Building, deploying and managing industrial Machine Vision and Computer Vision software systems.
Thursday, 8 January 2009
Sub-Pixel Maximum
Ever needed to find the position of the maximum of a peak in an image to sub-pixel precision? Examples are finding the peak energy in Fourier space, or the location of a peak in Hough space. One simple method that I have used with some success is a simple quadratic interpolation, which is performed once for each axis.
Sub-Pixel Peak finding recipe... Firstly, you take the value of your highest point and its neighbours (Pa, Pb and Pc). Assuming the three points are equally spaced and bracket the top of the peak, you can solve for the position of the 'true' maximum using:
x = 0.5*(Pa - Pc) / (Pa-2Pb+Pc)
The value of x will range between -0.5 and +0.5, which is relative to your maximum pixel. Jason Dale. www.visionexperts.co.uk
“We are dedicating all of our future product development to multicore designs. … This is a sea change in computing” - Paul Otellini, President, Intel (2005)
“Multicore: This is the one which will have the biggest impact on us. We have never had a problem to solve like this. A breakthrough is needed in how applications are done on multicore devices.” - Bill Gates, Microsoft
“When we start talking about parallelism and ease of use of truly parallel computers, we're talking about a problem that's as hard as any that computer science has faced. … I would be panicked if I were in industry.” - John Hennessy, President of Stanford
Jason Dale received his master’s degree in Astrophysics in 1997 from the University of Kent, England, before researching a PhD in real-time neuromorphic vision systems at University College London. He has since worked as an application engineer, developing industrial and military machine vision systems.
He is now a contract machine vision engineer (http://www.visionexperts.co.uk) writing custom machine vision algorithms for industry. He is also software architect for Vision4ce Ltd (http://www.vision4ce.com) where he continues to develop GPU software for image processing in rugged applications.
No comments:
Post a Comment