Unlock the true potential of the BME680.

Automising ML training for scent identification with MOX sensor, complete with convenient GUI for streamlined data collection and training.

Abstract

Metal Oxide Sensors (MOX sensors) have become popular in recent years for their versatility and low cost, for monitoring temperature, humidity, pressure, and VOCs present in the environment in real time. However, most use cases are limited to classifying few chemical compounds with laboriously hand crafted features. Our collaborative project with OWidgets Smell Digital introduces a complete package, forgoing the resource intensive process and narrow use cases for a streamlined and automised process for data synthesis, feature extraction, training, and deployment. Using OWidget's proprietary olfactometer for scent emission, we complement our client's low cost and customisable olfactory experience by introducing simplified process that lays the foundational work for quantifying and analysing unique scent experiences that is suited for our client's various pursuits in fragrance design, healthcare, environmental monitoring, and immersive VR experiences.

Automated Data Synthesis and Feature Extraction

Streamlining the data synthesis and feature extraction process for further analysis of any olfactory experience.

Refined User Interface

One destination for a straightforward and navigable access to view current and historical readings, setup and configure the gas sensor and olfactometer, train classification algorithms, and view database for all previously identified scents database.

Scent Database

Providing a digital library of olfactory experiences for further analysis and quality control.

Real Time Identification

Facilitate prediction and identification of scents present in the environment using machine learning algorithms trained specifically for the use case.

Featuring an easy to use and polished UI

View current and historical readings, set up and configure the gas sensor and olfactometer, and view all previously identified scents database all in one.

Demo Video

The Team

Charlotte Yeo

Second Year Computer Science Student at UCL.
Frontend and database developer.

Divyanshu Verma

Second Year Computer Science Student at UCL.
Frontend developer, system integration.

Stanley Hsu

Second Year Computer Science Student at UCL.
Data synthesis system developer, data analyst, ML developer.

Project Timeline