基于改进的活动轮廓图像分割

Image Segmentation Based on An Improved Active Contour

图像分割是由图像处理到图像分析的关键步骤,也是一种基本的计算机视觉技术。针对传统的活动轮廓模型在分割过程中具有处理速度慢,运算量大,对凹陷轮廓处理效果差等缺点,提出了一种改进的活动轮廓图像分割技术.在传统的活动轮廓中,外部能量通常由目标点的梯度势能场给出,然而梯度势能场存在着一些难以克服的缺点,即不能够很好地指导曲线的移动。把梯度向量流场(GVF)作为外部能量场,有效地克服了传统梯度势能场捕捉范围小以及难以处理凹平面的缺点,并通过实验证明了该方法的有效性。

Image segmentation is a key process from image processing to image analysis, which is also a basic technique in computer vision. Because the traditional active contour model has many defects such as low speed , huge computation , the paper proposes an improved parametric active contour model .The external force of the traditional active contour is given by the gradient potential energy, which has some insurmountable shortcomings, thus it cannot direct the contour to move to the destination correctly .The GVF field maintains the merit of diffusing the energy field to the slowly changed range , which the traditional gradient energy is very httle. The GVF not only expands the effective range of the energy field but also enhances the ability to deal with the concave surface . The experimental results show the validity of the method.