Retrieving geometric information from images: the case of hand-drawn diagrams

Abstract : This paper addresses the problem of retrieving meaningful geometric information implied in image data. We outline a general algorithmic scheme to solve the problem in any geometric domain. The scheme, which depends on the domain, may lead to concrete algorithms when the domain is properly and formally specified. Taking plane Euclidean geometry E as an example of the domain, we show how to formally specify E and how to concretize the scheme to yield algorithms for the retrieval of meaningful geometric information in E. For images of hand-drawn diagrams in E, we present concrete algorithms to retrieve typical geometric objects and geometric relations, as well as their labels, and demonstrate the feasibility of our algorithms with experiments. An example is presented to illustrate how nontrivial geometric theorems can be generated from retrieved geometric objects and relations and thus how implied geometric knowledge may be discovered automatically from images.
Type de document :
Article dans une revue
Data Mining and Knowledge Discovery, Springer, 2017, 31 (4), pp.934 - 971. 〈10.1007/s10618-017-0494-1〉
Liste complète des métadonnées

Littérature citée [55 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01646878
Contributeur : Dongming Wang <>
Soumis le : vendredi 24 novembre 2017 - 14:37:47
Dernière modification le : mardi 9 octobre 2018 - 12:54:04

Identifiants

Collections

Citation

Dan Song, Dongming Wang, Xiaoyu Chen. Retrieving geometric information from images: the case of hand-drawn diagrams. Data Mining and Knowledge Discovery, Springer, 2017, 31 (4), pp.934 - 971. 〈10.1007/s10618-017-0494-1〉. 〈hal-01646878〉

Partager

Métriques

Consultations de la notice

79