A. Madhan Kumar et al.; International Journal of Advance Research, Ideas and Innovations in Technology © 2019, www.IJARIIT.com All Rights Reserved Page | 176 ISSN: 2454-132X Impact factor: 4.295 (Volume 5, Issue 2) Available online at: www.ijariit.com Drowsy driving warning and traffic collision information system using IoT Madhan Kumar A. madhan7kumar22@gmail.com Panimalar Institute of Technology, Chennai, Tamil Nadu Kabilan M. masilamanikabi@gmail.com Panimalar Institute of Technology, Chennai, Tamil Nadu Karthikeyan S. B. sb.karthikeyan98@gmail.com Panimalar Institute of Technology, Chennai, Tamil Nadu Sathiyapriya S. sathiyapriya.anbunathan@gmail.com Panimalar Institute of Technology, Chennai, Tamil Nadu ABSTRACT Driver lassitude induced drowsiness accounts for a major of road accidents. To avoid such accidents, a precautionary system is needed. The idea is to develop driver sleep detection, drunk and drive monitoring system. Digital Image Processing (DIP) technique is used to detect fatigue. Haar cascaded algorithm is used for eye detection. In this paper, a system is proposed to avoid road accidents using the alert system along with collision information system. IOT module is used to get location information and to send messages. Wi-Fi module is used to update location information in the server. Keywords— Haar cascade algorithm, DIP, IOT, Wi-Fi 1. INTRODUCTION In today’s world, most people can afford to buy cars and bikes, as it becomes one of the most important and useful things. As vehicle count is increasing, it also leads to heavy traffic and accidents. Nowadays vehicle riding has become very painstaking, as accidents are now frequent in many places. There are many reasons for the accident; one of the reasons is drowsy driving. Drowsy driving can lead to Serious accidents. Other major reason is drunk and drive. This is not only unsafe for the drivers, but also for others who travel on the road. In India, more than 150,000 people are killed each year in traffic accidents year which is about 10% of road crash fatalities worldwide. That's about 400 fatalities a day and far more than developed auto markets like the US. In 2017 As many as 1,47,000 people died on Indian roads in 4,64,000 accidents as per the reports of the Ministry of Road Transport and Highways. According to the reports, drowsy driving was responsible for 72,000 crashes, 44,000 injuries and 800 deaths. Drunk and Drive is also responsible for many accidents, claiming more lives. As per National statistics, an average of nearly 12,000 people dies every year. The main objective of this system is to detect the drowsy state of the driver and to create a warning to the driver. Drowsiness detection can reduce accidents due to this problem. For this more efficient technologies must be used. Accurate detection of eyes must be done to identify the driver’s condition. Gas sensors can detect whether the driver is drunk. Constant Observation of eyes and alcohol gas in the air can be used to avoid accidents. 2. EXISTING SYSTEM In the existing system, ECG and EEG sensor based Drowsiness detection is implemented and sensors are suitable in laboratory monitoring. But during driving, it’s not suitable as it connected to the driver body. So there is no comfort to the person during driving and the sensor value may vary depending upon light intensity. It produces less accurate results. Goggles with Eyeball sensor also used for drowsiness detection, but it is not comfortable to use while driving. 3. PROPOSED SYSTEM In the proposed system, Drowsy driving detection and Drunk and drive detection is used to avoid accidents. In addition to that Traffic collision information system is also used. 3.1 Drowsiness Detection Drowsiness detection is detecting the eyes of the driver. The camera is used for detection instead of eyeball sensors, which is used in the older method. It detects the driver’s eyes and sends the information to raspberry pi 3 which process the information. 3.2 Drunk and drive and Smoke detection This system uses Gas sensors to detect whether the driver is drunk or smoking. Gas sensors react to the gases present, thus changes in the concentration of molecules at the gaseous state are updated. It reacts to the breath of the driver; if Ethanol or