Analysis of Relevant Features for Pollen Classification

Lozano-Vega, Gildardo; Benezeth, Yannick; Boochs, Frank; Marzani, Franck

doi:10.1007/978-3-662-44654-6_39

Cited by 4 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…15 The performance was comparable to that of Chen et al, 13 and superior to Boucher et al, 12 who also employed aperture detection on their pollen classification experiments. However, the results are promising considering the high difficulty of learning three different aperture types with multiple appearances and adding two pollen types in which apertures are not easily visible.…”

Section: Evaluation Of the Classification Performancesupporting

confidence: 56%

See 1 more Smart Citation

Modular method of detection, localization, and counting of multiple-taxon pollen apertures using bag-of-words

Lozano-Vega

Benezeth

Marzani

et al. 2014

J. Electron. Imaging

Self Cite

View full text Add to dashboard Cite

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 enable the detection of discriminant characteristics. Apertures are among the important characteristics which have not been adequately explored until now. A flexible method of detection, localization, and counting of apertures of different pollen taxa with varying appearances is proposed. Aperture description is based on primitive images following the bag-of-words strategy. A confidence map is estimated based on the classification of sampled regions. The method is designed to be extended modularly to new aperture types employing the same algorithm by building individual classifiers. The method was evaluated on the top five allergenic pollen taxa in Germany, and its robustness to unseen particles was verified.

show abstract

Section: Evaluation Of the Classification Performancesupporting

confidence: 56%

“…15. The method relies on the bag-of-words (BOW) approach for the description of apertures in terms of primitive images.…”

Section: Introductionmentioning

confidence: 99%

Modular method of detection, localization, and counting of multiple-taxon pollen apertures using bag-of-words

Lozano-Vega

Benezeth

Marzani

et al. 2014

J. Electron. Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another example is [ 19 ], which uses a neural network approach to perform pollen classification for the reconstruction of remote paleo environments, reporting over 90% of the grains being correctly identified. BOW-related techniques have also been applied in [ 20 ], with a 70% reported correct classification rate over 9 pollen types, and in [ 21 ], with a reported 97.2% of accuracy over just 1 pollen variety.…”

Section: State Of the Artmentioning

confidence: 99%

“…For example, many of the aforementioned studies have been made with sets of data with a very limited amount of images or pollen types. For example, [ 10 – 13 , 15 , 20 , 21 ] work with sets of data containing less than 10 pollen types. The results presented in [ 6 – 9 , 14 , 18 ] work with larger sets of data.…”

Section: State Of the Artmentioning

confidence: 99%

Improving classification of pollen grain images of the POLEN23E dataset through three different applications of deep learning convolutional neural networks

Sevillano¹,

Aznarte²

2018

PLoS ONE

View full text Add to dashboard Cite

In palynology, the visual classification of pollen grains from different species is a hard task which is usually tackled by human operators using microscopes. Its complete automatization would save a high quantity of resources and provide valuable improvements especially for allergy-related information systems, but also for other application fields as paleoclimate reconstruction, quality control of honey based products, collection of evidences in criminal investigations or fabric dating and tracking. This paper presents three state-of-the-art deep learning classification methods applied to the recently published POLEN23E image dataset. The three methods make use of convolutional neural networks: the first one is strictly based on the idea of transfer learning, the second one is based on feature extraction and the third one represents a hybrid approach, combining transfer learning and feature extraction. The results from the three methods are indeed very good, reaching over 97% correct classification rates in images not previously seen by the models, where other authors reported around 70.

show abstract

“…Those slices are scanned in various layers and the images are then used in the analysis and the semi-automatic classification process. The process is further described inLOZANO-VEGA et al 2013 andLOZANO-VEGA et al 2014.…”

mentioning

confidence: 99%

Location Based Sensing for Health Diagnoses: A Prototype for Personalized Pollen Profiling

Böhm¹,

Bock²,

Sehlinger³

2015

giforum

View full text Add to dashboard Cite

Personalized information and forecasts of environmentally triggered medical symptoms are becoming more and more critical, since the numbers of patients suffering from environmentally induced symptoms is still increasing (e.g. allergy or asthma patients). This project aimed at developing a solution to enable gathering personalized pollen exposure information and mapping this information with geospatial data from the personal habitat. This enables personal forecasts based on geospatial information and, when distributed to other suffering individuals, gives them the possibility of avoiding harmful environments. Within an interdisciplinary team of researchers from the medical, engineering, and geoinformatics professions, a system was developed to gather, qualify, and quantify aeroallergen particles (pollen), monitor additional climate parameters, and map this information, together with geospatial information, to the daily habitat of a single individual.

show abstract

Analysis of Relevant Features for Pollen Classification

Cited by 4 publications

References 17 publications

Modular method of detection, localization, and counting of multiple-taxon pollen apertures using bag-of-words

Modular method of detection, localization, and counting of multiple-taxon pollen apertures using bag-of-words

Improving classification of pollen grain images of the POLEN23E dataset through three different applications of deep learning convolutional neural networks

Location Based Sensing for Health Diagnoses: A Prototype for Personalized Pollen Profiling

Contact Info

Product

Resources

About