AI Case Study
Students at the Indian Institute of Technology Ropar aim to eliminate garbage in the streets of India using deep learning
Students Authors: Gaurav Mittal, Kaushal B. Yagnik, Mohit Garg and Narayanan C. Krishnan at the Indian Institute of Technology Ropar have developed an app to combat a problem in their every day life in India, that of garbage in public spaces. The app, SpotGarbage, uses deep architecture of fully convolutional networks to detect and segment garbage regions in a user-clicked geo-tagged image. The aim is to engage citizens to track and report garbage in their local communities. The app send this information to the local municipality and works out the optimal route for garbage delivery. According to the paper, the model achieved a mean accuracy of 87.69%.
Industry
Public And Social Sector
Public Services
Project Overview
Gaurav Mittal and his classmates at IIT Ropar in India are the creators of the garbage recognition app. Photographs taken through the Spot Garbage app are submitted to the municipality for review. AI to detect segment and geotag garbage. It then uploads this information to the cloud and provide the authorities with the most efficient route for garbage collection.
"With the aim of engaging citizens to track and report on their neighborhoods, this paper presents a novel smartphone app, called SpotGarbage, which detects and coarsely segments garbage regions in a user-clicked geo-tagged image. The app utilizes the proposed deep architecture of fully convolutional networks for detecting garbage in images. The model has been trained on a newly introduced Garbage In Images (GINI) dataset, achieving a mean accuracy of 87.69%. The paper also proposes optimizations in the network architecture resulting in a reduction of 87.9% in memory usage and 96.8% in prediction time with no loss in accuracy, facilitating its usage in resource constrained smartphones" (paper)
This project won a prize at the Microsoft challenge in India last year.
Function
Background
"Total amount of garbage in the world is around 3.7m tones, by 2025 this will rise to more than 6.1m tones", explains Gaurav Mittal.
"Maintaining a clean and hygienic civic environment is an indispensable yet formidable task, especially in developing countries." (paper)
Reported Results
The model has achieved a mean accuracy of 87.69%.
Benefits
Technology
Data
real-time images