Impact of Image Blur on Classification and Augmentation of Deep Convolutional Networks

0571255 - ÚTIA 2024 RIV CH eng C - Konferenční příspěvek (zahraniční konf.)
Lébl, Matěj - Šroubek, Filip - Flusser, Jan
Impact of Image Blur on Classification and Augmentation of Deep Convolutional Networks.
Image Analysis: 23rd Scandinavian Conference, SCIA 2023. Cham: Springer, 2023 - (Gade, R.), s. 108-117. Lecture notes on computer science, LNCS 13886. ISBN 978-3-031-31437-7.
[Scandinavian Conference on Image Analysis 2023 /23./. Levi (FI), 18.04.2023-21.04.2023]
Grant CEP: GA ČR GA21-03921S
Institucionální podpora: RVO:67985556
Klíčová slova: Image recognition * Blur * Augmentation of the training set * Convolutional neural network
Obor OECD: Computer hardware and architecture
http://library.utia.cas.cz/separaty/2023/ZOI/lebl-0571255.pdf

Blur is a common phenomenon in image acquisition that negatively influences the recognition rate of most classifiers. This paper studies the influence of image blurring of various types and sizes on the recognition rate achieved by a deep convolutional network. We confirm that the blur significantly decreases the performance if the network has been trained on clear images only. When the training set is augmented with blurred samples, the recognition rate becomes sufficiently high even if the blur in query images is of different size than the blur used for training. However, this is mostly not true if query images contain blur of a different type from the one used for training.
Trvalý link: https://hdl.handle.net/11104/0342934