摘要
This paper introduces a new enhancement method for multi-spectral satellite remote sensing imagery,based on principal component analysis(PCA) and intensity-hue-saturation(IHS) transformations.The PCA and the IHS transformations are used to separate the spatial information of the multi-spectral image into the first principal component and the intensity component,respectively.The enhanced image is obtained by replacing the intensity component of the IHS transformation with the first principal component of the PCA transformation,and undertaking the inverse IHS transformation.The objective of the proposed method is to make greater use of the spatial and spectral information contained in the original multi-spectral image.On the basis of the visual and statistical analysis results of the experimental study,we can conclude that the proposed method is an ideal new way for multi-spectral image quality enhancement with little color distortion.It has potential advantages in image mapping optimization,object recognition,and weak information sharpening.
This paper introduces a new enhancement method for multi-spectral satellite remote sensing imagery, based on principal component analysis （PCA） and intensity-hue-saturation （IHS） transformations. The PCA and the THS transformations are used to separate the spatial information of the multi-spectral image into the first principal component and the intensity component, respectively. The enhanced image is obtained by replacing the intensity component of the IHS transformation with the first prin- cipal component of the PCA transformation, and undertaking the inverse IHS transformation. The objective of the proposed method is to make greater use of the spatial and spectral information contained in the original multi-spectral image. On the basis of the visual and statistical analysis results of the experimental study, we can conclude that the proposed method is an ideal new way for multi-spectral image quality enhancement with little color distortion. It has potential advantages in image mapping optimization, object recognition, and weak information sharpening.
