Related Concept Videos
Vision
Light Acquisition
Imaging Biological Samples with Optical Microscopy
In optical microscopy, the specimen to be viewed is placed on a glass slide and clipped on the stage...
Visual System
Once through the pupil, the light passes through the lens, a...
Infrared (IR) Spectroscopy: Overview
Different compounds display unique properties due to their...
Machines
A free-body diagram of the...
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Correction: He et al. An Edge-Computing-Based Emotion-Aware Adaptive Lighting System for Intelligent Cockpits. <i>Sensors</i> 2026, <i>26</i>, 3489.
Related Experiment Video
Updated: Jan 9, 2026

Simulation of a Scaled Assembly Process with Collaboration of a Robotic Arm and Monitoring through a Vision System for Quality Control
Published on: August 29, 2025
A Framework for Integration of Machine Vision with IoT Sensing.
Gift Nwatuzie1, Hassan Peyravi1
1Department of Computer Science, Kent State University, Kent, OH 44240, USA.
This study presents a unified edge-cloud framework integrating cameras into IoT networks for environmental monitoring. It achieves robust, context-aware sensing through synchronized data fusion and efficient cross-modal learning.
Area of Science:
- Computer Science
- Environmental Science
- Electrical Engineering
Background:
- Automated monitoring systems often struggle to integrate diverse sensing sources like IoT sensors and machine vision.
- Existing multimodal fusion frameworks lack tight synchronization and efficient cross-modal learning, limiting coherent environmental interpretation.
- Cameras and IoT sensors offer complementary data (spatial context vs. point measurements) but are often operated independently.
Purpose of the Study:
- To introduce a unified edge-cloud framework that deeply integrates cameras as active sensing nodes within an IoT network.
- To enable tight time synchronization between visual and IoT data streams for enhanced environmental interpretation.
- To facilitate efficient cross-modal learning and model training on resource-constrained edge devices.
Main Methods:
- Developed a unified edge-cloud framework with tight time synchronization between visual and IoT data streams.
- Employed cross-modal knowledge distillation for efficient model training on edge devices.
- Utilized a multi-task learning setup with dynamically adjusted loss weighting, incorporating EfficientNet, Vision Transformers, and U-Net derivatives.
Main Results:
- Achieved 94.8% classification accuracy and 87.6% segmentation quality (mIoU) on environmental monitoring tasks.
- Sustained sub-second inference latency on compact edge hardware (Jetson Nano, Coral TPU).
- Demonstrated the framework's robustness in classification, segmentation, and anomaly detection.
Conclusions:
- The proposed synchronized, knowledge-driven fusion framework provides a more adaptive, context-aware, and deployment-ready sensing solution.
- Significantly advances the practical integration of machine vision within Internet of Things (IoT) ecosystems.
- Highlights the effectiveness of deep integration and efficient cross-modal learning for environmental monitoring.
