Selective capture based high-speed optical vehicular signaling system

Abstract

The present article introduces a high-speed optical vehicular signaling system using a distinct and unique capturing strategy called selective capture (SC). The proposed scheme performs the SC within the frame while capturing, not after the frames are captured, thereby reducing the capturing area and thus increasing the camera capture speed. One of the major challenges in optical camera communication (OCC) based vehicular signaling system is low data transmission rate, due to a low sampling rate of a camera receiver, compared with very high-speed modulation of light emitting diodes (LEDs) used as the transmitter. We propose this SC-based technique for a high-speed optical vehicular signaling system. Experiments are conducted to verify the proposed scheme while analyzing various parameters such as margin setting to calculate a total SC area, considering the movement of a vehicle along the roadside, inter-vehicle distance measurement and LED detection from a Raspberry Pi camera (RaspiCam) module. The SC is performed using template matching technique on the RaspiCam module and then the margin is added with the SC to obtain the total SC area. This total SC area is utilized for capturing the taillights of the leading vehicle (LV). The optical vehicle-to-vehicle (V2V) signaling system utilizes two 4 × 4 red LED arrays mounted on LV as the taillights and a RaspiCam mounted on the following vehicle (FV). It is found that the use of RaspiCam set with the total SC area to capture the taillights significantly increases the capture speed of RaspiCam from 120 frames per second (fps) to 435 fps, yielding an efficient and high-speed V2V signaling with flicker-free taillights. Experiment results demonstrate that the proposed optical V2V provides a data rate of up to 6.912 kbps and achieves acceptable bit error rate performance at a distance of up to 175 cm.