Outside of having a point spread function with which to deconvolve an image, has anyone developed any successful image sharpening / deblurring routines they would wish to share?
Thanks for posting to the NI forums! Have you seen the KB linked below from the ni.com website? It uses Grey-Scale modification to achieve various results, one of which is sharpening an image.
I have access to grey and color table modification including contrast expansion / stretching. I probably need to explore convolutions as proposed by: http://www.dfanning.com/ip_tips/sharpen.html. I am building a signal and image processing custom software program for nondestructive evaluation applications, and techniques such as this will help in the detection of flaws in materials. I have convolution capability per IMAQ fxs built into the code - but I wanted to know if some on this forum (either thru vision or custom codes developed to work on intensity graph or picture control data) have found some good specific kernal filters to work well. I supposed it mostly going to be trial and error.
The convolution fx I have using IMAQ fx can do a 3x3 or any size kernal. The sharpening filter they suggest in the article does not work very well in my case. But thanks.
Hi anorallee, -Please check shipped example along with labview first. Goto>>Help>>Find Examples>>Search for Convolution>> you will get edge detection example with laplacian.
