中文摘要：

为了提高活动轮廓分割图像的精度,解决传统活动轮廓不能够收敛到深凹陷和弱边界对象分割效果不佳等问题,提出了自适应扩散梯度矢量流（AD-GGVF）算法。首先,在外部力场中,使用基于分量的归一化方法代替传统的基于矢量的归一化方法,提高活动轮廓曲线进入深凹陷的能力;然后,将拉普拉斯算子分解为切向和法向分量,并增加两个互相关的自适应权重函数,使轮廓曲线能够根据图像的局部特征自适应调节扩散过程;最后,以分割结果的量化误差为评价标准,和传统的活动轮廓分割效果进行对比和分析。实验结果表明,本文算法针对两幅不同的弱边界图像,量化误差分别降低到0.08和0.09,活动轮廓曲线能够收敛到深凹陷的底部;分割效果较为准确。

英文摘要：

In order to solve the problems that the active contours cannot converge into deep indentation and poor segmentation results of images with weak edge structures, we propose an adaptive diffusion gradient vector flow algorithm. First, in the external force field, we adopt the normalization method based on component instead of the normalization method based on vector. It can improve the ability of curve to converge into deep depression. Second, we decompose the Laplace operator into tangential and normal components, and add two cross-correlation adaptive weighting functions. These two functions can make the curve adjust the diffusion process adaptively on the basis of local image features. Finally, we compare and analyze the segmentation results of several active contour models using quantization errors. Experimental results show that the convergence problem of deep indentation has been solved effectively. Aiming at two different images which own different image structures and weak boundaries, quantitative errors of the segmentation results reduce to 0.08 and 0. 09 ,respectively. Active contours can converge to the bot- tom of different deep depressions. The curves can also converge to the edges and corners according to the local structure features of images. The results of segmentation on real images are much more accurate.