Abstract : The correct classification of airborne pollen is relevant for medical treatment of allergies, and the regular manual process is costly and time consuming. Aiming at automatic processing, we propose a set of relevant image-based features for the recognition of top allergenic pollen taxa. The foundation of our proposal is the testing and evaluation of features that can properly describe pollen in terms of shape, texture, size and apertures. In this regard, a new flexible aperture detector is incorporated to the tests. The selected set is demonstrated to overcome the intra-class variance and inter-class similarity in a SVM classification scheme with a performance comparable to the state of the art procedures.
Type de document :
Communication dans un congrès
Lazaros Iliadis; Ilias Maglogiannis; Harris Papadopoulos. 10th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2014, Rhodes, Greece. Springer, IFIP Advances in Information and Communication Technology, AICT-436, pp.395-404, 2014, Artificial Intelligence Applications and Innovations. 〈10.1007/978-3-662-44654-6_39〉
https://hal-univ-bourgogne.archives-ouvertes.fr/hal-01095828
Contributeur : Yannick Benezeth
<>
Soumis le : jeudi 3 novembre 2016 - 11:49:47
Dernière modification le : mercredi 12 septembre 2018 - 01:27:36
Document(s) archivé(s) le : samedi 4 février 2017 - 13:36:55
Gildardo Lozano-Vega, Yannick Benezeth, Frank Boochs, Franck Marzani. Analysis of Relevant Features for Pollen Classification. Lazaros Iliadis; Ilias Maglogiannis; Harris Papadopoulos. 10th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2014, Rhodes, Greece. Springer, IFIP Advances in Information and Communication Technology, AICT-436, pp.395-404, 2014, Artificial Intelligence Applications and Innovations. 〈10.1007/978-3-662-44654-6_39〉. 〈hal-01095828〉
