《计算机应用研究》|Application Research of Computers

基于相似性的中智学图像分割方法

Neutrosophic image segmentation approach based on similarity

免费全文下载 (已被下载 次)  
获取PDF全文
作者 赵鑫,王士同
机构 江南大学 a.物联网工程学院;b.数字媒体学院,江苏 无锡 214122
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2012)06-2371-04
DOI 10.3969/j.issn.1001-3695.2012.06.099
摘要 针对传统图像分割算法抗噪性差的问题, 提出基于相似性的中智学图像分割方法。该方法在中智学基础上, 利用图像信息的不确定性, 结合相似性运算对图像信息进行处理。根据像素点的不确定性, 图像在中智学领域内经相似性运算和图像增强后, 利用聚类将其分割。实验结果显示, 该方法可以有效剔除噪声, 提高图像的信噪比, 对合成图像分割错误率仅为0. 110 7, 低于其他方法, 表明本方法在抗噪性以及图像分割效果上比其他方法更为理想。
关键词 图像分割;中智学;不确定性;相似性运算;图像增强;聚类
基金项目 国家自然科学基金资助项目(60903100,60975027)
江苏省自然科学基金资助项目(BK2009067)
本文URL http://www.arocmag.com/article/01-2012-06-099.html
英文标题 Neutrosophic image segmentation approach based on similarity
作者英文名 ZHAO Xin, WANG Shi-tong
机构英文名 a. School of Internet of Things Engineering, b. School of Digital Media, Jiangnan University, Wuxi Jiangsu 214122, China
英文摘要 According to the noise disturbance of image segmentation, this paper proposed a similarity- based neutrosophic image segmentation approach. Considering the indeterminacy of image information, the concept of similarity, as an operation, was introduced to processing of image information with neutrosophy theory. With indeterminacy of each pixel, it used similarity and enhancement operations to process an image in the field of neutrosophy. Then, it segmented the image using clustering. The experiments show that the proposed method can eliminate noise effectively and improve PSNR of image. The error rate of synthetic image segmentation is only 0. 110 7 and lower than other methods. The proposed method is better in the noise disturbance and segmentation effect.
英文关键词 image segmentation; neutrosophy; indeterminacy; similarity operation; image enhancement; clustering
参考文献 查看稿件参考文献
 
收稿日期
修回日期
页码 2371-2374
中图分类号 TP391
文献标志码 A