2020-05-11 13:44:55 -0600 | received badge | ● Famous Question (source) |
2018-04-18 04:19:46 -0600 | received badge | ● Notable Question (source) |
2017-05-10 03:08:04 -0600 | received badge | ● Popular Question (source) |
2015-09-10 10:05:20 -0600 | commented question | strategy to build asynchronic subpixel registration analysis 6.64 ms if I use codes inside int main and not sepparate function |
2015-09-10 09:57:15 -0600 | commented question | strategy to build asynchronic subpixel registration analysis tried linear case, with single Stream but that is roughly same speed as w/o with stream: 7.4745 ms w/o: 7.57892 ms (averaged over 1000 readouts) compared with functions without myStream. I also made sepparate function so need to check whether this dos not influence speed |
2015-09-10 07:31:31 -0600 | commented question | strategy to build asynchronic subpixel registration analysis I will try. My analysis has 5x dft + 2x mulSpectrum + 1x magnitude per each image 512x512 (zero padded due to the features close to image border) and these are done without CPU code in-line, therefore I thought if CUDA may be better here. It may be faster on CPU but not so parallel (at least I think) |
2015-09-09 16:56:53 -0600 | received badge | ● Scholar (source) |
2015-09-09 13:46:04 -0600 | answered a question | cross-correlation of 2 same sized images to cross-correlate same size images:
|
2015-09-09 13:13:44 -0600 | asked a question | strategy to build asynchronic subpixel registration analysis Hi, I am analysing set of images for subpixel image shifts. I have code which essantially loops through: loop(){
//next 2 points are based on dft, mulSpectrums, magnitude (all cuda "Streamable")
// next are locating maximum of correlation pattern with subpixel precision
All this is computed ~65000 times, it takes about 8 minutes to compute (256x256 base 16 bit B&W images). Cuda card is not even heating up (nvidia-smi shows 6% GPU-Util). Any suggestions on how to parallelize (the faster the better) this? (also thanks to L.Berger who got me this far) |
2015-09-01 05:14:01 -0600 | received badge | ● Enthusiast |
2015-08-29 06:58:47 -0600 | commented answer | cross-correlation of 2 same sized images cuda::createLinearFilter() is the way to go |
2015-08-28 15:27:13 -0600 | commented answer | cross-correlation of 2 same sized images yes, Ptr<cuda::convolution> conv = cuda::createConvolution(); conv->convolve(zero_padded_source1, source2 ,result, true); But filter 2D will do all this better I guess |
2015-08-28 10:14:27 -0600 | answered a question | cross-correlation of 2 same sized images just zero pad one of the images to (double -1) size of the other and do convolve(zero_padded_source1, source2 ,result, true); |
2015-08-28 10:10:38 -0600 | commented answer | cross-correlation of 2 same sized images Thanks, other software I used did not say how is the cross-correlation produced. I realised you just need to zero pad one of the images and use openCV convolution(ccor=true) |
2015-08-23 18:36:57 -0600 | asked a question | cross-correlation of 2 same sized images Is there a way to compute full cross-correlation (or phase correlation) for two images of same size? -resulting image should be same size as 2 source images. Convolution will only give me one pixel image the way it is implemented. Or do I have to compute it by dft and therefore code it manually? Essentially I am looking for subpixel template matching (for 2 same sized images where an object within shifts) |
2015-06-09 21:30:02 -0600 | received badge | ● Necromancer (source) |
2015-06-09 14:24:56 -0600 | commented answer | 16bit raw file stream analysis cool, I am not sure how much is this applicable but will give it a try. Images here are from Scanning TEM detector so it might be a little bit different |
2015-06-09 13:57:00 -0600 | commented answer | 16bit raw file stream analysis thanks for help, camera itself has 12bit counter and this is saved as 16bit raw file. It depends on our experiment how much of this space is filled. Counts are <20 in my example file (waste of space), but we do not have direct access to drivers or FPGA card which saves datasets. I'm doing analysis of data in proprietary software but I want to try something different. |
2015-06-09 13:37:31 -0600 | commented answer | 16bit raw file stream analysis ok, I am not programmer as is clear I think :) is your suggestion faster? |
2015-06-09 13:19:05 -0600 | received badge | ● Editor (source) |
2015-06-09 13:17:09 -0600 | answered a question | 16bit raw file stream analysis for anyone doing simmilar, to display normalised first image from 16bit 256x256 raw binary file with 256 bytes header, there is also endian conversion (many thanks to L.Berger) |
2015-06-09 12:20:41 -0600 | commented answer | 16bit raw file stream analysis this seems to work: |
2015-06-09 11:39:44 -0600 | commented answer | cv::Mat incorrect reading binary data Can you please comment on this, I have same trouble. |
2015-06-09 11:27:53 -0600 | commented answer | 16bit raw file stream analysis OK thanks, that one works fine now. Other problem is now endians. Value 1 is given as 256 in my Mat. Is there any simple workaround? (if I multiply whole array by 256 in imageJ it gives me 1 for same pixel) |
2015-06-09 10:24:33 -0600 | commented answer | 16bit raw file stream analysis Hi, I got back to this and I am stuck bit further.. this code: creates tmp array with everything 256 x bigger, then if it is read in Mat imAcq=Mat(... it gets to segmentation fault. I am pretty sure my file is 16bit (unsigned short) and if I display values of tmp as in example it gets it right, any suggestions for quick, non all array fix? |
2015-06-09 10:13:09 -0600 | received badge | ● Supporter (source) |