eduzhai > Applied Sciences > Engineering >

Frost filtered scale-invariant feature extraction and multilayer perceptron for hyperspectral image classification

  • Save

... pages left unread,continue reading

Document pages: 11 pages

Abstract: Hyperspectral image (HSI) classification plays a significant in the field ofremote sensing due to its ability to provide spatial and spectral information.Due to the rapid development and increasing of hyperspectral remote sensingtechnology, many methods have been developed for HSI classification but still alack of achieving the better performance. A Frost Filtered Scale-InvariantFeature Transformation based MultiLayer Perceptron Classification (FFSIFT-MLPC)technique is introduced for classifying the hyperspectral image with higheraccuracy and minimum time consumption. The FFSIFT-MLPC technique performs threemajor processes, namely preprocessing, feature extraction and classificationusing multiple layers. Initially, the hyperspectral image is divided intonumber of spectral bands. These bands are given as input in the input layer ofperceptron. Then the Frost filter is used in FFSIFT-MLPC technique forpreprocessing the input bands which helps to remove the noise fromhyper-spectral image at the first hidden layer. After preprocessing task,texture, color and object features of hyper-spectral image are extracted atsecond hidden layer using Gaussian distributive scale-invariant featuretransform. At the third hidden layer, Euclidean distance is measured betweenthe extracted features and testing features. Finally, feature matching iscarried out at the output layer for hyper-spectral image classification. Theclassified outputs are resulted in terms of spectral bands (i.e., differentcolors). Experimental analysis is performed with PSNR, classification accuracy,false positive rate and classification time with number of spectral bands. Theresults evident that presented FFSIFT-MLPC technique improves the hyperspectralimage classification accuracy, PSNR and minimizes false positive rate as wellas classification time than the state-of-the-art methods.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...