Comparing machine learning and neural network-based approaches for sign detection and classification in autonomous vehicles

Authors

Sphurti More, Michigan Technological University
Jeremy Bos, Michigan Technological UniversityFollow

Document Type

Conference Proceeding

Publication Date

1-1-2020

Department

Department of Electrical and Computer Engineering

Abstract

We compare two algorithms, Histogram of Oriented Gradient (HOG) with linear Support Vector Machine (SVM) and You Look Only Once (YOLO), to the task of sign detection and classification from imagery from the LISA dataset. Comparisons are made in terms of execution time, accuracy, and readiness for use on GPU or FPGA hardware for acceleration. We find the neural network-based approaches like YOLO have superior accuracy but run slower on general purpose CPUs without acceleration. On the other hand, while less accurate the SVM-based are faster without acceleration.

Publisher's Statement

© 2020 SPIE. Publisher’s version of record: https://doi.org/10.1117/12.2558966

Publication Title

Proceedings of SPIE - The International Society for Optical Engineering

Recommended Citation

More, S., & Bos, J. (2020). Comparing machine learning and neural network-based approaches for sign detection and classification in autonomous vehicles. Proceedings of SPIE - The International Society for Optical Engineering, 11415. http://doi.org/10.1117/12.2558966
Retrieved from: https://digitalcommons.mtu.edu/michigantech-p/2734