Classification of Multispectral images helps in remote sensing applications. A method which uses Harmonic Analysis based classification is proposed. Harmonic Analysis (HA) is introduced to extract the feature from hyperspectral image. Amplitude and phase features has been obtained by derived HA. Then select best feature among extracted feature by Principle Component Analysis (PCA). Finally the bands were classified with the help of Fuzzy C means clustering algorithm. The performance of the process is measured with the help of classifier accuracy. The input multispectral images were obtained from dataset. The features were extracted from the images based on Harmonic analysis (HA). HA transforms the pixels from time domain into frequency domain. The images were divided into different spectral levels. Amplitude, phase and residual values were extracted as features from each spectral levels as a resulting in set of harmonic features. From the extracted Harmonic analysis features best features were selected based on Principle Component Analysis (PCA). The selected best features were then classified using Fuzzy C means (FCM) clustering process. The performance of the process is measured based on the performance metrics like Accuracy.