Automated driving will have a big impact on society, creating new possibilities for mobility and reducing road accidents. Current developments aim to provide driver assistance in the form of conditional and partial automation. The presence of Computer vision technologies inside the vehicles is expected to grow as the automation levels increase. However, embedding a vision-based driver assistance system supposes a big challenge due to the special features of vision algorithms, the existing constrains and the strict requirements that need to be fulfilled. The aim of this project is to leverage Vision based Artificial Intelligence for assisting the driver to the next level. Like Face Recognition Security Feature (Inside Vehicle).Drowsiness Detection System(Inside Vehicle) , Pedestrian Detection (Outside Vehicle).Just by using Camera with powerful Compute Vision and Machine Learning algorithms, the system itself will be very Effective and reduces overall cost for development. Keywords: Maximally Stable External Region, Convoluted Neural Networks, Histogram of Oriented Gradient I. INTRODUCTION With the advent of smart phones, Android has become the dominating mobile OS functioning on over 1.2 million devices worldwide[1]. Android provides an efficient SDK which when used with Android Studio IDE can help create applications quickly and easily. World Health Organisation estimated about 1.35 million death all around the globe due to road traffic which can be approximated to about 1 death in 25 seconds in the year 2016. A majority of these accidents occur due to lack of attentiveness while departing lanes / lane splitting. The purpose of this paper is to create a smart driver assistant which will help the driver make rational decisions based on the real-time environment. The system marks lanes in front of the car with image processed highlighting and colored tracking. Along with this, the system will ensure that the driver never misses a traffic sign. The system constantly grabs each and every traffic symbol along the path and makes the information available to the driver through voice assistant. The following sections provides sufficient background and insight into our objective to develop the Smart Driver Assistant II. RELATED WORKS 1) Title: Analyses Of Driver's Body Movement For Detection Of Hypovigilance Due To Non-Driving Cognitive Task a) Author: M. Itoh, H. Nagasaku and T. Inagaki This paper investigates how driver's body movement may be affected by a nondriving and possibly distractive cognitive task. We have collected data on body movement under several settings, such as cases in which cognitive tasks are given intermittently, or cases in which cognitive tasks are given for a relatively long time period. A driver-adaptable method is proposed for detection of increase in tension via body movement. The method is evaluated with experimental data.

Abstract:The Infrared non-destructive evaluation (IRNDE) is an emerging approach for non-contact inspection of various solid materials such as metals, composites and semiconductors for industrial and research interest. Data processing is an essential step in IRNDE in order to visualize the subsurface defects in the sample and determine the shapes and sizes of the same. The data processing intends to analyse temporal variations in the contrast of each pixel, relative to defect-free reference point on the sample. For that, several post processing algorithms are applied to the recorded thermograms. In this work, it is proposed to an advanced graphical user interface (GUI) for processing the recorded thermograms by IR camera using MATLAB, that supports preprocessing, processing and quantification is presented. Investigations arecarried out using the thermal image sequence recorded with different active thermographic methods: frequency modulated thermal wave imaging (FMTWI) Pulse thermography (PT), lock in thermography (LT). A comparative analysis of the results from different themographic techniques for defect visualization is presented.