This is because you are basically trying to differentiate between these 2 colors. We do 2 things to make this really easy.Ī- Scale the image to make it 100x100 (forced), this will give us 10000 pixels, easy for percentages.ī- We change the channel depth to 1 bit. The main idea is to "convert" an image from one format to another, in this case, the output format is a histogram. Which can be translated to 71.68% red, 26.16% lime and some yellow. Now, all you need to do is execute a command to get the info:Ĭonvert -resize 100x100\! image.jpg -depth 1 -format %c histogram:info:. With windows you might need to set up some path or something, but I'm sure you can find info out there. This should be easy with a Mac or Linux machine. Research you say? Well, in that case you won't be intimidated by some command line work. She has met her first $1000 promise for Alzheimer's research and is working on her second $1000. My wife is proud to participate in the Alzheimer's Quilt Initiative! % red area: 244,669/307,200 x 100 = 79.64%Īdjusting the fuzziness probably has less variability than the natural variation of the leaves in the field. Green selected with Color range and medium fuzziness=62531 pixels This could have other useful applications.
I had never noticed that the selected pixel count displayed in the histogram. Redcrown's suggestion works very well and should be accurate enough for your purposes.
0 kommentar(er)
