Self-organizing Map for Clustering of Remote Sensing Imagery
نویسندگان
چکیده
We present a neural unsupervised pattern recognition approach for two applications related to significant topics of Earth Observation (EO) imagery: (a) EO image region classification; (b) multispectral pixel classification. The proposed model is based on the Self-Organizing Map (SOM) clustering, which is compared to two benchmark unsupervised classifiers: k-means and fuzzy c-means. We propose to apply the Davies-Bouldin index for cluster separation measure. The best classification scores are obtained by the proposed SOM approach for both applications. The experimental results prove the efficiency of the Davies-Bouldin measure to automatically detect the number of clusters in an unclassified dataset.
منابع مشابه
Cluster Analysis in Remote Sensing Spectral Imagery through Graph Representation and Advanced SOM Visualization
The Self-Organizing Map (SOM), a powerful method for clustering and knowledge discovery, has been used effectively for remote sensing spectral images which often have high-dimensional feature vectors (spectra) and many meaningful clusters with varying statistics. However, a learned SOM needs postprocessing to identify the clusters, which is typically done interactively from various visualizatio...
متن کاملApplication of Data Mining Techniques for Remote Sensing Image Analysis
The paper studies the applicability of various data mining techniques on aerial remote sensing imagery for automatic land-cover classification. Four techniques are applied, namely the Adaptive Dynamic K-means (ADK), Self Organizing Feature Map (SOFM), Machine Learning Induction Algorithm (C4.5) and Support Vector Machines (SVM). Special attention is drawn to the usefulness of these data mining ...
متن کاملAn Artificial Immune System Approach for Unsupervised Pattern Recognition in Multispectral Remote-Sensing Imagery
This paper presents an improved Artificial Immune System (AIS) approach for unsupervised classification in multispectral remote-sensing imagery. For benchmarking, one has considered several unsupervised nature-inspired intelligent classifiers (AIS, neural, fuzzy) versus statistical ones. We have comparatively evaluated the following pattern recognition techniques: the proposed AIS model; Self-O...
متن کاملKohonen Self Organizing for Automatic Identification of Cartographic Objects
Automatic identification and localization of cartographic objects in aerial and satellite images have gained increasing attention in recent years in digital photogrammetry and remote sensing. Although the automatic extraction of man made objects in essence is still an unresolved issue, the man made objects can be extracted from aerial photos and satellite images. Recently, the high-resolution s...
متن کاملRemote Detection of Cerebral Pathologies in Magnetic Resonance Imagery: an Unsupervised Heuristic Approach
In this work, the attention has been focused to the field of “medical imaging”. The problem of pattern recognition in remote sensing with medical application has been discussed in order to detect significant lesions in encephalic non-invasive diagnostics. In particular, Nuclear Magnetic Resonance analysis have been considered. The aim is to propose an automatic way to recognize encephalic patho...
متن کامل