Very interesting video, this has really helped me out, thank you. Could you (or anyone!) explain how this relates to the inverse radon transform please?
I think there is a mistake on 3:09. He's saying "all the frequencies HIGHER than that are set to zero". I guess the right word was "LOWER" as we're talking about a High Pass filter, so lower frequencies than a certain amount get cut.
He does explain the reason for that. Higher frequencies contain more noise and less signal which results in more noise in the output. In order to reduce the noise in the output he uses a cutoff value.
Jay Bhanushali but the ram lak filter is a high pass filter. It cuts LOW frequencies as in the standard back projection low frequencies gain more importance than the high ones and this provokes the blurring noise. So you have to gain high frequencies, not cut them off...
@@elijak8263 I thought that too at first. But if you look closely at the frequency response of that ram lak filter, you can see that it first ramps up linearly and then falls down abruptly at that said cut-off frequency. So yes it is a High Pass filter but with a cut-off at a high-frequency to eliminate the high-frequency noise.
No the ramp filter does not CUT any low frequencies. His filter is modified so that frequencies above a certain frequency are cut off (although if you ask me that's a rather inelegant way to do things, he should multiply his raw ramp filter with a gaussian function) to suppress high frequency noise otherwise high frequency noise would be boosted to a ridiculous degree.
I think you guys just saved my CT exam on monday...
Such an intuitive explanation. My teacher spends 2 hours explaining it and I still didn't get it!
I'm doing my final year project on this and it has help me so much in understanding what my equations are actually doing. Thank you!!!!!
Best explaination to be found on youtube! Good job! This really helped me understanding
Wow this is exactly what I searched for very nice and easy explanation
2:53 can you make a video where you say Ram-Lak for 10 minutes?
Can I ask how you generated the projection images @1:30?
Brief and clear. Thanks a lot! :)
Nice video! clear and easy
Very interesting video, this has really helped me out, thank you.
Could you (or anyone!) explain how this relates to the inverse radon transform please?
Medical Physics exam in 12 hours... thanks!
Well explained! Thanks!
Thank you for the good explanation, are you still planning on releasing a video about the derivation of FBP from the Fourier slice theorem?
love this video
I think there is a mistake on 3:09. He's saying "all the frequencies HIGHER than that are set to zero". I guess the right word was "LOWER" as we're talking about a High Pass filter, so lower frequencies than a certain amount get cut.
He does explain the reason for that. Higher frequencies contain more noise and less signal which results in more noise in the output.
In order to reduce the noise in the output he uses a cutoff value.
Jay Bhanushali but the ram lak filter is a high pass filter. It cuts LOW frequencies as in the standard back projection low frequencies gain more importance than the high ones and this provokes the blurring noise. So you have to gain high frequencies, not cut them off...
@@elijak8263 I thought that too at first. But if you look closely at the frequency response of that ram lak filter, you can see that it first ramps up linearly and then falls down abruptly at that said cut-off frequency. So yes it is a High Pass filter but with a cut-off at a high-frequency to eliminate the high-frequency noise.
No the ramp filter does not CUT any low frequencies. His filter is modified so that frequencies above a certain frequency are cut off (although if you ask me that's a rather inelegant way to do things, he should multiply his raw ramp filter with a gaussian function) to suppress high frequency noise otherwise high frequency noise would be boosted to a ridiculous degree.
this is amazing video, but anyone can tell me that whats the difference back projection and forward projection?
Please give me answer if you know ?
i love you man
great !
Each point on the 'object domain'? In English world no one uses that complex terminology to explain relatively simple things.
you sound belgian
Hansen Isle