基于图像处理的目标物体最大内接矩形面积检测
DOI:
作者:
作者单位:

安徽大学电子信息工程学院,安徽大学电子信息工程学院;北京农业智能装备技术研究中心,安徽大学电子信息工程学院,安徽大学电子信息工程学院,北京农业智能装备技术研究中心

作者简介:

通讯作者:

中图分类号:

TN911.73

基金项目:

安徽省自然科学基金青年基金(1308085QC58);国家自然科学基金(41301471);安徽省高等学校省级自然科学研究项目(KJ2013A026)


Detecting Maximum Inscribed Rectangle Area of Target Object Based on Image Processing
Author:
Affiliation:

School of Electronic and Information Engineering of Anhui University;Beijing Research Center for Information Technology in Agriculture,,,

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    目标物体最大内接矩形的检测对生产矩形产品的工厂意义重大。目前的裁定手段都是人工实现,存在主观性强、效率低、精度差等缺点。本文研究了基于图像处理的遍历法、中心扩散法检测目标物体最大内接矩形,并结合两种方法提出一种新的检测方法:遍历中心扩散法。实验结果表明:遍历法对噪声具有很强的鲁棒性,但对凹面物体十分敏感;中心扩散法适用性较好,但检测的准确性不高;而遍历中心扩散法既有较强的鲁棒性,又拥有较高的准确率,其检测准确率均达到90%以上,实现了目标物体最大内接矩形的高精度检测。

    Abstract:

    It’s great significant for rectangular products factories to detect the maximum inscribed rectangular of target objects. The traditional method is artificial, with some shortages such as strong subjectivity, low efficiency and poor accuracy etc. In this paper, the traversing and center diffusion methods were utilized to detect the maximum inscribed rectangle of target objects-based on image processing, then on the basis, a new method was proposed. The experimental results showed that the traversing was strong robustness to noise, but was sensitive to the concave objects, the applicability of center diffusion was better, but the detection accuracy was lower; while the traversal center diffusion method revealed stronger robustness, and had more than 90% accuracy. Thus the proposed detection method in the work is effective for improving maximum inscribed rectangle of the object.

    参考文献
    相似文献
    引证文献
引用本文

谢新华,梁栋,张香倩,等. 基于图像处理的目标物体最大内接矩形面积检测[J]. 科学技术与工程, 2015, 15(17): .
Xie xinhua, Liang dong, Zhang xiangqian, et al. Detecting Maximum Inscribed Rectangle Area of Target Object Based on Image Processing[J]. Science Technology and Engineering,2015,15(17).

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2015-01-22
  • 最后修改日期:2015-03-20
  • 录用日期:2015-03-18
  • 在线发布日期: 2015-06-16
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注