ハイパースペクトルデータを用いた比演算画像・カラー合成処理におけるバンド選定アルゴリズムの一提案

抄録

This paper presents a GA(Genetic algorithms)-based band selection procedure for producing a Color composite Ratioing image with Maximum Entropy (CRME image) using hyperspectral data (e.g., hyperion data). The more number of the hyperspectral bands, the more dicult the ecient and eective processing for producing color composite images. In producing the CRME image, one of the requirements is to maximize the amount of information in the images (i.e., image entropy). Through the GA operations, the increasing of image entropy of CRME image (i.e., tness value) was conrmed, which corroborates the GA operations can be applied for band selection in producing the CRME image. Toward the end of run in the GA operations, three ratioing images in case of maximizing the entropy of CRME image are selected and based on the spectrum reection characteristic, those are assigned to red-, blue-, and green-plane, respectively. Compared with the true, natural and false color composite images, we conclude the produced CRME image is useful for interpretation of image features in terms of the amount of information in the image.