spot detector-how the surface and average intensity is calculated?

54 views
Skip to first unread message

Yuan ZHOU

unread,
Jun 17, 2015, 4:41:05 AM6/17/15
to icy-so...@googlegroups.com
Hi,

I am using spot detector to quantify the number and intensity of each spot in my image stack. I am using scale 2, threshold sensitivity 20, no filtering to detect spots. I have following questions:
1. I found that in some case, the surface of the spot is measured to be 1 pixel, which is apparently smaller than visual impression (for example, the No.8 spot in this image). So how to solve this problem and how is the surface of a spot calculated? 
2. Is the average intensity calculated by the sum intensity of all pixels in a spot dividing by the surface? How can I get the sum intensity of a spot in 3D? 
3. The detector fail to differentiate two close spots (No.3 in this image). Can I manually add a detected spot? or any other method to differentiate these two?

Thanks a lot!

Lillian
p5.tif.xls
p5.tif

Stephane

unread,
Jun 29, 2015, 4:44:27 AM6/29/15
to icy-so...@googlegroups.com, lilian...@gmail.com
Hello Yuan,

Actually using 20 as sensitivity parameter is really low and so the spot detector will under estimate the real size of spot.
If you want to check what the spot detector is actually detecting, you can enable the "display binary image" in Detector parameters so you exactly see what the spot detector find.
The mean intensity is indeed calculated by sum of intensity divided by number of pixel / voxel. It does not make any difference if you're doing 2D or 3D (the spot are then 2D or 3D).
To have better separation, you can play with the scale & sensitivity parameters. Sometime using 2 scales together can help but honestly, you will never obtain a 100% perfect detection. You always have to accept some small errors here and there and just repeat the experience enough time to get the statistics good enough :)

Best,

-  Stephane

Yuan ZHOU

unread,
Jul 10, 2015, 3:50:50 AM7/10/15
to icy-so...@googlegroups.com, lilian...@gmail.com
Thanks for your reply, Stephane. 

Lillian

在 2015年6月29日星期一 UTC+8下午4:44:27,Stephane写道:
Reply all
Reply to author
Forward
0 new messages