Abstract: Weak visual environment problems holes and shock mutation adhesion area, traditional method is prone to over-segmentation phenomenon, the segmentation results in larger deviation, low precision segmentation.Weak to do this, put forward a new visual environment adhesion area and high precision satellite image segmentation method, through the Bernsen algorithm of satellite image binarization processing, based on this, will be able to reflect the shape of the target satellite image boundary complexity factor as the discrimination on the basis of the satellite image of adhesion area to determine the adhesion area, the distance of the satellite images captured by the distance transformation adhesion area diagram, of the seed point selection, on the basis of the seed point to rebuild from the figure, through rapid immersion watershed transformation model of reconstruction image segmentation adhesion area.The experimental results show that the proposed method has high segmentation accuracy and effectiveness.