On the Mapping of Burned Areas and Burn Severity Using Self Organizing Map and Sentinel-2 Data
2020
In this letter, an approach based on the use of Sentinel-2 spectral indices and self-organizing map (SOM) is proposed to automatically map burned areas and burned severity. The methodological approach proposed is based on three steps: 1) indices computation; 2) maps of the difference of the three indices computed using the data acquired from prefire and postfire occurrences; and 3) unsupervised classification obtained processing all the difference maps using the SOM.
Unsupervised Change Detection in Multispectral Remote Sensing Images via Spectral-Spatial Band Expansion
Content: Due to the limited number of spectral channels in multispectral remote sensing images, change information, especially the multiclass changes, may be insufficiently represented, resulting in inaccurate detection of changes. In this paper, we propose to use unsupervised band expansion techniques to generate artificial spectral and spatial bands to enhance the change representation and discrimination for change detection (CD) from multispectral images. In particular, in the proposed approach, two simple nonlinear functions, i.e., multiplication and division, are applied for spectral expansion. Multiscale morphological reconstruction is used to extend the band spatial information.
Unsupervised Multitemporal Domain Adaptation With Source Labels Learning
2019
In order to achieving accurate multitemporal alignment on a few spectral bands’ high-resolution images, source label learning step is proposed in this letter and used to optimize traditional manifold alignment (MA). The core of this method is to improve the erroneous manifold structure by combining majority voting and weighting coefficients. Besides, this method is a universal step and can be used for optimizing all MA methods. Two groups of data sets captured by Chinese GF1 and GF2 satellites are used for performance evaluation.
Automatic and Unsupervised Water Body Extraction Based on Spectral-Spatial Features Using GF-1 Satellite Imagery
2019 垃圾文章，表示不清
Log-Based Transformation Feature Learning for Change Detection in Heterogeneous Images
Cite: T. Zhan, M. Gong, X. Jiang and S. Li, “Log-Based Transformation Feature Learning for Change Detection in Heterogeneous Images,” in IEEE Geoscience and Remote Sensing Letters, vol. 15, no. 9, pp. 1352-1356, Sept. 2018, doi: 10.1109/LGRS.2018.2843385.
Content: In this letter, we propose an unsupervised change detection method for heterogeneous synthetic aperture radar (SAR) and optical images based on the logarithmic transformation feature learning framework. First, the logarithmic transformation is applied to the SAR image that aims to achieve similar statistical distribution properties as the optical image. Then, high-level feature representations can be learned from the transformed image pair via joint feature extraction, which are used to select reliable samples for training a neural network classifier. When it is trained well, a robust change map can be obtained, thus identifying changed regions accurately.
Joint Dictionary Learning for Multispectral Change Detection
2017
In this paper, an improved sparse coding method for change detection is proposed. The intuition of the proposed method is that unchanged pixels in different images can be well reconstructed by the joint dictionary, which corresponds to knowledge of unchanged pixels, while changed pixels cannot. First, a query image pair is projected onto the joint dictionary to constitute the knowledge of unchanged pixels. Then reconstruction error is obtained to discriminate between the changed and unchanged pixels in the different images. To select the proper thresholds for determining changed regions, an automatic threshold selection strategy is presented by minimizing the reconstruction errors of the changed pixels.
Toward Automated Land Cover Classification in Landsat Images Using Spectral Slopes at Different Bands
2017
Content: This paper proposes a spectral-slope-based classification technique and subsequently summarize the changes in temporal image sets. Using the properties of spectral slopes, we propose a set of rules for selection of training samples from Landsat imageries for classifying the land cover. The images are initially classified into three classes: water, vegetation, and vegetation-void. Further, vegetation and vegetation-void regions are classified into proper vegetation and dry cropland, and urban land and bare land, respectively. The initial classification is performed by support vector machines, and the second-level classification is performed through k-means clustering and subsequent labeling of clusters to subclasses. At last, considering temporal images of the same scene, postclassification change summarization is carried out to quantize the land variations, both qualitatively (type of change) and quantitatively (volume of change).
Implementation of machine-learning classification in remote sensing: an applied review
2018
This article provides an overview of machine learning from an applied perspective. it focuses on the relatively mature methods of support vector machines, single decision trees (DTs), Random Forests, boosted DTs, artificial neural networks, and k-nearest neighbours (k-NN). Issues considered include the choice of algorithm, training data requirements, user-defined parameter selection and optimization, feature space impacts and reduction, and computational costs. We illustrate these issues through applying machine-learning classification to two publically available remotely sensed data sets.
A Method for the Analysis of Small Crop Fields in Sentinel-2 Dense Time Series
2017
This paper presents a method suitable for the analysis of small crop fields in Sentinel-2 dense satellite image time series that accounts for Sentinel-2 characteristics. The method fuses spatiotemporal information, analyzes data spatio-temporal evolution, and extracts relevant spatio-temporal information.
Remote Sensing Scene Classification by Unsupervised Representation Learning
2017
Content: In this paper, an unsupervised representation learning method is proposed to investigate deconvolution networks for remote sensing scene classification.
Unsupervised Multilayer Feature Learning for Satellite Image Scene Classification
2017
Content: This letter proposes a simple but effective approach to automatically learn a multilayer image feature for satellite image scene classification.
Iterative Classifiers Combination Model for Change Detection in Remote Sensing Imagery
2016
Content: In this paper, we propose a new unsupervised change detection method designed to analyze multispectral remotely sensed image pairs. The proposed method is in the category of the committee machine learning model that utilizes an ensemble of classifiers (i.e., the set of segmentation results obtained by several thresholding methods) with a dynamic structure type. More specifically, in order to obtain the final “change/no-change” output, the responses of several classifiers are combined by means of a mechanism that involves the input data (the difference image) under an iterative Bayesian–Markovian framework.
Change detection of bitemporal multispectral images based on FCM and D-S theory
2016
Content: In this paper, a change detection method of bitemporal multispectral images based on the D-S theory and fuzzy c-means (FCM) algorithm is proposed. Firstly, the uncertainty and certainty regions are determined by thresholding method applied to the magnitudes of difference image (MDI) and spectral angle information (SAI) of bitemporal images. Secondly, the FCM algorithm is applied to the MDI and SAI in the uncertainty region, respectively. Then, the basic probability assignment (BPA) functions of changed and unchanged classes are obtained by the fuzzy membership values from the FCM algorithm. In addition, the optimal value of fuzzy exponent of FCM is adaptively determined by conflict degree between the MDI and SAI in uncertainty region. Finally, the D-S theory is applied to obtain the new fuzzy partition matrix for uncertainty region and further the change map is obtained.
Adaptive Multiobjective Memetic Fuzzy Clustering Algorithm for Remote Sensing Imagery
2015 code
This paper proposes an adaptive multiobjective memetic fuzzy clustering algorithm (AFCMOMA) for remote sensing imagery. In AFCMOMA, a multiobjective memetic clustering framework is devised to optimize the two objective functions, i.e., Jm and the Xie-Beni (XB) index. First, in order to balance the local and global search capabilities in memetic algorithms, an adaptive strategy is used, which can adaptively achieve a balance between them, based on the statistical characteristic of the objective function values. In addition, in the multiobjective memetic framework, in order to acquire more individuals with high quality, a new population update strategy is devised, in which the updated population is composed of individuals generated in both the local and global searches.
A New Self-Training-Based Unsupervised Satellite Image Classification Technique Using Cluster Ensemble Strategy
2015
Content: This letter addresses the problem of unsupervised land-cover classification of remotely sensed multispectral satellite images from the perspective of cluster ensembles and self-learning. The cluster ensembles combine multiple data partitions generated by different clustering algorithms into a single robust solution.
A Multisensor Multiresolution Method for Mapping Vegetation Status, Surficial Deposits, and Historical Fires Over Very Large Areas in Northern Boreal Forests of Quebec, Canada
Content: In this context, a mapping method was developed and was capable of dealing with very large areas with a lack of support datasets, which could characterize the current forest, surficial deposits, and forest disturbance history.It involved five steps: 1) mapping the vegetation based on unsupervised classification, imputation, and segmentation methods; 2) mapping the history of fires that occurred over the mapping area based on archives Landsat images; 3) determining the dominant species characterizing forest stands; 4) mapping surficial deposits; and 5) accuracy assessment of map attributes based on video dataset.
Soft clustering - fuzzy and rough approaches and their extensions and derivatives
2013
Content: This paper compares k-means to fuzzy c-means and rough k-means as important representatives of soft clustering. On the basis of this comparison, important extensions and derivatives of these algorithms are then surveied; the particular interest here is on hybrid clustering, merging fuzzy and rough concepts. Some examples are also given where k-means, rough k-means, and fuzzy c-means have been used in studies.
Classification of Satellite Images using New Fuzzy Cluster Centroid for Unsupervised Classification Algorithm
2013 Conference
Content: In this work a, new objective function is formulated by adding the new term along
with the distance between the pixels and cluster centers in the spectral domain. This new term is formulated by multiplying the Lagrange’s multiplier with the membership values of the pixel for a particular class is subtracted with one. It gives weightage to the instance of a particular pixel.
Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and k-Means Clustering
2009
Cite: T. Celik, “Unsupervised Change Detection in Satellite Images Using Principal Component Analysis and -Means Clustering,” in IEEE Geoscience and Remote Sensing Letters, vol. 6, no. 4, pp. 772-776, Oct. 2009, doi: 10.1109/LGRS.2009.2025059.
Content: In this letter, we propose a novel technique for unsupervised change detection in multitemporal satellite images using principal component analysis (PCA) and k-means clustering. The difference image is partitioned into h h nonoverlapping blocks. The orthonormal eigenvectors are extracted through PCA of h h nonoverlapping block set to create an eigenvector space. Each pixel in the difference image is represented with an S-dimensional feature vector which is the projection of h h difference image data onto the generated eigenvector space. The change detection is achieved by partitioning the feature vector space into two clusters using k-means clustering with k = 2 and then assigning each pixel to the one of the two clusters by using the minimum Euclidean distance between the pixel’s feature vector and mean feature vector of clusters.
Visualization and unsupervised classification of changes in multispectral satellite imagery
2006 劣
Content: The statistical techniques of multivariate alteration detection, minimum/maximum autocorrelation factors transformation, expectation maximization and probabilistic label relaxation are combined in a unified scheme to visualize and to classify changes in multispectral satellite data. The methods are demonstrated with an example involving bitemporal LANDSAT TM imagery.
Unsupervised classification of satellite imagery: choosing a good algorithm
2002
Content: In the context of land-cover classification with multispectral satellite data several unsupervised classification (clustering) algorithms are investigated and compared with regard to their ability to reproduce ground data in a complex landscape. Ground data is extended to the entire scene using a supervised neural network classification algorithm. The clustering algorithms examined are K-means, extended K-means, agglomerative hierarchical, fuzzy K-means and fuzzy maximum likelihood. Fuzzy clustering is found to perform best relative to a reference scene obtained with the Landsat Thematic Mapper 5 (TM5) platform.