Number of the records: 1
Automated Object Labeling For CNN-Based Image Segmentation
- 1.
SYSNO ASEP 0533825 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title Automated Object Labeling For CNN-Based Image Segmentation Author(s) Novozámský, Adam (UTIA-B) RID, ORCID
Vít, D. (CZ)
Šroubek, Filip (UTIA-B) RID, ORCID
Franc, J. (CZ)
Krbálek, M. (CZ)
Bílková, Zuzana (UTIA-B) ORCID
Zitová, Barbara (UTIA-B) RID, ORCIDNumber of authors 7 Source Title 2020 IEEE International Conference on Image Processing (ICIP). - Piscataway : IEEE, 2020 - ISSN 1522-4880 - ISBN 978-1-7281-6396-3 Pages s. 2036-2040 Number of pages 5 s. Publication form Print - P Action 2020 IEEE International Conference on Image Processing (ICIP) Event date 25.10.2020 - 28.10.2020 VEvent location Abu Dhabi Country AE - United Arab Emirates Event type WRD Language eng - English Country US - United States Keywords CNN ; SURF ; U-net ; automated object labeling ; image segmentation Subject RIV JC - Computer Hardware ; Software OECD category Computer hardware and architecture R&D Projects GA18-05360S GA ČR - Czech Science Foundation (CSF) TN01000024 GA TA ČR - Technology Agency of the Czech Republic (TA ČR) EID SCOPUS 85098623422 DOI 10.1109/ICIP40778.2020.9191320 Annotation Deep learning-based methods for classification and segmentation require large training sets. Generating training data is often a tedious and expensive task. In industrial applications, such as automated visual inspection of products in an assemble line, objects for classification are well defined yet labeled data are difficult to obtain. To alleviate the problem of manual labeling, we propose to train a convolutional neural network with an automatically generated training set using a naive classifier with handcrafted features. We show that when the naive classifier has high precision then the trained network has both high precision and recall despite the low recall of the naive classifier. We demonstrate the proposed methodology on real scenario of detecting a car coolant tank. However, the proposed methodology facilitates collection of train data for a wider type of CNN based methods such as near-duplicate image detection or segmenting tampered areas of images. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2021
Number of the records: 1