Lanes Detection in PCR Gel Electrophoresis Images

This study aims at development of methods to track the center of and detect lanes as the first step of automated tool to analyze rose DNA using PCR gel electrophoresis images. Although several research results have been previously reported using projection profiles in a whole image, it is still challenge to track the center of and detect bent lanes using projection profiles. To resolve the problem, we partitioned an input image into small images and found local maxima, potential lane centers, on the vertical projection in each partitioned image. We used 29 PCR gel images including 934 lanes to evaluate the performance of the proposed scheme. Proposed scheme achieved the performance of the sensitivity of 99.0% and the precision of 98.9%.