Feature Extraction by a Wavelet Algorithm for Land Classification

G. Luo (UK)


Pattern recognition, Feature extraction,Wavelet algorithm, Land classification


: Remotely sensed data and the land classification set their own requirements for feature extraction. Algorithms that exploit the potential of the narrow bands while being computationally efficient are needed to capture the features of interest from large imagery data set. This paper presents a wavelet algorithm based on spatial-frequency localisation property of wavelet transform, to reveal the spectral characteristics of image data. Experimental result shows highly effective and accurate feature extraction for land classification using the proposed approach.

Important Links:

Go Back