Application of statistical learning theory to plankton image analysis
Citable URI
https://hdl.handle.net/1912/1237DOI
10.1575/1912/1237Keyword
Plankton; Measurement; Machine learningAbstract
A fundamental problem in limnology and oceanography is the inability to quickly
identify and map distributions of plankton. This thesis addresses the problem by
applying statistical machine learning to video images collected by an optical sampler,
the Video Plankton Recorder (VPR). The research is focused on development
of a real-time automatic plankton recognition system to estimate plankton abundance.
The system includes four major components: pattern representation/feature
measurement, feature extraction/selection, classification, and abundance estimation.
After an extensive study on a traditional learning vector quantization (LVQ)
neural network (NN) classifier built on shape-based features and different pattern
representation methods, I developed a classification system combined multi-scale cooccurrence matrices feature with support vector machine classifier. This new method
outperforms the traditional shape-based-NN classifier method by 12% in classification
accuracy. Subsequent plankton abundance estimates are improved in the regions of
low relative abundance by more than 50%.
Both the NN and SVM classifiers have no rejection metrics. In this thesis, two
rejection metrics were developed. One was based on the Euclidean distance in the
feature space for NN classifier. The other used dual classifier (NN and SVM) voting as
output. Using the dual-classification method alone yields almost as good abundance
estimation as human labeling on a test-bed of real world data. However, the distance
rejection metric for NN classifier might be more useful when the training samples are
not “good” ie, representative of the field data.
In summary, this thesis advances the current state-of-the-art plankton recognition
system by demonstrating multi-scale texture-based features are more suitable
for classifying field-collected images. The system was verified on a very large realworld
dataset in systematic way for the first time. The accomplishments include developing a multi-scale occurrence matrices and support vector machine system, a dual-classification system, automatic correction in abundance estimation, and ability to get accurate abundance estimation from real-time automatic classification. The methods developed are generic and are likely to work on range of other image classification applications.
Description
Submitted to the Joint Program in Applied Ocean Science and Engineering
in partial fulfillment of the requirements for the degree of Doctor of Philosophy
At the Massachusetts Institute of Technology
and the Woods Hole Oceanographic Institution
June 2006
Collections
Suggested Citation
Thesis: Hu, Qiao, "Application of statistical learning theory to plankton image analysis", 2006-06, DOI:10.1575/1912/1237, https://hdl.handle.net/1912/1237Related items
Showing items related by title, author, creator and subject.
-
Machine learning techniques to characterize functional traits of plankton from image data
Orenstein, Eric C.; Ayata, Sakina Dorothée; Maps, Frédéric; Becker, Érica C.; Benedetti, Fabio; Biard, Tristan; de Garidel-Thoron, Thibault; Ellen, Jeffrey S.; Ferrario, Filippo; Giering, Sarah L. C.; Guy-Haim, Tamar; Hoebeke, Laura; Iversen, Morten H.; Kiørboe, Thomas; Lalonde, Jean-François; Lana, Arancha; Laviale, Martin; Lombard, Fabien; Lorimer, Tom; Martini, Séverine; Meyer, Albin; Möller, Klas O.; Niehoff, Barbara; Ohman, Mark D.; Pradalier, Cédric; Romagnan, Jean-Baptiste; Schröder, Simon-Martin; Sonnet, Virginie; Sosik, Heidi M.; Stemmann, Lars; Stock, Michiel; Terbiyik-Kurt, Tuba; Valcárcel-Pérez, Nerea; Vilgrain, Laure; Wacquet, Guillaume; Waite, Anya M.; Irisson, Jean-Olivier (Association for the Sciences of Limnology and Oceanography, 2022-06-30)Plankton imaging systems supported by automated classification and analysis have improved ecologists' ability to observe aquatic ecosystems. Today, we are on the cusp of reliably tracking plankton populations with a suite ... -
Techniques for studying vocal learning in bottlenose dolphins, Tursiops truncatus
Fripp, Deborah R. (Massachusetts Institute of Technology and Woods Hole Oceanographic Institution, 1999-02)The objective of this thesis is to develop the methods necessary for evaluating the role of learning in the natural whistle development of bottlenose dolphins. Bottlenose dolphins provide a unique opportunity to study ... -
Automated open circuit scuba diver detection with low cost passive sonar and machine learning
Cole, Andrew M. (Massachusetts Institute of Technology and Woods Hole Oceanographic Institution, 2019-06)This thesis evaluates automated open-circuit scuba diver detection using low-cost passive sonar and machine learning. Previous automated passive sonar scuba diver detection systems required matching the frequency of diver ...