Shadow compensation method of remote sensing image based on improved gamma transform
Image shadow is an important interference item in the application and interpretation of remote sensing images.In order to improve the accuracy of shadow compensation and restore more truthfully the feature information of objects on remote sensing images,an improved shadow compensation method based on extracting homogeneous regions is proposed.Two remote sensing images with different features were se-lected as the experimental area.First,the image of the experimental area is suppressed by blue light.Secondly,the homogeneous shaded and non-shaded areas are obtained through expansion and corrosion operations to construct the structural elements needed in the experimental area.Finally,a Gamma transformation is used to compensate the shadows of remote sensing images for homogeneous regions,com-pare the accuracy with Retinex algorithm and local compensation algorithm.The experimental results show that the average gradient of the improved Gamma transformation is better than the other two algorithms in each color channel,which verifies the validity of this method to improve the shadow compensation accuracy of remote sensing images,and provides a new idea to improve the shadow compensation accuracy of remote sensing images.
remote sensing imageshadow compensationGamma transformhomogeneous region
NETL
NSTL
万方数据
