Login  Register

Re: Image Processing in Art (Is Imagej the right tool?)

Posted by Derin Korman on Oct 05, 2011; 3:28am
URL: http://imagej.273.s1.nabble.com/Image-Processing-in-Art-Is-Imagej-the-right-tool-tp3682462p3682465.html

Thank you for your opinion on the platforms, I am not funded at the
moment but I will keep you in mind if I get funding / start looking
for programming assistance.

Best,
-derin



On Tue, Oct 4, 2011 at 11:13 PM, Doug S <[hidden email]> wrote:

> Derin,
>    I've just started using OpenCV through its python bindings, which solves
> some of the issues of using c & c++. If you use c & c++ I guess you have to
> be careful to to avoid memory leaks and maybe other problems. Python is
> nice, its a pleasure to code in and all the memoty management is taken care
> of. Python is a lot like Java but I have found it easier and faster to to
> code in in terms of development time. You get the ease of coding in python
> but the most expensive computations on images are done in c/c++, and
> potentially even on the GPU, making it faster in terms of the performance of
> the code as well. I use OpenCV as part of a larger package called PythonXY
> which is a scientific computing platform that runs on Windows. Its an easy
> Windows install that installs the main framework and packages like OpenCV
> that you choose. Whats nice is that, like Juanjo was talking about, it
> provides a nice interface, with an interactive programming environment like
> MATLAB plus a good GUI package called Qt, lots of scientific funcitonality
> already built in thru numpy/scipy and nice image/histogram or other
> visualization stuff built in with matplotlib. I find this interface much
> easier to work with than JAVA though JAVA is a step up from using OpenCV in
> pure c/c++. PythonXY does not install on Linux, but all the pieces that you
> would need: nump/scipy, Qt, matplotlib and OpenCV do install on Linux. All
> of these installations are easy except OpenCV, which I have not done yet
> myself, but I have found several sets of instructions on how to do it on the
> net. You could possibly develop the code on WIndows and end up with cross
> platform python code. If you where going to distribute the code to many
> linux or mac machines then looking into how the OpenCV install goes on these
> platforms would be good idea.
> I haven't used Processing before so I can't compare it to PythonXY or
> ImageJ.
>    If you're project was funded and you could afford a little bit of money
> to hire me as a undergrad researcher to handle ( all or part of ) the
> computational side of the project, I might be able to help you out. I am
> math student from the Univ of Connecticut and have done independent study in
> imaging. I finished my BS and am taking time off for health reasons and
> getting ready for doctoral studies and right now I am looking for part time
> projects to keep going while officially out of school. I live in Storrs, CT
> which is about 2 hours from Boston so coming in physically now and then is
> quite possible. If you think this could be helpful for you or the projects
> you're working on, don't hestitate to send me an email, otherwise good luck
> on your projects and finding the right platform.
> Best,
> Doug Snyder
> [hidden email]
>
>
>
> On 10/2/2011 1:32 PM, Derin Korman wrote:
>>
>> Hello Everyone,
>>
>> I am a Teaching Assistant at Harvard's Visual and Environmental
>> studies and I have several projects that need image processing. They
>> mainly revolve around Camera/Image culture, authorship, questions of
>> singularity (not the positivist one).
>>
>> I have no previous programming experience apart from some Actionscript
>> and modifying some github projects, I was wondering if you had a
>> suggestion as to what language to learn. Processing offers some object
>> import functions but I couldn't tell if it would suffice(if it does,
>> that's great as it encompasses some libraries that would ease some
>> work), ImageJ was recommended to my by Nitin Sampat, C++ naturally has
>> various image processing libraries, Matlab I know to be very powerful
>> but one might say it is too difficult a hill to climb before I reach
>> image processing.
>>
>>  example tasks: averaging 1000+ images, doing subtraction/difference
>> calculation, generation of random images, and extracting sequential
>> pixels/parts from images and combining them in another
>> ([x1,x2,x3][y1,y2,y3][z1,z2,z3] ->  [x1,y2,z3])
>>
>> Best,
>> -derin
>> Visual and Environmental Studies
>> Harvard University
>