6533b83afe1ef96bd12a7b87
RESEARCH PRODUCT
Modular Method of Detection, Localization and Counting of Mutliple-Taxon Pollen Apertures Using Bag of Words
Frank BoochsYannick BenezethGildardo Lozano-vegaFranck Marzanisubject
Contextual image classificationComputer sciencebusiness.industryLocal binary patternspattern recognitionaperturesCognitive neuroscience of visual object recognition[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Image segmentationmedicine.disease_cause[ INFO.INFO-CV ] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Atomic and Molecular Physics and OpticsComputer Science Applicationsbag of wordsRobustness (computer science)Bag-of-words modelPollenLBPPattern recognition (psychology)medicineComputer visionArtificial intelligenceElectrical and Electronic Engineeringbusinesspalynologydescription
International audience; Accurate recognition of airborne pollen taxa is crucial for understanding and treating allergic diseases, which affect an important proportion of the world population. Modern computer vision techniques enables the detection of discriminant characteristics. Apertures is one of these characteristic that has been little explored up to now. In this paper, a flexible method of detection, localization and counting of apertures of different pollen taxa with varying appearances is proposed. Apertures are described based by primitive images following the Bag-of-Words strat-egy. A confidence map is estimated based on the classification of sampled regions. The method is designed to be extended modularly to new aper-ture types employing the same algorithm by building individual classi-fiers. The method was evaluation on the top 5 allergenic pollen taxa in Germany and robustness to unseen particles was verified.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2014-10-01 |