Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 26, 2026

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
07:34

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

A Deep Learning Approach for Pixel-Level Material Classification via Hyperspectral Imaging.

Savvas Sifnaios1,2, George Arvanitakis3, Fotios K Konstantinidis1

  • 1Institute of Communication and Computer Systems, National Technical University of Athens, 9 Iroon Polytechniou Str., GR-157 73 Athens, Greece.

Journal of Imaging
|June 25, 2026
PubMed
Summary

Related Concept Videos

Classification of Signals01:30

Classification of Signals

In signal processing, signals are classified based on various characteristics: continuous-time versus discrete-time, periodic versus aperiodic, analog versus digital, and causal versus noncausal. Each category highlights distinct properties crucial for understanding and manipulating signals.
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Multi-Area, Multi-Service and Multi-Tier Edge-Cloud Continuum Planning.

Sensors (Basel, Switzerland)·2025
Same author

Wood Waste Valorization and Classification Approaches: A systematic review.

Open research Europe·2025
Same author

Data Analytics to Support Policy Making for Noncommunicable Diseases: Scoping Review.

Online journal of public health informatics·2024
Same author

A Robust End-to-End IoT System for Supporting Workers in Mining Industries.

Sensors (Basel, Switzerland)·2024
Same author

Correction: Gligoric et al. IOTA-Based Distributed Ledger in the Mining Industry: Efficiency, Sustainability and Transparency. <i>Sensors</i> 2024, <i>24</i>, 923.

Sensors (Basel, Switzerland)·2024
Same author

IOTA-Based Distributed Ledger in the Mining Industry: Efficiency, Sustainability and Transparency.

Sensors (Basel, Switzerland)·2024
This summary is machine-generated.

Hyperspectral imaging combined with deep learning accurately classifies plastic materials. This approach offers a significant advancement for industries requiring detailed material identification beyond visual cues.

Area of Science:

  • Computer Vision
  • Material Science
  • Spectroscopy

Background:

  • RGB imaging limitations for material characterization.
  • Hyperspectral (HS) imaging captures spatial and spectral data for enhanced analysis.
  • Need for advanced classification in waste sorting, pharmaceuticals, and defence.

Purpose of the Study:

  • Evaluate combining HS imaging with deep learning for plastic material classification.
  • Develop a robust and efficient classification model.
  • Address limitations of current computer vision systems.

Main Methods:

  • Designed an experimental setup with HS line-scan camera, conveyor, and controlled illumination.
  • Created an object-disjoint dataset of HDPE, PET, PP, and PS samples.
Keywords:
deep learninghyperspectral imagingmaterial classificationpixel-level classificationreal-time object detection

Related Experiment Videos

Last Updated: Jun 26, 2026

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals
07:34

Excitation-Scanning Hyperspectral Imaging Microscopy to Efficiently Discriminate Fluorescence Signals

Published on: August 22, 2019

  • Developed P1CH, a lightweight pixel-wise 1D convolutional hyperspectral classifier.
  • Utilized Raman spectroscopy for sample labelling.

    • Main Results:

    • Achieved 97.44% all-pixel accuracy on object-disjoint test images with P1CH.
    • Reached 99.94% accuracy in boundary sensitivity analysis after excluding border pixels.
    • Demonstrated effectiveness for small fragments, visually similar, and overlapping plastics.

    Conclusions:

    • Pixel-wise spectral analysis using deep learning is effective for plastic classification.
    • The P1CH model shows high accuracy and robustness.
    • Challenges remain for classifying black or very dark plastics with current configurations.