Smart Fridge Add-on

2019 ELE Engineering Design Project (NMK04)

---

Faculty Lab Coordinator

Naimul Mefraz Khan

Topic Category

Signal Processing / Communication

Preamble

The rise of internet of things and machine learning provides us with the opportunity to even make existing fridges "smarter" through integration of simple add-ons with intelligent cloud processing. Through simple image analysis, we can identify fridge inventory and assist the customer in making grocery lists/coming up with recipes through the convenience of their mobile phones.

Objective

The objective of this project is to develop an add-on package for existing fridges through an easy to install hardware rig coupled with an intelligent cloud-based analysis software. The add-on will be able to turn any fridge into a "smart" one through intelligent image analysis to identify the ingredients inside it.

Partial Specifications

1. The hardware package should consist of a micro computer coupled with a camera module. Additional sensors can be added if image analysis becomes difficult (e.g. pressure sensor, counter), however, the rig should be kept as simple as possible for easy installation.

2. Images taken inside the fridge will be sent to a cloud-based API for object identification so that a list of ingredients can be prepared.

3. The add-on should come with an accompanying app, where the user can query simple things (e.g. "What's in my fridge?") through voice control by Google Assistant/Amazon Alexa/similar APIs.

Suggested Approach

1. Raspberry Pi/NVIDIA Jetson with camera module. Additional sensors only if necessary.

2. Google vision/Amazon AWS for object identification. This is the key component of the project, so custom object recognition library should be developed if necessary.

3. Google Assistant/ Amazon Alexa for voice processing.

Group Responsibilities

1. Study literature on existing approaches for object recognition through image analysis.
2. Identify best approach (develop custom approach if necessary).
2. Study required SDKs to develop the system.
3. end-to-end development for a demonstrable product.
4. Test the final prototype for robustness (sensitivity to angles, illumination, speed of performance, accuracy).

Student A Responsibilities

Hardware rig development

Student B Responsibilities

vision algorithm for object recognition

Student C Responsibilities

vision algorithm for object recognition

Student D Responsibilities

app development and voice command processing

Course Co-requisites