Mostrar comentarios más antiguos
Hi,
This is NICK.I want to selct a pixel os size 100*100 or 50*50.
I have to do A) Define Mask size (Y*Y) square input in the code I used. B) Move the Mask systematically around image(either in upward or side ward direction). c) Within each area,identify Blank area(white zone)-Brown Infill material (white in picture)-green identify shallow pores -yellow Rock-Blue
http://i39.tinypic.com/wtho5j.jpg. Please give me one example or sample code to solve it.
Thank you, NICK
Respuestas (3)
Image Analyst
el 9 de Dic. de 2011
1 voto
Check out the help for rbbox, and imcrop. Then check out my BlobsDemo:
Check out my other submissions to do color segmentation.
By the way, I'm not sure what step B is or why you think it is necessary. You can find the white blobs simply by thresholding the image - you don't need to scan any mask around.
2 comentarios
Nick ALan
el 10 de Dic. de 2011
Image Analyst
el 10 de Dic. de 2011
Like I said I see no reason at all for systematically moving a window around when you can get everything all in one shot, in one single line of code. Doing your Step B won't help one bit.
Image Analyst
el 10 de Dic. de 2011
0 votos
See Steve's demo:
1 comentario
Nick ALan
el 11 de Dic. de 2011
Walter Roberson
el 10 de Dic. de 2011
0 votos
Your image does not have sufficient resolution to be able to detect the pores in the basalt sample.
In a series of earlier threads since deleted by the poster (aggh!) I posted specific references to studies of pore size ranges in basalt. I then demonstrated that optical wavelengths were too long by a minimum factor of 200 to be able to detect the basalt pore positions even at only 1 pixel per pore.
2 comentarios
Nick ALan
el 11 de Dic. de 2011
Walter Roberson
el 15 de Dic. de 2011
I never got a clear answer as to why that poster kept deleting the older questions as they went along, even though we requested that they not do that.
I already told you to use ginput() or rbbox() to get the input point. Get the coordinate of the lower-left corner. Add 10 to both the x and y in order to get the upper right coordinate. Extract the block between and do whatever you want with it.
I am quite busy these days, and will not be able to assist further in this matter.
Categorías
Más información sobre Blocked Images en Centro de ayuda y File Exchange.
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!