首页> 外文期刊>Journal of Imaging Science and Technology >Preferred Racial Skin Color Reproduction in an Image based on Race Classification
【24h】

Preferred Racial Skin Color Reproduction in an Image based on Race Classification

机译:基于种族分类的图像中的首选种族肤色再现

获取原文
获取原文并翻译 | 示例
           

摘要

Preferred skin color reproduction is an important factor for improving the image quality of display devices and systems. Therefore, this article presents a race-selective preferred skin color reproduction method for the case of multiple races in a single image. First, the proposed method detects skin regions using a focus map and skin candidate map. Next, the race of each detected skin region is classified by comparing the Euclidean distance of the average chromaticity for each detected skin region with a database of racial skin colors. Based on this classification, the racial preferred skin color for each detected skin region is then selected from candidate racial preferred skin colors that are predefined in a database generated through an observer's preference test. Finally, the skin color is reproduced toward the preferred skin color. Experimental results confirm that the proposed method can achieve race-selective preferred skin color reproduction in the case of multiple races in a single image. (C) 2015 Society for Imaging Science and Technology.
机译:优选的肤色再现是改善显示设备和系统的图像质量的重要因素。因此,本文针对单个图像中存在多个种族的情况,提出了一种种族选择的首选肤色再现方法。首先,所提出的方法使用聚焦图和皮肤候选图来检测皮肤区域。接下来,通过将每个检测到的皮肤区域的平均色度的欧氏距离与种族皮肤颜色的数据库进行比较,对每个检测到的皮肤区域的种族进行分类。然后,基于此分类,从通过观察者的偏好测试生成的数据库中预定义的候选种族偏爱肤色中选择每个检测到的皮肤区域的种族偏爱肤色。最终,皮肤颜色朝着优选的皮肤颜色再现。实验结果证实,在单个图像中存在多个种族的情况下,所提出的方法可以实现种族选择性的首选肤色再现。 (C)2015年影像科学与技术学会。

著录项

  • 来源
    《Journal of Imaging Science and Technology》 |2015年第2期|020504.1-020504.9|共9页
  • 作者单位

    Kyungpook Natl Univ, Sch Elect Engn, Daegu 702701, South Korea;

    Kyungpook Natl Univ, Sch Elect Engn, Daegu 702701, South Korea;

    Kyungpook Natl Univ, Sch Elect Engn, Daegu 702701, South Korea;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号