Dorozynski, M.; Clermont, D.; Rottensteiner, F.: Multi-task deep learning with incomplete training samples for the image-based prediction of variables describing silk fabrics. In: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences 4 (2019), Nr. 2/W6, S. 47-54. DOI: https://doi.org/10.5194/isprs-annals-IV-2-W6-47-2019
Abstract: | |
This paper presents a method for the classification of images of silk fabrics with the aim to predict properties such as the placeand time of origin and the production technique. The proposed method was developed in the context of the EU project SILKNOW(http://silknow.eu/). In the context of classification, we address the problem of limited as well as not fully labelled data andinvestigate the connection between the distinct variables. A pre-trained Convolutional Neural Network (CNN) is used for thefeature extraction and a classification network realizing Multi-task learning (MTL) is trained based on these features. The trainingprocedure is adapted to enable the consideration of images that do not have a label for all tasks. Additionally, MTL with fullylabeled training data is investigated for the classification of silk fabrics. The impact of both MTL approaches is compared to singletask learning based on two different class structures. We achieve overall accuracies of 92-95% and average F1-scores of 88-90% inour best experiments. © 2019 Authors. | |
License of this version: | CC BY 4.0 Unported |
Document Type: | Article |
Publishing status: | publishedVersion |
Issue Date: | 2019 |
Appears in Collections: | Fakultät für Bauingenieurwesen und Geodäsie |
pos. | country | downloads | ||
---|---|---|---|---|
total | perc. | |||
1 | Germany | 85 | 43.37% | |
2 | United States | 26 | 13.27% | |
3 | China | 15 | 7.65% | |
4 | India | 9 | 4.59% | |
5 | No geo information available | 6 | 3.06% | |
6 | Israel | 6 | 3.06% | |
7 | Japan | 5 | 2.55% | |
8 | Indonesia | 4 | 2.04% | |
9 | France | 4 | 2.04% | |
10 | Iran, Islamic Republic of | 3 | 1.53% | |
other countries | 33 | 16.84% |
Hinweis
Zur Erhebung der Downloadstatistiken kommen entsprechend dem „COUNTER Code of Practice for e-Resources“ international anerkannte Regeln und Normen zur Anwendung. COUNTER ist eine internationale Non-Profit-Organisation, in der Bibliotheksverbände, Datenbankanbieter und Verlage gemeinsam an Standards zur Erhebung, Speicherung und Verarbeitung von Nutzungsdaten elektronischer Ressourcen arbeiten, welche so Objektivität und Vergleichbarkeit gewährleisten sollen. Es werden hierbei ausschließlich Zugriffe auf die entsprechenden Volltexte ausgewertet, keine Aufrufe der Website an sich.