, , , ,

Women’s Safety and Air Pollution: Top Issues for Indian Undergrads

By Aakanksha Chowdhery

We are proud to complete the third year of the Marconi Society’s Celestini Program in India.  

The Celestini Program is a flagship effort in the Marconi Society’s mission to celebrate, inspire and connect individuals building tomorrow’s technologies in service of a digitally inclusive world. 

The Celestini Program’s objective is to inspire and connect the next generation of networking and communications innovators in developing countries by empowering undergraduate students to create social and economic transformation in their communities through information and communications technology. The Marconi Society and our Young Scholars achieve this by selecting universities with promising telecommunications and engineering undergrads and providing them with support and mentorship to help tap their students’ true potential.

2019 is the third successful year of the Celestini program in India (projects from previous years are showcased here). 

More than three hundred students from IIT Delhi applied to participate in the program this year and five student teams (comprising fifteen students) were selected to compete for prizes funded by the Marconi Society. The students chose problem statements related to women’s safety and air pollution and designed smartphone applications leveraging machine learning.

On the theme of women’s safety, two student teams designed solutions to detect potential danger triggers in the audio from the user’s smartphone. On the theme of air pollution, one team built privacy-aware smartphone application that measures air quality using smartphone camera and machine learning, while two other teams focused on understanding the sources of air pollution, for example, investigating the effect of vehicular traffic on the air quality. 

Making Women Safer

India is considered to be the most dangerous country in the world for women, according to the Thompson Reuters Foundation’s 2018 Annual Poll.  Women rightfully feel unsafe and two teams chose to focus on this issue.

The award-winning student team, of Aniket Sharma,  Subham Banga, Piyush Agrawal and Ujjwal Upadhyay, designed a smartphone application called Rakshak that detects speech commands via the audio microphone of the user’s smartphone. When the application  detects audio snippets with speech commands requesting help or saying “stop” in distressed tones, it generates SOS alerts –  along with the location of the user – and sends them to emergency contacts specified by the user. The app is now available on Google Play store in beta release.

During the prototyping phase, the students started with publicly available speech command datasets, such as the Google Speech command dataset, then added speech commands specific to the scenario of women’s safety. They crowd-sourced additional data and open-sourced it as the Indian EmoSpeech Command dataset. This enabled them to detect emotion, background noise, and Indian accents in the audio with improved precision. 

Complementary to using speech commands such as `help` or `bachao` in smartphone audio, another student team used an alternate approach where they train a machine learning classifier to identify signals of distress in the streaming audio to the microphone of the user’s smartphone.

Measuring Air Quality

According to World Health Organization Global Ambient Air Quality Database, India has 14 out of the 15 most polluted cities in the world in terms of PM 2.5 concentrations. One approach to solving the air pollution issues is enabling users to conduct fine-grained measurements of the air quality that affects them and understand the sources of pollution so that they can take preventive measures.

India’s Central Pollution Control Board (CPCB) makes the air quality concentrations of pollutants such as CO, CO2, and NO2 available for a dozen locations within Delhi. Government-operated air quality monitoring stations provide accurate, but average, metrics on pollution exposure experienced by the majority of the population due to the complex spatial heterogeneity of pollutants. These monitoring stations are mainly used for city planning and policy decision-making.

To make air quality measurements available at the fine-grained scale of every location in Delhi over a large area of 1484 square kilometers, the teams needed crowdsourced measurements collected by Delhi residents (16.8 million). In addition, they needed to understand the correlation of these measurements with potential causes that generate the pollutants.

The team winning the second prize, of Shivam Grover, Shivani Jindal, Harshita Diddee and Divyanshu Sharma, built a privacy-aware smartphone application called VisionAir which uses photos of the horizon taken from a smartphone to measure the air quality.

Work by last year’s participants showed that a machine learning model can be built to estimate air quality from an image by extracting image features such as transmission index or haziness and combining them with meteorological data and historical air quality data. The innovative aspect is to leverage federated learning to train the machine learning model in a privacy-aware manner instead of uploading photos from each user as shown in this tutorial. Federated learning only uploads the features extracted from the images without uploading the smartphone images to train the machine learning model.

The VisionAir team, winning the second prize, also enable other developers to develop new machine learning models by open-sourcing a diverse dataset of  smartphone images taken across several locations in Delhi,  from different phones with ground truth air quality data from Central Pollution Control Board monitors and Airveda sensors. 

In addition to tracking air pollution using smartphone cameras, two other student teams focused on understanding the causes of air pollution. One of the team formulated this as the problem of personalizing air quality measurement forecasts by fusing air quality measurements from multiple CPCB stations based on user’s location to emphasize the major pollutants and enable users to contribute additional measurements using wearable sensors.

A second team conjectured that Delhi’s vehicular traffic is a major cause of the air pollution and investigated its effect on the concentration of pollutants such as NO2 in different locations at peak and off-peak traffic hours over different seasons.

Award ceremony

The concluding ceremony of this year was held on October 21 at IIT Delhi where alumnus and business leader Padmasree Warrior gave an inaugural address. IIT Delhi’s Dean Alumni Affairs and International Programmes, several faculty members from Electrical Engineering and Computer Science, as well as industry partners attended the ceremony. The first prize-winning team, Rakshak, was awarded a cash prize of $1500 and the second prize-winning team, VisionAir, was awarded a cash prize of $500. 

The Celestini Program is designed to bring hands-on learning and critical decision-making skills to participating students and to inspire these students to pursue STEM-related studies to solve the problems they see in their communities. The participating students highly valued the experience of developing prototypes that solve real-world problems with the potential to improve the well-being of their communities. We have already had the privilege to work with outstanding students in the Celestini Program India and look forward to seeing the exceptional ways that they use STEM in their careers.