Improved segmentation of meteorite micro-CT images using local histograms

Authors:
L. D. Griffin;P. Elangovan;A. Mundell;D. C. Hezel
Affiliations:
Computer Science, University College London, London WC1E 6BT, UK;Natural History Museum, Department of Mineralogy, Cromwell Road, SW7 5BD London, UK;Computer Science, University College London, London WC1E 6BT, UK;Natural History Museum, Department of Mineralogy, Cromwell Road, SW7 5BD London, UK
Venue:
Computers & Geosciences
Year:
2012

Citing 9
Cited 2

Statistical Pattern Recognition: A Review

IEEE Transactions on Pattern Analysis and Machine Intelligence
Acquisition, optimization and interpretation of X-ray computed tomographic imagery: applications to the geosciences

Computers & Geosciences - Special issue on three-dimensional reconstruction, modelling and visualization of geologic materials
Towards 3-D petrography: application of microfocus computer tomography in geological science

Computers & Geosciences - Geological Applications of Digital Imaging
The Structure of Locally Orderless Images

International Journal of Computer Vision
Skin Segmentation Using Color Pixel Classification: Analysis and Comparison

IEEE Transactions on Pattern Analysis and Machine Intelligence
A study of Gaussian mixture models of color and texture features for image classification and segmentation

Pattern Recognition
Image registration by local histogram matching

Pattern Recognition
Volume and surface-area measurements using tomography, with an example from the Brenham pallasite meteorite

Computers & Geosciences
Using Basic Image Features for Texture Classification

International Journal of Computer Vision

PhaseQuant: A tool for quantifying tomographic data sets of geological specimens

Computers & Geosciences
Form From Projected Shadow (FFPS): An algorithm for 3D shape analysis of sedimentary particles

Computers & Geosciences

Quantified Score

Hi-index	0.00

Visualization

Abstract

In micro-CT images of meteorites individual components such as matrix, chondrules, Ca,Al-rich inclusions (CAIs), and opaque phases (metal and sulfide) are visually distinguishable. Automated classification of the components is desirable to deal with the large amount of data in a 3-D CT image. Classification by pixel intensity achieves a performance only 25% of the way from baseline to perfect. The poor performance is explained by an overlap in the range of intensities present in the different components. An improved method of semiautomated classification is presented, based on local histograms of the intensity. This achieves a performance 60% of the way from baseline to perfect.