Camera-based document image spotting system for complex linguistic maps
Quoc Bao Dang, Mickaël Coustaty, Muhammad Muzzamil Luqman, Silvia Gally, Paule‐Annick Davoine, Jean-Marc Ogier, Jean-Christophe Burie · 2016
This paper proposes a camera-based document retrieval systems using various local features as well as indexing methods in order to locate a region from dialectology data. Dialectology addresses the study of the linguistic features of languages having a strong oral tradition like local dialects. In order to transcribe ancient maps of Linguistic Atlas of France into geolinguistic data, and to automatically map iso-glosses in interpreted maps, this work aims at identifying the region spot by a camera or a user. This method relies on a new feature, named as Scale and Rotation Invariant Features (SRIF), which is computed based on geometrical constraints between pairs of nearest points around a keypoint. Our systems are applied on dataset including 400 heterogeneous-content complex linguistic map images (9800 × 11768 pixels resolution) and the experimental results show that SRIF outperforms the state-of-the-art in terms of retrieval time with 91.9% retrieval accuracy.