Using Data Pre-Processing and Convolutional Neural Network (CNN) to Mitigate Light Deficient Regions in Visible Light Positioning (VLP) Systems

Hsu, Ling-Yuan; Tsai, Deng-Cheng; Chow, Chi‐Wai; Liu, Yang; Chang, Yun-Han; Lin, Yuan-Zeng; Yeh, Chien‐Hung; Wang, Yichang; Chen, Yiyuan

doi:10.1109/jlt.2022.3184931

Cited by 22 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…VLC-based localization 3 takes advantage of known data associations conveyed by VLC. Often, such systems use high-speed modulated LEDs which broadcast their unique IDs as location beacons, using photodetectors 4 , 5 or cameras 6 – 8 as reception sensors. Photodetector-based methods 4 , 5 usually require accurate propagation channel modeling to estimate the ranges between the receivers and transmitters, so they are not as accurate and robust as camera-based methods due to the elusiveness of RF signals.…”

Section: Related Workmentioning

confidence: 99%

“…Often, such systems use high-speed modulated LEDs which broadcast their unique IDs as location beacons, using photodetectors 4 , 5 or cameras 6 – 8 as reception sensors. Photodetector-based methods 4 , 5 usually require accurate propagation channel modeling to estimate the ranges between the receivers and transmitters, so they are not as accurate and robust as camera-based methods due to the elusiveness of RF signals. For camera-based systems, 6 – 8 a more accurate pose estimation can be achieved for vision-only methods, which need at least three LED features to fix a 6-DoF pose.…”

Section: Related Workmentioning

confidence: 99%

Section: System Overviewmentioning

confidence: 99%

“…In the previous VL localization systems, the whole environmental scenario, i.e., the drive circuit or the luminaire arrangement, has to be redesigned, which is time-and effort-consuming. [4][5][6][7][8] Inspired by the work [9][10][11][12] that uses artificial fiducial landmarks for indoor localization, we use a set of luminaire distribution patterns as the visual landmarks to avoid complicated operations with no degradation in localization accuracy. The system architecture proposed consists of different luminaire distribution patterns located on the ceiling and the detection, identification, and localization algorithms on the receiver side, as shown in Fig.…”

Section: System Overviewmentioning

confidence: 99%

“…Besides their illumination function, LED lights can be reused as artificial visual landmarks for localization. For the current mainstream, LEDs are retrofitted to be modulated at high speed by the transmitter to broadcast their unique identities via visible light communication (VLC), which will later be recognized by photodetectors 4 , 5 or rolling-shutter cameras 6 – 8 The terminal’s position can be resolved trivially by trilateration or triangulation, given at least three individual LED features are detected concurrently by the receiver.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Indoor visible light localization system using luminaire distribution patterns

Zhang

2023

Opt. Eng.

View full text Add to dashboard Cite

High accuracy and low-cost indoor localization has received much attention for the past few years. Visible light (VL) localization, which reuses indoor LED luminaire infrastructure as artificial visual landmarks, has emerged as a promising technology. Using only lights as localization beacons, it can provide highly accurate indoor localization without reconstructing the whole environmental scenario. We propose a localization system that utilizes luminaire distribution patterns on the ceiling as visual landmarks without the costly luminaire circuits as in previous methods. For pattern detection, a unique descriptor for each luminaire cluster is derived through image processing followed by perspective rectification. Then, each pattern's ID in a lookup table is retrieved with earth mover's distance under rotation transformation. Using all individual luminaires in each pattern, our system achieves accurate and robust localization under noise and interference. Furthermore, our system can automatically generate pattern dictionaries of hundreds to thousands of unique IDs resulting in great scalability. Evaluations are conducted through real-world experiments. Compared with previous systems, the proposed method achieves superior performance in terms of detection range, localization, and orientation accuracy. Specifically, our system achieves high average three-dimensional localization accuracy of 2.47 cm and orientation accuracy of 0.87 deg under different heights for the best dictionary setting. Finally, an implementation on a mobile phone demonstrates the potential to run on resource-constrained platforms in real time.

show abstract