HSEG and PCA for hyper-spectral image classification

This paper presents an image classification method named hierarchical set of image segmentation (HSEG) for hyperspectral image. Hyperspectral imaging sensor measure the energy of the received light in tens of hundreds of narrow spectral band in each spatial position in an image. The high number of spectral bands acquired by spectral sensors increase the capability to distinguish physical material and objects. The processing of hyper spectral images is applied in two stages dimensionality reduction and unsupervised classification techniques. The high dimensionality of data has been reduced with the help of Principal Component Analysis (PCA). The selected dimensions are classified using Hierarchical set of image segmentation. Experiments on hyperspectral image indicate that this algorithm can give better result than conventional method. HSEG has high computational speed and better segmentation accuracy.