应用视觉显著性的快速有偏聚类超像素算法

您当前的位置：

首页 >

文章列表页 >

应用视觉显著性的快速有偏聚类超像素算法

更新时间：2025-07-09

- 应用视觉显著性的快速有偏聚类超像素算法
- Issue 1, Pages: 112-117+138(2015)
- 作者机构：
  
  1. 上海交通大学自动化系
  2. 西安交通大学电子与信息工程学院
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391.41
- Published：2015
- 稿件说明：
移动端阅览
[J]. 2015, (1): 112-117+138.
DOI：

[J]. 2015, (1): 112-117+138. DOI：

摘要

针对正则化超像素方法的超像素数随边缘拟合要求迅速增长的问题

提出了一种有偏聚类超像素算法。结合人类视觉对目标专注程度不一的特点

在SLIC算法框架下

提出了基于视觉显著性的非均匀初始化方法和有偏聚类距离函数。算法在图像的显著性区域进行密集的过分割

保持目标边缘的细节信息

而在非显著区域仅生成稀疏的超像素

以降低分割块数

再通过一步全局聚类和边缘逐步细化过程

有效地保证了图像的边缘拟合

同时提高了算法的速度。实验表明

在相同超像素数下

所提算法在边缘查全率、欠分割错误率以及运行速度方面均优于传统算法。

Abstract

关键词

Keywords

references

{"db_type":"WWJDREF","db_name":"国际期刊","reference_articles":[{"unit":"International Journal of Computer Vision","flag":"[J]","year":2009,"author":"Jamie Shotton;John Winn;Carsten Rother;Antonio Criminisi","index":1,"title":"TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and Context"},{"unit":"International Journal of Computer Vision","flag":"[J]","year":2008,"author":"Stephen Gould;Jim Rodgers;David Cohen;Gal Elidan;Daphne Koller","index":2,"title":"Multi-Class Segmentation with Relative Location Prior."},{"unit":"International Journal of Computer Vision","flag":"[J]","year":2004,"author":"Pedro F. Felzenszwalb;Daniel P. Huttenlocher","index":3,"title":"Efficient Graph-Based Image Segmentation."}],"articles_count":3}

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Effects of Steam Addition on the Swirl Combustion and Emission Characteristics of Cracked Ammonia

Research on the Electrohydrodynamic Effects of Ionic Liquid-Modified Fuels

Numerical Simulation of the Energy Conversion Process in Thermoacoustic Engines with Different Stack Materials

Study on the Thermodynamic Properties in the Anomalous Region of Supercritical Water

Energy Efficiency Optimization of the Cryogenic Air Separation Process for Electronic-Grade Ultra Pure Nitrogen

Related Author

No data

Related Institution

No data

⁰