针对室内成像定位技术受随机噪声的影响较大、定位误差较高的问题, 提出了一种基于像素距离加权的室内成像定位技术。在室内屋顶布设多个红外LED, 依靠成像传感器获得红外LED信标的像点, 将成像点到成像传感器中心的像素距离作为加权因子引入室内成像定位算法中, 可以有效地提高室内定位精度。并进行了仿真实验, 实验选择4m×4m×3m的空间区域模拟室内环境, 当布设的红外LED信标数量为3时, 应用改进后的算法可以获得10cm以内的定位误差性能, 并且误差波动不超过5cm。另外, 随着布设信标数量的增加, 定位误差继续减小。改进后的定位算法有效地提高了室内定位的精度以及成像定位算法的普适性。

Aiming at the problems that the indoor imaging localization is easily affected by random noise and the localization error is high, an indoor imaging localization technique based on pixel distance weighting is proposed. A number of infrared LEDs were layed on the indoor roof, the image points of infrared LED beacon were obtained with the imaging sensor, and the pixel distance between the imaging point and center of the imaging sensor was considered as the weight factor, thus the indoor positioning accuracy can be effectively improved by introducing indoor imaging positioning algorithm. Simulation experiments were carried out to simulate the spatial environment of 4m×4m×3m. When the number of infrared LED beacons is 3, the algorithm can obtain the positioning error performance within 10cm and the error fluctuation will not exceed 5cm. In addition, with the number increase of laying beacons, the positioning error decreases. The optimized positioning algorithm effectively improves the accuracy of indoor positioning and universality of imaging localization algorithm.