arrow
Return

Intelligent drone-based framework for autonomous inventory data inspection

delete2026-05-01
delete0
PRE
AI
S
Singkhamfu, Phudinan
P
Phasit Charoenkwan
T
Teerawat Kamnardsiri
N
Noamna, Somkeit
M
Malang, Chommaphat
R
Ratapol Wudhikarn *
DOI:10.1016/j.iswa.2026.200648delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
This study introduces an innovative drone-based framework for autonomous inventory inspection, designed to address the inherent inefficiencies and limitations of conventional inventory management practices. The proposed system uniquely combines indoor drone navigation with advanced deep learning-based barcode detection and decoding, further enhanced by the creation of a manually annotated barcode dataset acquired directly from drone imagery in warehouse environments. In contrast to previous studies, which have typically examined these technological components in isolation, this research presents a comprehensive and unified approach that facilitates efficient, contactless, and dependable inventory data collection. State-of-the-art deep learning models, specifically EfficientDet, Faster R-CNN, and YOLO, are utilized to accurately identify and decode both onedimensional and two-dimensional barcodes, thereby streamlining inventory identification and data acquisition processes. Experimental evaluations conducted under controlled warehouse conditions have demonstrated robust performance, achieving barcode detection rates of up to 100 percent, with a minimum of 74.1 percent, and decoding accuracy ranging from 81.3 percent to 98.6 percent. The primary contributions of this work include: (1) the development of the first unified framework that integrates indoor drone navigation with deep learning (DL)-based barcode reading in real-world settings, and (2) the creation of a new annotated barcode dataset tailored for warehouse environments. These results highlight the significant potential of this approach to transform inventory management practices and enable scalable, efficient, and autonomous operations for industrial and commercial applications.
Keywords:
Intelligence framework
Drone
Unmanned aerial vehicle (UAV)
Inventory inspection
Inventory management
Autonomous inspection

Journal

I
Intelligent Systems with Applications
IF:
4.3
Papers:
90
Citations:
0

Organization

C
Chiang Mai University
Scholars:
1.5W
Papers: 9.3K
Citations: 7.9K