Satellite images multiple data using clustering techniques

Abstract EN

Clustering is considered one of the complex tasks in data mining and plays an important role in many applications such as image processing.

Different types of algorithms have been appeared for clustering.

In this paper two unsupervised classification algorithms will apply on Landsat-8 satellite images, k-means clustering and fuzzy c-means with two approaches pixel based clustering and block based clustering.

In block based clustering color features and texture features are extracted.

In texture features gray level co-occurrence matrix (GLCM) is used.

Finally, the results are used for comparison between the two algorithms.

The obtained results according to the proposed method for the satellite images clustering shows that k-means clustering algorithm gave better results with (74.2615 and 83.5906), while fuzzy c-means algorithm gave results with (71.06933 and 81.7031).