Distracted Driver Detection and Classification
Journal Title: International Journal of engineering Research and Applications - Year 2018, Vol 8, Issue 4
Abstract
The number of road accidents due to distracted driving has been on a rise in the recent years. As per the Union Road Transport and Highways Ministry Report 2016, 17 people were killed each hour in India due to road accidents. This makes it imperative to take measures to curb the number of road fatalities. The major cause of these accidents is driver error. This paper proposes solution to detect the distraction of driver, thus averting the possible accidents. The use of different Convolutional Neural Network (CNN) models namely: Small CNN, VGG16, VGG19, Inception for classification of distracted drivers according to State Farm Distracted Driver Detection challenge on Kaggle are depicted in this paper. The deep learning library used for the purpose is Keras running on top of TensorFlow. Our best result is a categorical cross entropy loss of 0.899 on the validation set.
Authors and Affiliations
Prof. Pramila M. Chawan, Shreyas Satardekar, Dharmin Shah, Rohit Badugu, Abhishek Pawar
Keywords
Related Articles
- EP ID EP394014
- DOI 10.9790/9622-0804036064.
- Views 112
- Downloads 0
Study Of Mechanical Properties Of LM-4 Reinforcement With Zro2 For Structural Application
Composite materials are engineered or naturally occurring materials made from two or more constituent materials with significantly different physical or chemical properties which remain separate and distinct within the f...
Consequences and Influence of Loads That Are Not contemplated In the Structure Calculation Process In some regulations.
The approach to the calculation of the structure of a building, apart from the geometry and its peculiarities always in the first instance, the first approach is to analyze the state of loads, according to current legisl...
Software Reuse metrics and their impacts
Reusability is not only the process of using already existing artifacts to develop new software; it is the process of applying existing solution to the new arrived problem. Reusability not only reduces the rework, develo...
A Novel Approach to Analyze Satellite Images for Severe Weather Events
Severe weather events (e.g. cyclones, thunderstorms, fog, floods etc.) have catastrophic effects on human life and aggrandize the impacts on agriculture, economy etc. Therefore information related to severe weather, in a...
A Review on Fluoride in Ground Water: Causes, Effects and Solutions
Water, the most vital resource for all kinds of life on this planet, is also the resource, adversely affected both quantitatively and qualitatively by all kind of human activities on land, in air or in water or naturally...