A new algorithm for lane detection and tracking on pulsed field gel electrophoresis images

In this paper, a new method is presented for lane detection and extraction on the pulsed field gel electrophoresis (PFGE) images. Average lane width and lane curvature are the most important parameters that are required for automatic image processing of PFGE images. For this purpose, a new algorithm based on computing the power spectrum density (PSD) is proposed for automatic lane detection and tracking. The PSD is used to calculate the average lane width and then partitioning original images to sub-images, tolerating some overlaps. The performance of the proposed algorithm has been evaluated on 30 PFGE images which totally form 300 lanes. Results show that the new algorithm has better performance in lane width detection compared to the other methods such as intersection of the horizontal lines and reaches 99.66%. Finally, it can also be used as a software tool for automatic analysis of PFGE images. (C) 2016 Elsevier B.V. All rights reserved.