AccidentPrevention

This project involves detection and prevention of road side accidents using the library known as OpenCv

This involves creation of some recognising technology that can detect the chances of collision if two cars are close to each by buzzing of certain alarm this also takes in consideration of fact that certain minimum velocity is required for the two cars to start detection

In this module which is set to minimum value of 50m/s to start the process of detection.

This approach involves certain steps for detection of certain cars

Step 1

This step involves detection of certain cars using certain datasets of images which when converted into

xml file of certain information about the file which is in the form of Haarcascade classifiers

Then these Haars of certain images are taken from the dataset and then for detection of certain cars we

turn on our camera for detection purpose we are using Ip camera for capturing of video

Then within the process of capturing certain videos we are converting every frame of certain image to

grayscale inorder to detect the cars.

Step 2

After detection of certain cars we are finding edges where there are edges in that cars

by certain methods such as erosion for shrinking of borders of an image then dilation to expand borders

for perfect detection of certain canny edges in an image.Then after finding edges in an images we are

plotting contours in order to find certain end points of that image so as to get sharp entries of an image

then we use imutils for grabbing of all contours regions in order to get all the points of an image

Then we sort all the contours of an image.Then based on all those contour areas we are plotting

all the box points of certain images.Then we minimze this area inorder to calculate the minimum area for that

certain image then after defining thos minimum area region we have taken a reference point in our case it is

our camera which is taken as one reference object from which the distance between object is calculated then

after calculating certain distance using euclidian distance formula for finding the distance between two refernce objects

Step 3

Power Sytem Analysis

This Step involves finding of certain readings using a sample dataset for generation of power when a car is travelling from one place to another via road as by travelling on road generates certain power and then using these power criteria we had found certain estimates that will lead to generation of certain power when we are travelling certain distance and then after travelling that particular distance we are creating a plot where we will analyse our power system.

Some steps that are needed to be implemented

After finding the distance from the reference object we need to calculate the distance between two objects so

as to find the distance between then and in order to do that we also need to set certain velocity so as that

threshold velocity os required for object detection to be started.And we also need to construct a code so as to

make kind of buzzer alarm code so as that will buz when car will reach at that certain minimum distance

This project is a direct implementation of team DARK KNIGHTS for geekathon 2019 which was not possible without

the help of other members:-

1. Ishank Tiwari ()

2. Sparsh Jain (sparshjain99)