Hyper spectral Image Analysis using Harmonic Analysis with BFO Optimized RVM

Image processing is playing a vital role in all fields like satellite, medical, telecommunication, and missile. Hyper spectral images show similar statistical properties to natural grayscale or color photographic images. HSI (Hyper Spectral Image) are a more challenging area because of high spectral bands and dimensionality. As well as it is very easy to learn. It will be used to identify the problem in various fields like Signal Processing, and moreover used to determine the complex manifolds. There are several algorithms have been proposed to classify the hyper spectral image. In our paper new methods have been introduced that is Harmonic Analysis based classification such as HA-BFO-RVM (Bacterial Foraging Optimization – Relevance Vector Machine) approach. This new approach accurately classifies the cluster band by respect to their amplitude and phase. Harmonic Analysis (HA) is introduced to extract the features of hyper spectral image. Amplitude and phase features have been obtained by deriving HA. Then select the best features, among extracted features by Bacterial Foraging Optimization. Finally, classify the respective band by a related cluster, which are performed by the help of Relevance Vector Machine (RVM). This classifier accurately classifies the band to respective cluster form. In prior work, instead of HA, used MNF, PCA and ICA could extract features and also in combination of PSO-SVM could use CV-SVM and GA-SVM. In this process will be carried out by integrating HA-BFO-RVM. This combination leads to provide good accuracy and also limited computer time because of using the BFO Method. Keywords—Harmonic analysis (HA); hyper spectral image classification (HSI); Bacterial Foraging optimization (BFO); Relevance vector machine (RVM).