中文摘要：

利用现势性较好的真彩色遥感影像,基于已有水域矢量边界数据,提出了一种改进的GVF蛇模型水域矢量边界数据更新方法。将已有水域矢量边界数据作为GVF蛇模型初始位置,用真彩色遥感影像彩色梯度作为GVF求解中的边缘梯度,通过轮廓演化实现水域矢量边界精化,以满足水域矢量边界数据更新的要求。实验表明,本方法相比传统的GVF蛇模型、四元数及数学形态学方法具有更高的精度,而且在一定距离范围内对初始边界位置不敏感。

英文摘要：

An automated updating of water vector edge based on improved GVF Snakes model from true color remote sensing image is presented in this paper.The existing water vector edge data are used to initialize the contour.The color gradient from true color remote sensing image is used to generate the gradient vector flow.Compared to the original GVF Snakes model,quaternion and morphological method,the proposed method has the better updating accuracy and the same efficiency.And it is not sensitive to the initial vector edge in the range of a certain distance.