基于弱小边缘剔除的区域增长改进算法  

Improved Algorithm of Region Growing Based on Weak Edge Reject

作　　者：唐湘成[1] 黄自力[1] 霍建亮[1] 肖柯[1] 刘怡[1] 

机构地区：[1]西南技术物理研究所,四川成都610041

出　　处：《电视技术》2015年第6期18-21,共4页Video Engineering

摘　　要：提出了一种基于弱小边缘剔除的区域增长改进算法。该算法引入了两幅图像主要结构相似度判别公式,在图像预处理时,剔除了同一目标中的细小纹理和弱小边缘,保留了图像中目标主要结构。仿真实验表明,该算法能够避免在区域增长算法目标分割时,同一个目标被分割为多个小区域的情况,提高了目标分割和提取的准确性。An improved algorithm of region growing based on weak edge reject is proposed in this paper, which introduces the discriminant formula of similarity about main structure between two images. On the image-processing stage, the small texture and weak edge of the same goal are rejected while the main structure remains there. It is indicated from the simulation experiment that the algorithm has the capability to avoid the possibility that the same goal is divided into several small regions when the goal segmentation through the traditional region-growing algorithm is conducted, thus the veracity of segmentation and extraction can be further enhanced.

关 键 词：区域增长 弱小边缘剔除 图像分割 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]