Pattern-Enhanced RT-DETR for Multi-Class Battery Detection

TL;DR

This paper benchmarks various CNN and transformer-based detectors for multi-class battery detection and introduces PaQ-RT-DETR, a pattern-based dynamic query method that improves detection accuracy with minimal overhead.

Contribution

It provides a comprehensive benchmark of detectors and proposes PaQ-RT-DETR, enhancing RT-DETR with pattern-based dynamic queries for better multi-class battery detection.

Findings

YOLOv8n is the fastest detector (~1,667 FPS).

YOLO11n achieves the highest CNN-based accuracy (mAP@50: 0.779).

PaQ-RT-DETR-X surpasses RT-DETR-X with 0.782 mAP@50.

Abstract

Accurate and efficient battery detection is increasingly important for applications in electronic waste recycling, industrial quality control, and automated sorting systems. In this paper, we present both a comprehensive benchmark and a novel method for multi-class battery detection. We systematically compare three CNN-based detectors (YOLOv8n, YOLOv8s, YOLO11n) and two transformer-based detectors (RT-DETR-L, RT-DETR-X) on a publicly available dataset of approximately 8,591 annotated images under identical experimental conditions, and further propose PaQ-RT-DETR, which introduces pattern-based dynamic query generation into RT-DETR to alleviate query activation imbalance with negligible computational overhead. Among baselines, YOLO11n achieves the best CNN-based accuracy (mAP@50: 0.779) at only 2.6M parameters, while YOLOv8n delivers the fastest inference at ~1,667 FPS. PaQ-RT-DETR-X achieves the highest overall mAP@50 of 0.782, surpassing RT-DETR-X by +2.8% with consistent per-class gains across all six battery categories including the data-scarce Bike Battery class. Our findings provide practical guidance for selecting object detection models in battery-related industrial applications.