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 Concept Videos

Classification of Signals01:30

Classification of Signals

411
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...
411
Aggregates Classification01:29

Aggregates Classification

305
Aggregate classification is generally based on its size, petrographic characteristics, weight, and source. Size classification ranges from coarse to fine aggregates, defined by the size of the particles. Coarse aggregates are particles that do not pass through ASTM sieve No. 4, and aggregates that pass through the sieve are fine aggregates.
Petrographic classification groups aggregates based on common mineralogical characteristics. Some of the common mineral groups found in aggregates are...
305
Difference from Background: Limit of Detection01:05

Difference from Background: Limit of Detection

5.9K
The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...
5.9K
Classification of Systems-II01:31

Classification of Systems-II

136
Continuous-time systems have continuous input and output signals, with time measured continuously. These systems are generally defined by differential or algebraic equations. For instance, in an RC circuit, the relationship between input and output voltage is expressed through a differential equation derived from Ohm's law and the capacitor relation,
136
Classification of Systems-I01:26

Classification of Systems-I

176
Linearity is a system property characterized by a direct input-output relationship, combining homogeneity and additivity.
Homogeneity dictates that if an input x(t) is multiplied by a constant c, the output y(t) is multiplied by the same constant. Mathematically, this is expressed as:
176
Force Classification01:22

Force Classification

1.1K
Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
1.1K

You might also read

Related Articles

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

Sort by
Same author

Cellulose-Based Superabsorbent Hydrogel with Recyclable Moisture Regulation for Agriculture in Arid Regions.

Journal of agricultural and food chemistry·2026
Same author

Class prototype rectification and multi-scale feature measurement for few-shot classification of bearing surface defects.

Scientific reports·2026
Same author

Serum cystatin C levels are independently correlated with cognitive impairment in individuals with cerebral small vessel disease.

Frontiers in neuroscience·2026
Same author

Postoperative molecular residual disease status refines recurrence risk stratification beyond pathologic response after neoadjuvant chemoradiotherapy for esophageal squamous cell carcinoma.

The Journal of thoracic and cardiovascular surgery·2026
Same author

Resolution improvement in speckle wavemeters via mechanically induced multimode fiber mode-mixing method.

Optics express·2026
Same author

Robust Fluorine- and Silicon-Free Slippery Surface toward Marine Antibiofouling and Anticorrosion.

ACS applied materials & interfaces·2026

Related Experiment Video

Updated: Jun 10, 2025

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

657

Context-aware feature reconstruction for class-incremental anomaly detection and localization.

Jingxuan Pang1, Chunguang Li1

  • 1College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China.

Neural Networks : the Official Journal of the International Neural Network Society
|October 18, 2024
PubMed
Summary

This study introduces a new method for class-incremental anomaly detection and localization (CADL) using a context-aware feature reconstruction model. The approach effectively retains knowledge of previously learned classes when encountering new ones, crucial for industrial applications.

Keywords:
Anomaly detectionAnomaly localizationClass-incremental learningFeature reconstruction

More Related Videos

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

6.9K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.8K

Related Experiment Videos

Last Updated: Jun 10, 2025

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

657
Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

6.9K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.8K

Area of Science:

  • Computer Vision
  • Machine Learning
  • Deep Learning

Background:

  • Unsupervised visual anomaly detection and localization is crucial in industry.
  • Traditional methods train on all data simultaneously, unsuitable for incremental data availability.
  • Class-incremental anomaly detection and localization (CADL) is needed for evolving industrial product lines.

Purpose of the Study:

  • To develop a method for class-incremental anomaly detection and localization (CADL).
  • To effectively retain knowledge of old classes when learning new classes with limited exemplars.
  • To address the challenge of inter-class context conflict in incremental learning scenarios.

Main Methods:

  • Proposed a context-aware feature reconstruction (CFR) model to capture anomaly-identification knowledge from input context.
  • Designed an intermediate feature organization strategy to prevent context conflict across incremental classes.
  • Implemented dual constraints (feature organization and output-level knowledge distillation) to regularize the model.

Main Results:

  • The proposed CFRDC method effectively retains old-class knowledge while learning new classes.
  • Demonstrated outstanding performance on the MVTec-AD dataset for the CADL task.
  • The CFR model successfully captures context-aware knowledge for anomaly identification.

Conclusions:

  • The CFRDC method provides an effective solution for class-incremental anomaly detection and localization.
  • Leveraging context-aware features and dual constraints is key to successful incremental learning in anomaly detection.
  • The approach is well-suited for practical industrial scenarios with sequentially available training data.