Texture based prelens tear film segmentation in interferometry images

Authors:
Dijia Wu;Kim L. Boyer;Jason J. Nichols;Peter E. King-Smith
Affiliations:
Rensselaer Polytechnic Institute, Department of Electrical, Computer and Systems Engineering, 12180, Troy, NY, USA;Rensselaer Polytechnic Institute, Department of Electrical, Computer and Systems Engineering, 12180, Troy, NY, USA;The Ohio State University, College of Optometry, 43210, Columbus, OH, USA;The Ohio State University, College of Optometry, 43210, Columbus, OH, USA
Venue:
Machine Vision and Applications
Year:
2010

Citing 6
Cited 1

Spatial frequency channels and perceptual grouping in texture segregation

Computer Vision, Graphics, and Image Processing - Special issue on human and machine vission, part II
Multichannel Texture Analysis Using Localized Spatial Filters

IEEE Transactions on Pattern Analysis and Machine Intelligence
Texture Segmentation using 2-D Gabor Elementary Functions

IEEE Transactions on Pattern Analysis and Machine Intelligence
Design of multiple Gabor filters for texture segmentation

ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 04
Unsupervised texture segmentation of images using tuned matched Gabor filters

IEEE Transactions on Image Processing
Optimal Gabor filters for texture segmentation

IEEE Transactions on Image Processing

Resilient subclass discriminant analysis with application to prelens tear film interferometry

MCPR'11 Proceedings of the Third Mexican conference on Pattern recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Interferometric imaging has been identified as a novel approach to the evaluation of prelens tear film (PLTF) thickness in contact lens patients. In this paper, we present a texture based segmentation approach for the detection of tear film breakup regions on interferometry images. First, the textural information was extracted from the studied images using a bank of Gabor filters. A novel classifier, EM-MDA, which integrates traditional Expectation-Maximization with Multiple Discriminant Analysis, was then trained for the recognition of breakup regions of the PLTF. Experimental results provided a correct classification rate of 91.0% which proved significantly higher compared to traditional EM or well known Linear Discriminant Analysis.