Symbol Recognition: Current Advances and Perspectives
GREC '01 Selected Papers from the Fourth International Workshop on Graphics Recognition Algorithms and Applications
Integration of Local and Global Shape Analysis for Logo Classification
IWVF-4 Proceedings of the 4th International Workshop on Visual Form
Segmentation of Brazilian Bank Check Logos without a Priori Knowledge
ITCC '00 Proceedings of the The International Conference on Information Technology: Coding and Computing (ITCC'00)
Vehicle logo recognition using mathematical morphology
TELE-INFO'07 Proceedings of the 6th WSEAS Int. Conference on Telecommunications and Informatics
A polar-based logo representation based on topological and colour features
DAS '10 Proceedings of the 9th IAPR International Workshop on Document Analysis Systems
Hi-index | 0.00 |
Recently, much attention has been paid on the recognition of graphical objects, like company logos and trademarks. Recognizing these objects facilitates the recognition of document class. Some promising results have been achieved by using Autoassociator-based Artificial Neural Networks (AANN) also in the presence of homogeneously distributed noise. However, the performance drops significantly when dealing with spot-noisy logos, where strips or blobs produce a partial obstruction of the pictures.In this paper, we propose a new approach for training AANNs especially conceived for dealing with spot noises. The basic idea is that of introducing a new norm for assessing the reproduction error in AANNs. The proposed algorithm, which is referred to as Spot-Backpropagation (S-Bp), is significantly much more robust with respect to spot-noise than classical Euclidean norm-based Backpropagation (Bp). Our experimental results are based on a database of 88 real logos that are artificially corrupted by spot-noise.