A Study on Tiny YOLO for Resource Constrained Xray Threat Detection

TL;DR

This paper evaluates Tiny YOLO variants, including QCFS and SNN versions, for efficient threat detection in X-ray images on edge devices, achieving state-of-the-art results.

Contribution

It introduces QCFS and SNN Tiny YOLO models tailored for resource-constrained environments, enhancing X-ray threat detection performance.

Findings

QCFS Tiny YOLO achieves state-of-the-art accuracy.

Spiking Tiny YOLO demonstrates promising behavior.

Models are suitable for edge device deployment.

Abstract

This paper implements and analyzes multiple networks with the goal of understanding their suitability for edge device applications such as X-ray threat detection. In this study, we use the state-of-the-art YOLO object detection model to solve this task of detecting threats in security baggage screening images. We designed and studied three models - Tiny YOLO, QCFS Tiny YOLO, and SNN Tiny YOLO. We utilize an alternative activation function calculated to have zero expected conversion error with the activation of a spiking activation function in our Tiny YOLOv7 model. This \textit{QCFS} version of the Tiny YOLO replicates the activation function from ultra-low latency and high-efficiency SNN architecture. It achieves state-of-the-art performance on CLCXray, an open-source X-ray threat Detection dataset. In addition, we also study the behavior of a Spiking Tiny YOLO on the same X-ray threat Detection dataset.