通信与信息技术Issue(2)：5-10,16,7.
多信息原融合的目标实时定位导航算法
Real-time target location and navigation algorithm based on multi-information source fusion
摘要
Abstract
In order to improve the performance of the algorithm of target localization and navigation in indoor environment,a real-time target localization and navigation algorithm based on multi-information source fusion of smart phone is proposed.In the image pre-processing stage,firstly,adaptive mean filter and adaptive median filter are used to filter the grayscale image,and then edge detection of the filtering results of these two filters is carried out through the Canny operator,and the binary image of the image edge is generated re-spectively.Finally,the generated binary image is fused by OR operation.In the stage of straight-line detection,the Hough transform method is used to detect the long line of the image by designing reasonable angle and straight-line distance.In the stage of target position-ing and navigation,the processed image data is fused with the data collected by the inertial sensor,and the image and dead reckoning al-gorithm are combined to realize real-time target positioning and navigation.The experimental results show that,compared with the origi-nal algorithm,the proposed algorithm is more stable for detecting the depth,width and height data of the scene,and reduces the jitter of detected data.In term of mapping,the proposed algorithm has the advantage of more accurate and smaller offset of peripheral width trajec-tory.This means that the proposed algorithm of indoor localization and navigation get better robustness and effectiveness.关键词
自适应中值滤波/自适应均值滤波/霍夫变换/航迹推算/视觉定位Key words
Adaptive median filter/Adaptive mean filter/Hough transform/Dead reckoning/Visual positioning分类
计算机与自动化引用本文复制引用
王宇宁,裴氏莺,韦耿..多信息原融合的目标实时定位导航算法[J].通信与信息技术,2025,(2):5-10,16,7.基金项目
广西科技计划项目(项目编号:桂科AD21238038) （项目编号:桂科AD21238038）
