Patch-based deep learning approaches for artefact detection of endoscopic images

Gao, Xiaohong W. and Qian, Yu (2019) Patch-based deep learning approaches for artefact detection of endoscopic images. In: Endoscopic artefact detection challenge 2019 (EAD2019), 08 Apr 2019, Venice, Italy.

[img]
Preview
PDF - Published version (with publisher's formatting)
Download (13MB) | Preview
[img] PDF - Final accepted version (with author's formatting)
Download (667kB)

Abstract

This paper constitutes the work in EAD2019 competition. In this competition, for segmentation (task 2) of five types of artefact, patch-based fully convolutional neural network (FCN) allied to support vector machine (SVM) classifier is implemented, aiming to contend with smaller data sets (i.e., hundreds) and the characteristics of endoscopic images with limited regions capturing artefact (e.g. bubbles, specularity). In comparison with conventional CNN and other state of the art approaches (e.g. DeepLab) while processed on whole images, this patch-based FCN appears to achieve the best.

Item Type: Conference or Workshop Item (Speech)
Additional Information: Paper published in Proceedings of the 2019 Challenge on Endoscopy Artefacts Detection (EAD2019), co-located with the 16th International Symposium on Biomedical Imaging (ISBI) Edited by: Sharib Ali, Felix Zhou. CEUR Workshop Proceedings vol. 2366Published on CEUR-WS: 25-May-2019, http://ceur-ws.org/Vol-2366/
Research Areas: A. > School of Science and Technology > Computer Science
Item ID: 26886
Notes on copyright: Copyright © The authors. Copying permitted for private and academic purposes.
Useful Links:
Depositing User: Xiaohong Gao
Date Deposited: 26 Jun 2019 21:08
Last Modified: 08 Oct 2019 16:44
URI: https://eprints.mdx.ac.uk/id/eprint/26886

Actions (login required)

Edit Item Edit Item

Full text downloads (NB count will be zero if no full text documents are attached to the record)

Downloads per month over the past year