Talk:Hough transform: Difference between revisions
Content added Content deleted
(→Can I get some pseudo-code...and possibly an amen: Differences are fine, and we need more examples.) |
|||
| Line 2: | Line 2: | ||
I have read about the Hough Transform in an attempt to create a MATLAB implementation, but every resource I use has a different take on this transform. The background theory is the same, but the algorithms that I have seen are different. Even the TCL and C implementations on this site are somewhat different from some other implementations I have found. Can we possibly get some solid pseudo-code to implement to make sure any other solutions provided in different languages conform to some standard specification? Because, at this point, my MATLAB implementation is not going to come close to resembling the algorithm implemented in TCL and C. Can I get an Amen?!--[[User:Cferri|Chris Ferri]] |
I have read about the Hough Transform in an attempt to create a MATLAB implementation, but every resource I use has a different take on this transform. The background theory is the same, but the algorithms that I have seen are different. Even the TCL and C implementations on this site are somewhat different from some other implementations I have found. Can we possibly get some solid pseudo-code to implement to make sure any other solutions provided in different languages conform to some standard specification? Because, at this point, my MATLAB implementation is not going to come close to resembling the algorithm implemented in TCL and C. Can I get an Amen?!--[[User:Cferri|Chris Ferri]] |
||
: Not really. What's important at the surface is that the inputs match and the outputs match. How different examples achieve this will vary by how the problem best maps to each language. MATLAB's implementation ''should'' differ, because MATLAB looks at mathematical problems differently from TCL or C. Though, yes, we definitely need more examples. --[[User:Short Circuit|Michael Mol]] 18:35, 9 August 2010 (UTC) |
: Not really. What's important at the surface is that the inputs match and the outputs match. How different examples achieve this will vary by how the problem best maps to each language. MATLAB's implementation ''should'' differ, because MATLAB looks at mathematical problems differently from TCL or C. Though, yes, we definitely need more examples. --[[User:Short Circuit|Michael Mol]] 18:35, 9 August 2010 (UTC) |
||
::But, I thought the point of Rosetta Code was to be able to compare implementations of a common task between languages. I understand that this definition doesn't invalidate your response, but I don't see the point of ''not'' specifying a pseudo-code to implement. It would be much easier for: not only novices trying to figure out how to do something using the solutions as examples, but also developers who are confused about how to implement this particular algorithm, if there were pseudo-code to refer to. I say this especially since my implementation requires an "edge-detected" version of the image as the input (a boolean array where the array member at the index of an "edge" pixel is true). And, from what I can tell, the TCL and C solutions don't. So, I am not exactly sure what is going on in those algorithms which allows them to produce valid output. This is why pseudo-code is important.--[[User:Cferri|Chris Ferri]] |
|||
==PNG Image== |
==PNG Image== |
||