Feature Extraction and Machine Learning for the Classification of Brazilian Savannah Pollen Grains

The classification of pollen species and types is an important task in many areas like forensic palynology, archaeological palynology and melissopalynology. This paper presents the first annotated image dataset for the Brazilian Savannah pollen types that can be used to train and test computer visio...

Full description

Saved in:
Bibliographic Details
Published in:PloS one Vol. 11; no. 6; p. e0157044
Main Authors: Gonçalves, Ariadne Barbosa, Souza, Junior Silva, Silva, Gercina Gonçalves da, Cereda, Marney Pascoli, Pott, Arnildo, Naka, Marco Hiroshi, Pistori, Hemerson
Format: Journal Article
Language:English
Published: United States Public Library of Science 08.06.2016
Public Library of Science (PLoS)
Subjects:
Accuracy
Archaeology
Artificial intelligence
Automation
Biology and Life Sciences
Biotechnology
Brazil
Classification
Computer and Information Sciences
Computer science
Computer vision
Criminal investigations
Data mining
Environmental science
Feature extraction
Forensic engineering
Forensic science
Grassland
Honey
Human performance
Identification
Identification and classification
Information retrieval
Learning algorithms
Machine Learning
Medicine and Health Sciences
Palynology
Physiological aspects
Pollen
Pollen - anatomy & histology
Pollen - classification
Research and Analysis Methods
Researchers
Savannahs
Species classification
Sustainability
Wavelet transforms
Brazil
ISSN:1932-6203, 1932-6203
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The classification of pollen species and types is an important task in many areas like forensic palynology, archaeological palynology and melissopalynology. This paper presents the first annotated image dataset for the Brazilian Savannah pollen types that can be used to train and test computer vision based automatic pollen classifiers. A first baseline human and computer performance for this dataset has been established using 805 pollen images of 23 pollen types. In order to access the computer performance, a combination of three feature extractors and four machine learning techniques has been implemented, fine tuned and tested. The results of these tests are also presented in this paper.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Conceived and designed the experiments: JSS ABG HP. Performed the experiments: GGS ABG AP. Analyzed the data: MPC ABG HP. Contributed reagents/materials/analysis tools: AP MPC MHN. Wrote the paper: ABG JSS GGS MPC AP MHN HP.
Competing Interests: The authors have declared that no competing interests exist.
These authors also contributed equally to this work.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0157044