computing, University of Eastern Finland, 80101 Joensuu, Finland
Interests: clustering; machine learning; data mining; location-based applications
Special Issues and Collections in MDPI journals

Special Issue Information

Dear Colleagues,

Clustering methods have been actively developed for decades for applications in databases, data analysis, web mining, recognition systems, pattern recognition, and also image processing. Clustering depends on two things: Objective function such as sum-of-squared errors (SSE), and the algorithm that tries to optimize this function.

Simple algorithms like k-means are still widely used. Recent results have shown that with change of the initialization technique and by repeating the algorithm 100 times, one can reduce the error so that the algorithm is well suited for most pattern recognition applications. Some other applications may require more accurate clustering, and better methods like random swap are needed. These techniques can be evaluated on benchmarking datasets.

An open question is to what extent these results apply to clustering for image processing. What clustering algorithm and which objective function should be used in remote sensing? It is expected that the role of the algorithm is less critical in image processing, and the choice of the features and objective function are more important in the clustering. In image segmentation, a simple extension of k-means adds the pixel location (x,y) to the color value (r,g,b) and then uses the existing clustering methods.

The number of clusters should also be solved. A common approach is to use the heuristic merge-based criterion, which effectively leads to agglomerative clustering with the stopping criterion. An open question is whether cluster validity indexes can be applied instead of the heuristic criterion.

This call for papers invites submissions of new methods and review papers that study how clustering methods are effectively applied in image processing. The application can be image enhancement, filtering, segmentation, object extraction or any other process that is used in remote sensing. Review papers based on systematic comparison of single components are especially welcome.

Prof. Pasi Fränti
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Clustering
  • Image processing
  • Segmentation
  • Satellite images
  • GPS

Published Papers (2 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Article
Improving K-Nearest Neighbor Approaches for Density-Based Pixel Clustering in Hyperspectral Remote Sensing Images
Remote Sens. 2020, 12(22), 3745; https://doi.org/10.3390/rs12223745 - 14 Nov 2020
Cited by 2 | Viewed by 684
Abstract
We investigated nearest-neighbor density-based clustering for hyperspectral image analysis. Four existing techniques were considered that rely on a K-nearest neighbor (KNN) graph to estimate local density and to propagate labels through algorithm-specific labeling decisions. We first improved two of these techniques, a KNN [...] Read more.
We investigated nearest-neighbor density-based clustering for hyperspectral image analysis. Four existing techniques were considered that rely on a K-nearest neighbor (KNN) graph to estimate local density and to propagate labels through algorithm-specific labeling decisions. We first improved two of these techniques, a KNN variant of the density peaks clustering method dpc, and a weighted-mode variant of knnclust, so the four methods use the same input KNN graph and only differ by their labeling rules. We propose two regularization schemes for hyperspectral image analysis: (i) a graph regularization based on mutual nearest neighbors (MNN) prior to clustering to improve cluster discovery in high dimensions; (ii) a spatial regularization to account for correlation between neighboring pixels. We demonstrate the relevance of the proposed methods on synthetic data and hyperspectral images, and show they achieve superior overall performances in most cases, outperforming the state-of-the-art methods by up to 20% in kappa index on real hyperspectral images. Full article
(This article belongs to the Special Issue Advances on Clustering Algorithms for Image Processing)
Show Figures

Graphical abstract

Article
Sketch-Based Subspace Clustering of Hyperspectral Images
Remote Sens. 2020, 12(5), 775; https://doi.org/10.3390/rs12050775 - 29 Feb 2020
Cited by 5 | Viewed by 1421
Abstract
Sparse subspace clustering (SSC) techniques provide the state-of-the-art in clustering of hyperspectral images (HSIs). However, their computational complexity hinders their applicability to large-scale HSIs. In this paper, we propose a large-scale SSC-based method, which can effectively process large HSIs while also achieving improved [...] Read more.
Sparse subspace clustering (SSC) techniques provide the state-of-the-art in clustering of hyperspectral images (HSIs). However, their computational complexity hinders their applicability to large-scale HSIs. In this paper, we propose a large-scale SSC-based method, which can effectively process large HSIs while also achieving improved clustering accuracy compared to the current SSC methods. We build our approach based on an emerging concept of sketched subspace clustering, which was to our knowledge not explored at all in hyperspectral imaging yet. Moreover, there are only scarce results on any large-scale SSC approaches for HSI. We show that a direct application of sketched SSC does not provide a satisfactory performance on HSIs but it does provide an excellent basis for an effective and elegant method that we build by extending this approach with a spatial prior and deriving the corresponding solver. In particular, a random matrix constructed by the Johnson-Lindenstrauss transform is first used to sketch the self-representation dictionary as a compact dictionary, which significantly reduces the number of sparse coefficients to be solved, thereby reducing the overall complexity. In order to alleviate the effect of noise and within-class spectral variations of HSIs, we employ a total variation constraint on the coefficient matrix, which accounts for the spatial dependencies among the neighbouring pixels. We derive an efficient solver for the resulting optimization problem, and we theoretically prove its convergence property under mild conditions. The experimental results on real HSIs show a notable improvement in comparison with the traditional SSC-based methods and the state-of-the-art methods for clustering of large-scale images. Full article
(This article belongs to the Special Issue Advances on Clustering Algorithms for Image Processing)
Show Figures

Graphical abstract

Back to TopTop