According to the features of high-resolution panchromatic imagery of Beijing-1 small satellite,an approach to extracting information of residential areas is proposed in this paper based on Gabor texture segmentation.T...According to the features of high-resolution panchromatic imagery of Beijing-1 small satellite,an approach to extracting information of residential areas is proposed in this paper based on Gabor texture segmentation.The algorithm extracts the features in different directions and different scales by building the Gabor filter,uses cluster analysis of multiple features to segment the image,and performs the fusion processing based on morphological scale space.It solves the problems in image processing resulting from low contrast between remote sensing objects and background,the blurring of image edges and high noise.It has the benefits of direction selection and frequency selection with strong self-adaptive ability.Our experiments prove the effectiveness of the approach for extracting information of residential areas from Beijing-1 high-resolution imagery.展开更多
基金supported by a grant from the National High Technology Research and Development Program of China(863 Program)(No.2005AA133013,No.2006CB701305)the Major State Basic Research Development Program of China(973 Program)(No.2006CB701305).
文摘According to the features of high-resolution panchromatic imagery of Beijing-1 small satellite,an approach to extracting information of residential areas is proposed in this paper based on Gabor texture segmentation.The algorithm extracts the features in different directions and different scales by building the Gabor filter,uses cluster analysis of multiple features to segment the image,and performs the fusion processing based on morphological scale space.It solves the problems in image processing resulting from low contrast between remote sensing objects and background,the blurring of image edges and high noise.It has the benefits of direction selection and frequency selection with strong self-adaptive ability.Our experiments prove the effectiveness of the approach for extracting information of residential areas from Beijing-1 high-resolution imagery.