Code for: An IoT-Integrated YOLOv8 Framework with Adaptive Domain Alignment for Real-Time Quality Inspection in Smart Logistics

Outdated Version: This repository contains the implementation accompanying the manuscript "An IoT-Integrated YOLOv8 Framework with Adaptive Domain Alignment for Real-Time Quality Inspection in Smart Logistics".

The repository includes:

• Modified YOLOv8 detector with enhanced micro-defect sensitivity.

• Lightweight style-transfer domain adaptation module.

• MQTT-based real-time defect alerting framework.

• Adaptive feedback and incremental retraining policy.

• Dataset conversion scripts for deriving YOLO-format bounding boxes from anomaly masks.

• Evaluation scripts for detection and IoT-performance metrics.

• Reproducibility instructions for the experiments reported in the manuscript.

The implementation supports experiments on MVTec AD and VisA datasets and demonstrates the integration of computer vision and Internet-of-Things technologies for Industry 4.0 smart logistics applications.

Note: Raw real-world logistics validation images are not included because they may contain commercially sensitive warehouse information. The repository provides code, configuration files, annotation guidelines, derived log templates, and instructions for reproducing the experiments using publicly available datasets.

Keywords:

YOLOv8
Smart Logistics
Industrial Inspection
Defect Detection
Domain Adaptation
Internet of Things
MQTT
MVTec AD
VisA
Industry 4.0
Computer Vision