Hazard Detection in Supermarkets using Deep Learning on the Edge

TL;DR

This paper introduces EdgeLite, a lightweight deep learning model designed for real-time hazard detection on supermarket floors using resource-constrained edge devices, improving accuracy over existing models.

Contribution

The paper presents EdgeLite, a novel, efficient deep learning model optimized for deployment on edge devices for hazard detection in supermarkets.

Findings

EdgeLite outperforms six state-of-the-art models in accuracy.

EdgeLite maintains comparable memory and inference time.

Effective hazard detection on resource-limited devices.

Abstract

Supermarkets need to ensure clean and safe environments for both shoppers and employees. Slips, trips, and falls can result in injuries that have a physical as well as financial cost. Timely detection of hazardous conditions such as spilled liquids or fallen items on supermarket floors can reduce the chances of serious injuries. This paper presents EdgeLite, a novel, lightweight deep learning model for easy deployment and inference on resource-constrained devices. We describe the use of EdgeLite on two edge devices for detecting supermarket floor hazards. On a hazard detection dataset that we developed, EdgeLite, when deployed on edge devices, outperformed six state-of-the-art object detection models in terms of accuracy while having comparable memory usage and inference time.