Here we present results on 2D electrophoresis gel image processing using methods that provide a measure of meaningfulness. This work is part of an ongoing project on biomedical image analysis. Biomedical sciences have a long tradition in our country and therefore there is extensive experience in this area. Unfortunately, economical restrictions sometimes do not allow our researchers to have the latest technology. As we will see later, 2D electrophoresis gels are an extremely useful and affordable technology, which can be used in our countries. From the image processing point of view we believe that there is room for improvements when thinking in the final software application. With this in mind we present an algorithm that covers all the steps for gel image registration. First we present a method for robust and meaningful detection of spots. Then we study two improvements on the computation of the distance between spots using shape contexts. Finally, we present an iterative random sampling process which deals with spot differences between images to give a gel registration. Across all the steps we address the easy interaction with the user based on a measure of meaningfulness of the results.
[1]
Jean Serra,et al.
Image Analysis and Mathematical Morphology
,
1983
.
[2]
V. Caselles,et al.
The Tree of Shapes of an Image
,
2003
.
[3]
Lionel Moisan,et al.
Edge Detection by Helmholtz Principle
,
2001,
Journal of Mathematical Imaging and Vision.
[4]
Haibin Ling,et al.
An Efficient Earth Mover's Distance Algorithm for Robust Histogram Comparison
,
2007,
IEEE Transactions on Pattern Analysis and Machine Intelligence.
[5]
Jean-Michel Morel,et al.
From Gestalt Theory to Image Analysis
,
2008
.
[6]
Jean-Michel Morel,et al.
From Gestalt Theory to Image Analysis: A Probabilistic Approach
,
2007
.
[7]
Jim Graham,et al.
Robust and Accurate Registration of 2-D Electrophoresis Gels Using Point-Matching
,
2007,
IEEE Transactions on Image Processing.
[8]
S. Varadhan,et al.
A probabilistic approach to
,
1974
.