Fuzzy Similarity Measure and Fractional Image Query for Large Scale Protein 2D Gel Electrophoresis

Spot matching is a challenging problem in two dimensional protein gel electrophoresis (2DGE) chromatography images analysis. It is necessary to provide a robust solution to the problem of querying and matching large scale for various size of 2DGE images. In this paper, we have developed a novel maximum relation spanning tree (MRST) algorithm which is capable of performing fast and accurate matching without the need for landmarks to be manually selected. In the matching process, we employ fuzzy similarity measuring technique to conclude the final decision of matching and location. The resultant system performs up to 94% correct matching for 225 2DGE test images. The additive value is the foundation of querying fractional gel images with large format gel images database.