A Lean Convolutional Neural Network for Vehicle Classification

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Authors

  • Jonathan J. Sanchez-Castro
  • Julio C. Rodriguez-Quinonez
  • Luis R. Ramirez-Hernandez
  • Guillermo Galaviz
  • Daniel Hernandez-Balbuena
  • Gabriel Trujillo-Hernandez
  • Wendy Flores-Fuentes
  • Paolo Mercorelli
  • Wilmar Hernandez-Perdomo
  • Oleg Sergiyenko
  • Felix Fernando Gonzalez-Navarro

Image classification is an important task in machine vision, in which vehicle classification is used for different applications like traffic analysis, autonomous driving, security, among others. Recent studies made with Convolutional Neural Networks (CNN) have shown that these networks have surpassed older algorithms like Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) in terms of accuracy, speed, and resources management. Even though that CNN have better accuracy and speed they still are heavy in resource consumption on computers which makes them not suitable to deploy on an embedded platform. This paper proposes a lean CNN that has a smaller number of parameters and still maintaining the best accuracy possible on vehicle classification.

Original languageEnglish
Title of host publication2020 IEEE 29th International Symposium on Industrial Electronics (ISIE) : 17 - 19 June, 2020, Delft, Netherlands, Proceedings
Number of pages5
Place of PublicationPiscataway
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Publication date01.06.2020
Pages1365-1369
Article number9152274
ISBN (print)978-1-7281-5636-1
ISBN (electronic)978-1-7281-5635-4, 978-1-7281-5634-7
DOIs
Publication statusPublished - 01.06.2020
Event29th IEEE International Symposium on Industrial Electronics, ISIE 2020 - TU Delft , Delft, Netherlands
Duration: 17.06.202019.06.2020
Conference number: 29
http://isie2020.org/

    Research areas

  • Artificial Intelligence, CNN, Vehicle Classification
  • Engineering

Recently viewed

Researchers

  1. Martin Hailer

Publications

  1. Explaining and controlling for the psychometric properties of computer-generated figural matrix items
  2. Resolving the Complexity-Flexibility Dilemma in Multi-Issue Negotiations: Nested Bracketing as a Strategy to Enhance Negotiation Outcomes
  3. A Playful Approach to Interactive Media in the Foreign Language Classroom
  4. Towards productive functions?
  5. Creep behavior of AE42 based hybrid composites
  6. Sliding Mode Control of an Inductive Power Transmission System with Maximum Efficiency
  7. Modeling the distribution of white spruce (Picea glauca) for Alaska with high accuracy: an open access role-model for predicting tree species in last remaining wilderness areas
  8. Modeling and simulation of the microstructural behaviour in thermal sprayed coatings
  9. Photodegradation of micropollutants using V-UV/UV-C processes
  10. Toward Automatically Labeling Situations in Soccer
  11. Exploring the implications of the value concept for performance assessment of sustainable business models
  12. Rapid Prototyping of a Mechatronic Engine Valve Controller for IC Engines
  13. Some surprising differences between novice and expert errors in computerized office work
  14. Framework for empirical research on science teaching and learning
  15. Teaching content and language in the multilingual classroom
  16. Global finite-time stabilization of a class of perturbed planar systems with actuator saturation and disturbances
  17. Disassembly and reassembly
  18. The importance of product lifetime labelling for purchase decisions
  19. Active suspensions decoupling by algebraic feedback
  20. Effectiveness and Moderators of an Internet-Based Mobile-Supported Stress Management Intervention as a Universal Prevention Approach
  21. Mathematical Modelling of molecular adsorption in zeolite coated frequency domain sensors
  22. Finding the Best Match — a Case Study on the (Text‑) Feature and Model Choice in Digital Mental Health Interventions
  23. An Integrative and Comprehensive Methodology for Studying Aesthetic Experience in the Field
  24. Individual differences and cognitive load theory