Spatially Aware Dictionary Learning and Coding for Fossil Pollen Identification

TL;DR

This paper presents a novel, robust method for automatic fossil pollen species recognition that combines global shape and local texture features using a spatially-aware dictionary learning and coding approach, achieving high accuracy.

Contribution

It introduces a new exemplar selection criterion optimized via submodular functions and a spatially-aware sparse coding method with accelerated soft-thresholding for fossil pollen identification.

Findings

Achieved 86.13% accuracy on fossil spruce pollen classification

Demonstrated effectiveness of exemplar selection and spatially-aware coding

Provided an efficient matching process for fine-grained species recognition

Abstract

We propose a robust approach for performing automatic species-level recognition of fossil pollen grains in microscopy images that exploits both global shape and local texture characteristics in a patch-based matching methodology. We introduce a novel criteria for selecting meaningful and discriminative exemplar patches. We optimize this function during training using a greedy submodular function optimization framework that gives a near-optimal solution with bounded approximation error. We use these selected exemplars as a dictionary basis and propose a spatially-aware sparse coding method to match testing images for identification while maintaining global shape correspondence. To accelerate the coding process for fast matching, we introduce a relaxed form that uses spatially-aware soft-thresholding during coding. Finally, we carry out an experimental study that demonstrates the effectiveness and efficiency of our exemplar selection and classification mechanisms, achieving $86.13\%$ accuracy on a difficult fine-grained species classification task distinguishing three types of fossil spruce pollen.