http://www.k-wave.org/forum/topic/applying-2d-fft-reconstruction-to-real-signal Support for the k-Wave MATLAB toolbox en-US Tue, 12 May 2026 23:36:24 +0000 http://bbpress.org/?v=1.0.2 q http://www.k-wave.org/forum/search.php http://www.k-wave.org/forum/topic/applying-2d-fft-reconstruction-to-real-signal#post-428 Tue, 03 Apr 2012 02:08:30 +0000 Seongjun Park 428@http://www.k-wave.org/forum/ <p>Hi Ben, </p> <p>It works!!! </p> <p>Thank you very much!!! </p> <p>Seongjun </p> http://www.k-wave.org/forum/topic/applying-2d-fft-reconstruction-to-real-signal#post-411 Mon, 26 Mar 2012 21:07:50 +0000 bencox 411@http://www.k-wave.org/forum/ <p>Hi Seongjun, </p> <p>The image reconstruction algorithms in k-Wave are specifically designed for photoacoustic imaging where you are essentially solving an inverse initial value problem (where the initial condition is the acoustic pressure generated thermally by the laser pulse). This is slightly different mathematically from the scattering problem of ultrasound imaging and so the reconstruction algorithms will not be the same for both. Beam forming has been used to do photoacoustic imaging, but there are exact algorithms for homogeneous media which will (usually) give superior results.</p> <p>Hope you can get it working!</p> <p>Ben </p> http://www.k-wave.org/forum/topic/applying-2d-fft-reconstruction-to-real-signal#post-410 Mon, 26 Mar 2012 20:16:53 +0000 Seongjun Park 410@http://www.k-wave.org/forum/ <p>Hi Ben, </p> <p>In fact, I did both experiments photoacoustic &amp; ultrasound imaging. However, I usually used datas from the ultrasound imaging experiment due to the few reasons. Does 2D FFT reconstruction algorithm work only for photoacoustic imaging? </p> <p>And the 'traditional' reconstruction algorithm means parallel beamformers which is commonly used in "Ultrafast imaging". And the following paper will help you out:<br /> [ref] Gabriel Montaldo, Mickaël Tanter, Jérémy Bercoff, Nicolas Benech, and Mathias Fink<br /> "Coherent Plane-Wave Compounding for Very High Frame Rate Ultrasonography and Transient Elastography"<br /> IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 56, no. 3, March 2009. </p> <p>The phantom was point-like absorbers for photoacoustic imaging and a wire for ultrasound imaging. </p> <p>If you need, I can upload results. Also I am going to do additional photoacoustic experiment to abtain the better image. </p> <p>Thank you very much. </p> <p>Seongjun Park </p> http://www.k-wave.org/forum/topic/applying-2d-fft-reconstruction-to-real-signal#post-407 Sat, 24 Mar 2012 23:11:39 +0000 bencox 407@http://www.k-wave.org/forum/ <p>Hi Seongjun, </p> <p>Can I check I understand: are you doing a photoacoustic imaging experiment with a linear array of 128 elements and using the 2D FFT reconstruction to obtain an image, and comparing the image to one obtained using another algorithm? Is that all correct? If so, what do you mean by 'traditional' reconstruction algorithm, and what does your initial acoustic pressure distribution look like?</p> <p>One thing that worries me is that experiments are usually 3D and you're using a 2D image reconstruction algorithm. Unless, of course, you've taken great care to ensure it is a two-dimensional experiment, i.e.. non varying in the direction perpendicular to the array.</p> <p>Ben </p> http://www.k-wave.org/forum/topic/applying-2d-fft-reconstruction-to-real-signal#post-405 Sat, 24 Mar 2012 05:47:16 +0000 Seongjun Park 405@http://www.k-wave.org/forum/ <p>Hello Ben, </p> <p>I am still studying this amazing tool.<br /> Right now, I am trying to apply 2D FFT Reconstruction algorithm to the experimental data obtained by 128 linear arrays and compare its performance to the other when using tranditional reconstruction algorithm. Unfortunately, there were much more artifacts in the result when using 2D FFT Reconstruction. Of course, the phantom used in experiment was homogeneous one and also, I did not consider any attenuation effect when using tranditional reconstruction algorithm. </p> <p>Also, to compare both algorithms in another way, I used the signal which is obtained by 'kwave' and applied it to both algorithms. However, in this case, traditional one didn't work well. </p> <p>But I found that some differences between those singals one obatained from real experiment and the other from kwave. The signal from experiment was shaped like circle but, the sinal from kwave was shaped like parabolic one. I think this is one of the reasons giving different results. Would you tell me why the sensor signals are parabolic shaped? </p> <p>Thank you. </p> <p>Seongjun Park </p>