Thanks for the video. As I don't have much background knowledge, sometimes I feel really lost on what I am listening. I think a feel explanation on the purpose/goal behind the functions and equations will be very helpful for many people. For example "Laplacian filters are derivative filters used to find areas of rapid change (edges) in images. Since derivative filters are very sensitive to noise, it is common to smooth the image (e.g., using a Gaussian filter) before applying the Laplacian. This two-step process is call the Laplacian of Gaussian (LoG) operation." (from academic.mu.edu/phys/matthysd/web226/Lab02.htm)
The output matrix at 04:59 is incorrect in the sense that it is not exactly calculated from the original matrix. Rather a shifted version of the original matrix. I'm guessing that when you were cropping the matrices from bigger matrices, the output matrix was cropped one column to the left of its actual position.
The change in matrix at 4:59 is well explained here th-cam.com/video/zQKNVept4bU/w-d-xo.html Also if you'll be making the computation on Matlab, use the full blown matrix(new matrix gotten from adding row and column values to your input matrix) shown in the link above and not the matrix of your input image, then your laplacian mask. The result gives the actual matrix of the filtered image
It is an operator that was invented by Laplace. There are very good explanations for this in the 3blue1brown channel in the differential calculus videos I think the videos where he speaks about curls or some video in around that one.
QUESTION REGARDING GRAPH : Hi , Thanks for all your videos. Am following you in TH-cam for long time, your videos were really clear and crisp to understand. I have a question regarding plotting of the intensity graph ,How to plot this graph ?? Could you please help me in explaining or it would be really great if you post a video about this, it will help everyone to know about how to analyse our own filter :) Thanks again, Cufu
rashi agrawal Thanks :) I have another question also, have you ever worked on Advanced filtering metods ( Variational approaches) something like Chan vese or Mumford shah Algorithm ? I would like to discuss a little bit about those which is a part of my ongoing research work.
Hi dear thank uuu so much for your videos , it's very helpfu , well there's a mistake in the matrix of the image with 2 lines in background, it's not the same as the matrix that you used it the Matlab sicript , and finally thank uuuuu for your efforts
Thank you very much for this lesson. I have one question: When you implement laplacian filter to image with pixels values at 4:56 you say that the result value of filter goes like this: 1*150 + 1*150 + 1*150 + 1*10 - 4*150 of witch result is -140. Why is then the results at the picture below - 280 ? I canot figure out how this goes exactly. I also explain my question at the picture below: shrani.si/f/3q/ob/3nanx9B0/problem.jpg Tnx very much for answer!
we get -140 because after apply kernal matrix is reduced by 2 columns and 2 rows,1 from up and down each and same 1 from left right each. because kernal is applied on valid matrix.
Thanks for the video. As I don't have much background knowledge, sometimes I feel really lost on what I am listening. I think a feel explanation on the purpose/goal behind the functions and equations will be very helpful for many people. For example "Laplacian filters are derivative filters used to find areas of rapid change (edges) in images. Since derivative filters are very sensitive to noise, it is common to smooth the image (e.g., using a Gaussian filter) before applying the Laplacian. This two-step process is call the Laplacian of Gaussian (LoG) operation." (from academic.mu.edu/phys/matthysd/web226/Lab02.htm)
Great explanation in 8 minute, thank you.
The output matrix at 04:59 is incorrect in the sense that it is not exactly calculated from the original matrix. Rather a shifted version of the original matrix.
I'm guessing that when you were cropping the matrices from bigger matrices, the output matrix was cropped one column to the left of its actual position.
Thanks Rashi, you solved my problem of understanding with easy words. Keep it up.
kia bat hai. very helpfull.
Very nice explanation! Thank you.
Nice video, very useful and well explained. Thank You
The change in matrix at 4:59 is well explained here
th-cam.com/video/zQKNVept4bU/w-d-xo.html
Also if you'll be making the computation on Matlab, use the full blown matrix(new matrix gotten from adding row and column values to your input matrix) shown in the link above and not the matrix of your input image, then your laplacian mask.
The result gives the actual matrix of the filtered image
Amazing still better than my college lecturers
You are so awesome.Thank you for this video.
why 8*10 matrix converted into 6*8 matrix at 5:06 ????
She does a valid convolution, only applying the filter where there is a full overlap between filter and image
excellent. thank you.
Vinod could you do an assignment for me ill pay
@@xych0938 Do You know Me?
No but if you'd do an assignment for me it'll be very great cuz i have no idea
@@xych0938 i too dont have any idea regarding these topics...
Ah my bad sorry for raking your time
The video explains very well! I have a question. How do you obtain the equation of d2f/dx2 and d2f/dy2? It's shown in 1:16. Thanks in advance! :D
It is an operator that was invented by Laplace. There are very good explanations for this in the 3blue1brown channel in the differential calculus videos I think the videos where he speaks about curls or some video in around that one.
Great Explanation
Thank you very much for this!
Thank you. Very helpful!
@3.43 why did u change the axis? I don't understand that
nice explanation
Thank you very much for this nice video
Hi, how about explaining to apply Laplacian filter on digital singal processing.
good work! can you put up a video on how to embed in contourlet steganography?
Thank you.
Thanks a lot for this
plz tell me what is zero crossing??
QUESTION REGARDING GRAPH :
Hi ,
Thanks for all your videos. Am following you in TH-cam for long time, your videos were really clear and crisp to understand.
I have a question regarding plotting of the intensity graph ,How to plot this graph ??
Could you please help me in explaining or it would be really great if you post a video about this, it will help everyone to know about how to analyse our own filter :)
Thanks again,
Cufu
Thanks for the positive feedback. Which intensity graph are u referring to?
rashi agrawal
at the start of the slide which you have shown.
We also use that one in previous lecture.
gopi cufu
I mean the one where you just put a line in an image & plot the varying intensity levels on that particular line.
gopi cufu oh sure! no problem..will do that
rashi agrawal
Thanks :)
I have another question also, have you ever worked on Advanced filtering metods ( Variational approaches) something like Chan vese or Mumford shah Algorithm ?
I would like to discuss a little bit about those which is a part of my ongoing research work.
do you have any paper? i want to refer you
Hi dear thank uuu so much for your videos , it's very helpfu , well there's a mistake in the matrix of the image with 2 lines in background, it's not the same as the matrix that you used it the Matlab sicript , and finally thank uuuuu for your efforts
Thank you very much for this lesson. I have one question:
When you implement laplacian filter to image with pixels values at 4:56 you say that the result value of filter goes like this: 1*150 + 1*150 + 1*150 + 1*10 - 4*150 of witch result is -140. Why is then the results at the picture below - 280 ? I canot figure out how this goes exactly. I also explain my question at the picture below:
shrani.si/f/3q/ob/3nanx9B0/problem.jpg
Tnx very much for answer!
Hi!
I have the same question but you seem to have the answer already? Can you share it?
did anyone found why? even on Matlab it says -140. I suppose the uploader is wrong.
we get -140 because after apply kernal matrix is reduced by 2 columns and 2 rows,1 from up and down each and same 1 from left right each. because kernal is applied on valid matrix.
hi guys he matrix of the image with 2 lines in background, it's not the same as the matrix that you used it the Matlab sicript
Enlarge the screen because what's going in the numerical not getting
FUCK! Thank you very much!
Mam your Voice so cute.... :) I jst forgot to concentrate on concept
nah gtfoh
gud job