Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Can Deep Learning Trigger Alerts from Mobile-Captured Images?

Pritisha Sarkar, Duranta Durbaar Vishal Saha, Mousumi Saha

TL;DR

The paper addresses the challenge of real-time, image-based air quality assessment by introducing HealthCamCNN, a regression-oriented CNN that predicts multiple pollutants from mobile camera images and delivers a user-centric dashboard with health-guided location recommendations. A core finding is that a two-stage, branched CNN architecture achieves low $MSE$ values ($0.0077$ for $2$ pollutants and $0.0112$ for $5$ pollutants) across multiple cities, while data augmentation provides minimal improvements in final accuracy. The study also demonstrates that augmenting images through vertical splits and mirroring does not significantly alter predictive performance, supporting dataset efficiency. Overall, the work offers a cost-effective, mobile-centric approach to environmental health monitoring and personalized decision support, with a deployable real-time dashboard for end users.

Abstract

Our research presents a comprehensive approach to leveraging mobile camera image data for real-time air quality assessment and recommendation. We develop a regression-based Convolutional Neural Network model and tailor it explicitly for air quality prediction by exploiting the inherent relationship between output parameters. As a result, the Mean Squared Error of 0.0077 and 0.0112 obtained for 2 and 5 pollutants respectively outperforms existing models. Furthermore, we aim to verify the common practice of augmenting the original dataset with a view to introducing more variation in the training phase. It is one of our most significant contributions that our experimental results demonstrate minimal accuracy differences between the original and augmented datasets. Finally, a real-time, user-friendly dashboard is implemented which dynamically displays the Air Quality Index and pollutant values derived from captured mobile camera images. Users' health conditions are considered to recommend whether a location is suitable based on current air quality metrics. Overall, this research contributes to verification of data augmentation techniques, CNN-based regression modelling for air quality prediction, and user-centric air quality monitoring through mobile technology. The proposed system offers practical solutions for individuals to make informed environmental health and well-being decisions.

Can Deep Learning Trigger Alerts from Mobile-Captured Images?

TL;DR

The paper addresses the challenge of real-time, image-based air quality assessment by introducing HealthCamCNN, a regression-oriented CNN that predicts multiple pollutants from mobile camera images and delivers a user-centric dashboard with health-guided location recommendations. A core finding is that a two-stage, branched CNN architecture achieves low values ( for pollutants and for pollutants) across multiple cities, while data augmentation provides minimal improvements in final accuracy. The study also demonstrates that augmenting images through vertical splits and mirroring does not significantly alter predictive performance, supporting dataset efficiency. Overall, the work offers a cost-effective, mobile-centric approach to environmental health monitoring and personalized decision support, with a deployable real-time dashboard for end users.

Abstract

Our research presents a comprehensive approach to leveraging mobile camera image data for real-time air quality assessment and recommendation. We develop a regression-based Convolutional Neural Network model and tailor it explicitly for air quality prediction by exploiting the inherent relationship between output parameters. As a result, the Mean Squared Error of 0.0077 and 0.0112 obtained for 2 and 5 pollutants respectively outperforms existing models. Furthermore, we aim to verify the common practice of augmenting the original dataset with a view to introducing more variation in the training phase. It is one of our most significant contributions that our experimental results demonstrate minimal accuracy differences between the original and augmented datasets. Finally, a real-time, user-friendly dashboard is implemented which dynamically displays the Air Quality Index and pollutant values derived from captured mobile camera images. Users' health conditions are considered to recommend whether a location is suitable based on current air quality metrics. Overall, this research contributes to verification of data augmentation techniques, CNN-based regression modelling for air quality prediction, and user-centric air quality monitoring through mobile technology. The proposed system offers practical solutions for individuals to make informed environmental health and well-being decisions.
Paper Structure (11 sections, 3 figures, 6 tables, 2 algorithms)

This paper contains 11 sections, 3 figures, 6 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Literature Survey
  3. System Overview
  4. Data Description
  5. Data Augmentation
  6. Methodology
  7. Steps of our proposed method :
  8. Results and Discussion
  9. Data augmentation results
  10. HealthCamCNN: A Real-Time Health Recommendation System
  11. Conclusion

Figures (3)

  • Figure 1: Workflow of our proposed approach
  • Figure 2: Data augmentation visualization : A Tamil Nadu B Delhi; $A_{1}$ and $A_{2}$ - followed the segment 1 method, $B_{1}$ and $B_{2}$ - followed the segment 2 method, $C_{1}$ and $C_{2}$ - followed the segment 3 method,
  • Figure 3: Some snapshot of real-time health-related symptoms used to predict AQI in our HealthCamCNN dashboard