摘要
由于图像的复杂性和模糊性进一步增强,传统的图像分割算法已经无法满足其对分割精度的要求。为了进一步提高图像分割的精度,本文提出了一种结合SIFT特征提取与Delaunay三角网表达的图像分割方法,该方法可以有效削弱噪声对图像分割结果的影响,与传统去噪滤波相比,平滑过程中模糊图像边缘的程度比较低,同时,运用超像素思想,将传统基于像素单元的分割方法运用到超像素上,对被三角网划分的子区域进行聚类,最后得出分割结果。与传统算法相比,本文算法在分割精度方面有显著提高。
Due to the complexity and ambiguity of the image,the traditional image segmentation algorithm cannot meet the requirement of segmentation precision. In order to further improve the accuracy of image segmentation,this paper proposes an image segmentation method combined with SIFT feature extraction and Delaunay triangulation,which can be used to reduce the influence of noise on image segmentation effectively. Compared with traditional denoising filter,the smoothing process of the edge of the blurred image is relatively small. At the same time,the traditional pixel-based segmentation method is applied to the super-pixel,and the sub-regions divided by the triangulation are clustered. Finally,the segmentation result is obtained. Compared with the traditional algorithm,the algorithm has improved significantly in terms of segmentation accuracy.
作者
王崇倡
韩旭
WANG Chongchang;HAN Xu(School of Geomatics,Liaoning Technical University,Fuxin 123000,China)
出处
《测绘与空间地理信息》
2019年第3期216-221,共6页
Geomatics & Spatial Information Technology