BSafe

Problem Statement

The course Big Data and Policing has given us a detailed account about the prominence of Data and how it can influence Policing and general safety. We as students had the chance to attend talks from policemen to lawyers who discussed their role in collecting and analysing data of any form to conduct policing in a smarter way.

Our focus was to try and develop something that can tackle the issue of safety and provide a service that helps in general policing. We decided to come up with an application that could aid the process.

Preliminary Idea

We started off with the idea to develop a web and mobile application primarily intended for women safety. We wanted to collect data about narrow streets and roads and understand how unsafe it would be for women mainly as pedestrians.

The application allows the users to mark a particular spot on the street which they deem as unsafe. It also allows them to enter a short description about the area and optionally upload a photo of how it looks like.

It was determined that the primary data collection and analysis would be through outsourcing within our college community as they would know the nearby areas better. The first phase of the analysis was focused along one particular strip of street.

Further Developments

On analysing the information we received from the initial data plot points, we realised that the application had the potential to serve as a general safety application with women safety still a dedicated sub category. Some of the users marked areas where they had experienced issues of thefts or accidents due to overcrowding or poor lighting. Such shared experiences would help other users to get an idea about that particular locality and be vigilant while travelling along that region.

Additionally, some more streets and roads were analysed using the application to compare and contrast between various regions.

This shows the application with all the existing data plot point from the users so far. Each point is represented by a photo taken and a description that they have given.

This is one particular area of a street where we can see each of the points that the users have marked.

Demonstration

A video demonstration of how a user can plot a point that he/she might think is unsafe is given below

The option to upload a photo has not been shown here as this was strictly for the demonstration. Moreover, this demo is taken from the desktop web browser. The same can be done through the mobile web browser of the user when they are on the move.

Issues being addressed

After getting considerable number of plot points we understood that the application address a few different categories of dangers and safety concerns.

The users shared some past experiences from certain regions of streets, which allows other users to take precautionary measures if they are to commute through that area.

This particular data point describes about an area that is unsafe for pedestrians due to the lack of lighting at night.

This particular plot point shows a usually crowded area which has a history of pick-pocketing.

This data point points out an area in a street that is prone for accidents. (Past experience of the user).

The LUX Analysis

The events of a user marking a particular spot as accident-prone/history of theft is completely user based and those are straight out facts that help the other users. But most of the date collected on this application are related to dim lit places that makes commute difficult and unsafe for pedestrians mainly.

The brightness of the images that each of the user uploads are subjective to the phone camera configurations that they take it in. While collecting the image date as a cue for people to get an idea of an area, we wanted to make more use of the images than just display them on the application. So, we decided to use these images to determine how well lit an area actually would be at night.

Just calculating the brightness of an image was not good enough because of several factors like quality of the camera, post processing etc. That's when we decided to calculate the lux :

Which is an absolute quantity to determine the amount of light in a location.

A lux meter is a standard used to measure ambient lighting in day to day use like in cricket matches or for building design.

When a photograph is captured using a phone, the metadata of the pictures saves the following among several other parameters :

ISO : It is the measure of the sensitivity of the sensor capturing the image
Shutter Speed : This measures how quickly the shutter clicked and thus time the sensor was exposed to the light. Less the shutter speed, more bright the photo would be
Aperture (or f-stop) : This defines the size of the aperture open for capturing a picture. More the aperture, more the exposure
Brightness compensation : The adjustment made by the software to make a photo look better in terms of brighness

All these can be used to determine an absolute quantity called lux, using the following formula

lux= f-stop² / (ISO x Shutter Speed)

Using the lux constant of a phone we then adjust the brightness compensation as :

lux=(f-stop²/ ( ISO x ShutterSpeed)) - (BrightnessCompensation / constant)

Proof of concept

Lux = 0.029

Lux = 0.028

The Lux value gives the absolute brightness of a region. Here we can see two images of the same spot having almost the same lux value. When this happens, it implies that both images have similar lighting irrespective of the camera or where the user focused while clicking the photo.

Thus, we came up with a particular safety rating scale corresponding to the lux values of each image.
The safety ratings are integer values from 1 - 3, with 1 being least lit and thus least safe.

A lux value between 0 - 0.002 would have a safety rating of 1
A lux value between 0.002 - 0.004 would have a safety rating of 2
A lux value between 0.004 - 0.006 would have a safety rating of 3

Contains all the points plotted from Gachibowli and Manikonda. The red circles signify a lower safety rating compared to the orange ones.

A heat map depicting the intensity and safety of the areas that are plotted using the application.

Ready to Use

We started out with the objective to create an application service that is fully functional and allows the users to make use of the features we created to help our bit towards safety and general policing.

The application is accessible on desktop and mobile web browsers.

Scan the QR code to get the link