A Hyperspectral remote sensing image (HSI) composed of various intrinsic components such as shading, albedo, noise and continuous narrow bands in different wavelengths. The classification of the HSI image is one of the challenging tasks in the area of Remote Sensing as it has numerous applications on environment, mineral exploration, target detection and anomaly detection. The present paper identifies a novel approach in classifying the image by incorporating the albedo intrinsic component retrieved from the image on principal components and factor analysis obtained through the dimensionality reduction. The obtained results are classified via Support Vector Machine classifier. The proposed algorithm tested on the benchmark datasets available worldwide such as Indian Pines, University of Pavia and Salinas. The extraction of albedo intrinsic components helps in effective classification of HSI image and outperforms the results with state of the art techniques, achieved the overall accuracy (OA) on these datasets.