Multi-Thread, Increment-Bandwidth and Weighted Mean-Shift Algorithm for Neural Stem Cells Tracking

For studying their activities in time lapse image sequences, as some limitations of traditional mean-shift algorithm (MSA) applied in large population cells´ tracking exist, an improved MSA with increment bandwidth and adjusted weight is presented for the neural stem cells tracking. Meanwhile, as the limitation of serial computation of MSA, multithreading technique is applied to save tracking time. The presented algorithm has been tested in three image sequences. Tracking result from comparison between improved and traditional MSA shows that the former has increased the accuracy rate and saved time, especially in large populationpsilas cells tracking.