*Using Image J to quantify green pixels (plant leaves) using a digital
camera images Hi everybody, I need to measure the green area pixels (leaves) in the images. Can ImageJ do this? If not which free programs would you recommend? Thanks Hilal -- ********************************** Hilal Ilarslan, Ph.D Adjunct Assistant Professor Department of Genetics, Development and Cell Biology Iowa State University, Room 1 Bessey Ames, Iowa 50011-1020 USA (515) 294-3872 FAX: (515) 294-1337 email: [hidden email] http://www.gdcb.iastate.edu/ http://www.public.iastate.edu/~ihilal/ ************************************ Sent from Ames, IA, United States |
Hi Hilal,
I think Imagej can certainly do this, however it depends what you mean by "green" and what you mean by "area": In the very simplest way you could seperate the green channel and measure the number of pixels over a certain threshold or the integrated (average) intensity. However, as you will know, other colors as white and so on also have a "green" content. If you really want the pixels that only belong to a certain range of "green" hue you will need some extra plugin for "color deconvolution" or the like IIRC, I have not yet used them. And for the "area" you might probably need to use one of the many plugin that can generate (a) selection area(s), if you only need to count connected parts starting from a certain size. It all depends on your problem, there will be no "fully automatic" way! You my first thoughts! Cheers Mit freundlichen Grüßen / Best regards Joachim Wesner Hilal iLARSLAN <[hidden email] DU> An Gesendet von: [hidden email] ImageJ Interest Kopie Group <[hidden email]. Thema GOV> measurement of green area (green pixels) in the digital camera images 27.08.2009 18:21 Bitte antworten an ImageJ Interest Group <[hidden email]. GOV> *Using Image J to quantify green pixels (plant leaves) using a digital camera images Hi everybody, I need to measure the green area pixels (leaves) in the images. Can ImageJ do this? If not which free programs would you recommend? ______________________________________________________________________ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email ______________________________________________________________________ |
I've used ImageJ to analyze leaf area from digital images.
Separating out the green channel does not work well, but as Joachim suggested you can use hue. However, I've found that comparing the red and blue channels to the green channel in a manner similar to NDVI works rather well (leaves typically have very little red and only some blue), but requires some adjustment between species. For soybean leaves I use double red_add_blue = r*.6+b*.4; double g_sub_rb = g-red_add_blue; double g_add_rb = g+red_add_blue; double mask = g_sub_rb/g_add_rb; The higher the value, the more likely the pixel belongs to a leaf. You didn't describe much about what you were doing, but typically lighiting is very important. Even lighting produces much better results. Justin |
In reply to this post by Hilal iLARSLAN
Try Landini's "Threshold Color" plugin (see plugins page for download) with
the HSB button set. In my test case (i.e forest with trees, rocks, and people) the leaves had very well defined green hue in ther hue histogram. I think this is better than RGB as everything tends to have some green in it. David Webster On Thu, Aug 27, 2009 at 9:21 AM, Hilal iLARSLAN <[hidden email]> wrote: > *Using Image J to quantify green pixels (plant leaves) using a digital > camera images > > > > > > Hi everybody, > > > > I need to measure the green area pixels (leaves) in the images. > > > > Can ImageJ do this? > > > > If not which free programs would you recommend? > > > > > > Thanks > > > Hilal > > -- > > ********************************** > Hilal Ilarslan, Ph.D > Adjunct Assistant Professor > > Department of Genetics, Development and Cell Biology > Iowa State University, Room 1 Bessey > Ames, Iowa 50011-1020 USA > (515) 294-3872 FAX: (515) 294-1337 > email: [hidden email] > http://www.gdcb.iastate.edu/ > http://www.public.iastate.edu/~ihilal/ > > ************************************ > Sent from Ames, IA, United States > |
Free forum by Nabble | Edit this page |