Trying to figure this out, but it looks like a convolution flips the kernel around. If this is the case, then this lib has a bug. It should flip the kernel around before it passes it to the convolution function, so it ends up correct. This did not appear as a bug because the Ricker wavelet is symmetric. I would like confirmation from a math guy.
The convolution is "t-b", while the CWT is "b-t". Note, this is confusing, but the "t" variable in the convolution is the "b" variable in the CWT.
Trying to figure this out, but it looks like a convolution flips the kernel around. If this is the case, then this lib has a bug. It should flip the kernel around before it passes it to the convolution function, so it ends up correct. This did not appear as a bug because the Ricker wavelet is symmetric. I would like confirmation from a math guy.
The convolution is "t-b", while the CWT is "b-t". Note, this is confusing, but the "t" variable in the convolution is the "b" variable in the CWT.
Convolution equation CWT equation