Anomaly detection in formed sheet metals using convolutional autoencoders

Research output: Journal contributionsConference article in journalResearchpeer-review

Authors

Autoencoders are used to compress data and learn how to reliably reconstruct it. Through comparing the reconstruction error with a set threshold, they are able to detect anomalies in unseen datasets where the data does not quite match the reconstructed input samples. In this work, we attempt to investigate the use of convolutional autoencoders in the field of visual quality inspection, where images of formed sheet metals from a real production line are inspected for the occurrence of cracks and wrinkle formation. This approach tackles the problem of needing enough defective samples to attain reliable detection accuracies
Original languageEnglish
JournalProcedia CIRP
Volume93
Pages (from-to)1281-1285
Number of pages5
ISSN2212-8271
DOIs
Publication statusPublished - 01.01.2020
Event53rd CIRP Conference on Manufacturing Systems - CMS 2020 - Chicago, United States
Duration: 01.07.202003.07.2020
Conference number: 53

Bibliographical note

The authors gratefully acknowledge the financial support given by the Federal Ministry for Economic Affairs and Energy (BMWi) under the Central Innovation Programme for SMEs IM)(Zorfe thopjrectartSPm“”ress on the basis of a decision of the German Bundestag (grant number: ZF4084503GR7).

    Research areas

  • Engineering - Artifical intelligence, Deep autoencoders, Process monitoring, Surface quality inspection

Documents

DOI

Recently viewed

Publications

  1. Analysis of semi-open queueing networks using lost customers approximation with an application to robotic mobile fulfilment systems
  2. Evaluating the construct validity of Objective Personality Tests using a multitrait-multimethod-Multioccasion-(MTMM-MO)-approach
  3. Continuous 3D scanning mode using servomotors instead of stepping motors in dynamic laser triangulation
  4. Intersection tests for the cointegrating rank in dependent panel data
  5. Algebraic combinatorics in mathematical chemistry. Methods and algorithms. I. Permutation groups and coherent (cellular) algebras.
  6. A Wavelet Packet Tree Denoising Algorithm for Images of Atomic-Force Microscopy
  7. A New Framework for Production Planning and Control to Support the Positioning in Fields of Tension Created by Opposing Logistic Objectives
  8. Introducing parametric uncertainty into a nonlinear friction model
  9. Volume of Imbalance Container Prediction using Kalman Filter and Long Short-Term Memory
  10. Age effects on controlling tools with sensorimotor transformations
  11. Using protochirons for three-dimensional coding of certain chemical structures.
  12. Second language learners' performance in mathematics
  13. A discrete approximate solution for the asymptotic tracking problem in affine nonlinear systems
  14. Improving students’ science text comprehension through metacognitive self-regulation when applying learning strategies
  15. A guided simulated annealing search for solving the pick-up and delivery problem with time windows and capacity constraints
  16. Text Comprehension as a Mediator in Solving Mathematical Reality-Based Tasks
  17. Analysis and Implementation of a Resistance Temperature Estimator Based on Bi-Polynomial Least Squares Method and Discrete Kalman Filter
  18. Fixed-term Contracts and Wages Revisited Using Linked Employer-Employee Data from Germany
  19. Partitioned beta diversity patterns of plants across sharp and distinct boundaries of quartz habitat islands
  20. 'SPREAD THE APP, NOT THE VIRUS’ – AN EXTENSIVE SEM-APPROACH TO UNDERSTAND PANDEMIC TRACING APP USAGE IN GERMANY
  21. Distributed robust Gaussian Process regression
  22. Passive Peak Voltage Sensor for Multiple Sending Coils Inductive Power Transmission System
  23. Combining linked data and statistical information retrieval
  24. Inversion of fuzzy neural networks for the reduction of noise in the control loop
  25. Simulation based comparison of safety-stock calculation methods
  26. Using Wikipedia for Cross-Language Named Entity Recognition
  27. Control versus Complexity
  28. Selecting and Adapting Methods for Analysis and Design in Value-Sensitive Digital Social Innovation Projects: Toward Design Principles
  29. Convolutional Neural Networks
  30. Integrating the underlying structure of stochasticity into community ecology
  31. Cognitive load and instructionally supported learning with provided and learner-generated visualizations
  32. Globally asymptotic output feedback tracking of robot manipulators with actuator constraints