DESIGN AND IMPLEMENTATION OF VEHICLE DETECTION ON SMART HEADLIGHT FOR CAR BASED ON VEHICLE ATTITUDE AND SURROUNDING USING IMAGE PROCESSING

• Main Author: ABDUH TAUFAN - NIM : 13213004, MUHAMAD
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/29066
• Institution: Institut Teknologi Bandung

<p align="justify">Smart headlight for car is an intelligent headlight system in the car that can adjust its behavior according to the input of the car attitude and visual condition in front of the car. One of the features of this system is to lower the headlight from High Beam condition to Low Beam if there are other vehicles in front of the car. To support these features, it is required that the vehicle has a sensing sensor system in front of the car that can be processed quickly enough to the movement of the headlights, for the drivers’ convenience. One method of such sensors is to use image processing to detect the presence of vehicles in front of the car. The algorithm used are Red Detection (as a red color detection on the vehicle's rear lights for vehicles in the same direction) and Bright Detection (as a vehicle headlight detection for vehicles in opposite directions). A webcam is used to captre the image and it is forwarded to Raspberry Pi 3, a microcomputer that will process the incoming video and send commands to Arduino Nano, a microcontroller that will adjust the direction of the motion of the lamp. Using this system, 82% detection accuracy was obtained, with a processing time of 0.111 s. It is also obtained that the farthest distance detection is 72 m for Red Detection and 100 m for Bright Detection.<p align="justify">