• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>2011 IEEE 3rd International Conference on Communication Software and Networks >A novel Features from Accelerated Segment Test algorithm based on LBP on image matching
【24h】

A novel Features from Accelerated Segment Test algorithm based on LBP on image matching

机译：基于LBP的加速段测试算法在图像匹配中的新特点。

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Since the FAST (Features from Accelerated Segment Test) detector is much faster than any of the commonly used detection algorithms, and the repeatability and efficiency of FAST-9 are best in the FAST family, we proposed a novel LBP descriptor based on FAST-9 algorithm. Firstly, feature points are detected by FAST-9 and then, the texture information of the interesting point neighborhood is studied with the proposed Fast LBP descriptor named F-LBP, structured the 256-bit descriptor of the interesting point, which effectively reduced the computational complexity of the descriptor. Finally, the descriptor matching is carried out with the logical XNOR. The experimental results show that the proposed algorithm has got good repeatability, efficiency, the rotation invariance, the affine invariance and the illumination invariance in the process of image matching.
机译：由于FAST（加速段测试的特征）检测器比任何常用的检测算法都快得多，并且FAST-9的可重复性和效率在FAST系列中是最好的，因此我们提出了一种基于FAST-9的新型LBP描述符算法。首先通过FAST-9对特征点进行检测，然后使用提出的F-LBP快速LBP描述符对感兴趣点邻域的纹理信息进行研究，构造感兴趣点的256位描述符，从而有效地减少了计算量。描述符的复杂性。最后，使用逻辑XNOR进行描述符匹配。实验结果表明，该算法在图像匹配过程中具有良好的重复性，效率，旋转不变性，仿射不变性和照度不变性。

著录项

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