Classical Algorithm vs. Machine Learning in Objects Recognition

Jakub Czygier , Piotr Tomaszuk , Aneta Łukowska , Paweł Straszyński , Kazimierz Dzierżek


This article focuses on two most popular methods of detecting regular shapes in the pictures. Over the past years, image processing and object recognition are entering our lives more and more. The difference between a classical algorithm and machine learning was analyzed, in case of a tennis ball recognition on photos. It has been created on own dataset to avoid a uniform background and bright colors. Images were taken with a low-cost camera in different conditions. Creating the classical algorithm and machine learning and comparing the accuracy, false flags and performance of the two methods are described.
Author Jakub Czygier (FME)
Jakub Czygier,,
- Faculty of Mechanical Engineering
, Piotr Tomaszuk (FME)
Piotr Tomaszuk,,
- Faculty of Mechanical Engineering
, Aneta Łukowska (FME)
Aneta Łukowska,,
- Faculty of Mechanical Engineering
, Paweł Straszyński (FME)
Paweł Straszyński,,
- Faculty of Mechanical Engineering
, Kazimierz Dzierżek (FME / DACR)
Kazimierz Dzierżek,,
- Department of Automatic Control and Robotics
Publication size in sheets0.55
Book Kohei Arai, Kapoor Supryia (eds.): Advances in Computer Vision: Proceedings of the 2019 Computer Vision Conference (CVC). Vol. 2, Advances in Intelligent Systems and Computing, vol. 944, 2020, Springer, ISBN 978-3-030-17797-3, [978-3-030-17798-0], 766 p., DOI:10.1007/978-3-030-17798-0
Keywords in Englishobject recognition, machine learning, image processing, classification, neural network
Internal identifierROC 19-20
Languageen angielski
Score (nominal)20
Score sourcepublisherList
ScoreMinisterial score = 20.0, 17-01-2020, MonographChapterAuthor
Citation count*
Share Share

Get link to the record

* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Are you sure?