Filter seeds by finding a local minimum in the kernel density estimation. The first local minimum represents the boundary between the noise at the beak and the real data.
A maximum log KDE of -10 is imposed as otherwise the algorithm will filter real seeds in the case where there is no beak noise.
Filter seeds by finding a local minimum in the kernel density estimation. The first local minimum represents the boundary between the noise at the beak and the real data.
A maximum log KDE of -10 is imposed as otherwise the algorithm will filter real seeds in the case where there is no beak noise.