Classification performance of reduced color band hyperspectral imagery from JPEG2000 decompressed data

This paper presents a study of the impact of the compression of a hyperspectral image (HSI) using the JPEG2000 standard. The impact measure was the accuracy of the classification. The goal of this study is to present an approach that proposes to reduce the computational and memory costs of processing this type of data. The results presented show a savings of storage and savings in the computational time by decompressing only a small portion of the color-bands.