首页> 外文期刊>Applied Sciences >High Precision Indoor Visible Light Positioning Algorithm Based on Double LEDs Using CMOS Image Sensor
【24h】

High Precision Indoor Visible Light Positioning Algorithm Based on Double LEDs Using CMOS Image Sensor

机译:基于CMOS图像传感器的双LED高精度室内可见光定位算法。

获取原文
           

摘要

Visible Light Positioning (VLP) is widely recognized as a cost-effective solution for indoor positioning with increasing demand. However, the nonlinearity and highly complex relationship between three-dimensional world coordinate and two-dimensional image coordinate hinders the good performance of image-sensor-based VLP. Therefore, there is a need to develop effective VLP algorithms to locate the positioning terminal using image sensor. Besides, due to the high computational cost of image processing, most existing VLP systems do not achieve satisfactory performance in terms of real-time ability and positioning accuracy, both of which are significant for the performance of indoor positioning system. In addition, the accurate identification of the ID information of each LED (LED-ID) is important for positioning, because if the LED-ID is not recognized well, the positioning can only be achieved in a particular positioning unit and cannot be applied to a large scene with many LEDs. Therefore, an effective image-sensor-based double-light positioning system is proposed in this paper to solve the above problems. We also set up relevant experiments to test the performance of the proposed system, which utilizes the rolling shutter mechanism of the Complementary Metal Oxide Semiconductor (CMOS) image sensor. Machine learning was used to identify the LED-ID for better results. Simulation results show that the proposed double-light positioning system could deliver satisfactory performance in terms of both the real-time ability and the accuracy of positioning. Moreover, the proposed double-light positioning algorithm has low complexity and takes the symmetry problem of angle into consideration, which has never been considered before. Experiments confirmed that the proposed double-light positioning system can provide an accuracy of 3.85 cm with an average computing time of 56.28 ms, making it a promising candidate for future indoor positioning applications.
机译:可见光定位(VLP)被广泛认为是一种对室内定位具有更高性价比的解决方案。但是,三维世界坐标和二维图像坐标之间的非线性和高度复杂的关系阻碍了基于图像传感器的VLP的良好性能。因此,需要开发有效的VLP算法以使用图像传感器来定位定位终端。此外,由于图像处理的高计算成本,大多数现有的VLP系统在实时能力和定位精度方面均不能获得令人满意的性能,这两者对于室内定位系统的性能都是重要的。另外,准确识别每个LED的ID信息(LED-ID)对于定位很重要,因为如果不能很好地识别LED-ID,则只能在特定的定位单元中实现定位,并且无法应用于具有许多LED的大型场景。因此,为解决上述问题,本文提出了一种有效的基于图像传感器的双光定位系统。我们还建立了相关实验来测试所提出系统的性能,该系统利用了互补金属氧化物半导体(CMOS)图像传感器的滚动快门机制。机器学习被用来识别LED-ID,以获得更好的结果。仿真结果表明,所提出的双光定位系统在实时性和定位精度上都可以提供令人满意的性能。此外,所提出的双光定位算法具有较低的复杂度,并且考虑了角度的对称性问题,这是以前从未考虑过的。实验证实,所提出的双光定位系统可以提供3.85 cm的精度,平均计算时间为56.28 ms,使其成为未来室内定位应用的有希望的候选者。

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