http://imagej.273.s1.nabble.com/Measuring-images-in-screen-pixels-tp5005667p5005708.html
currently the ImageJ macro language does not support ovals with subpixel resolution, neither for drawing nor as overlays.
I would suggest using short lines perpendicular to the long line, i.e., small tick marks. You can have these in the ROI Manager (show all) or as overlayer, both with subpixel resolution.
By the way, please make sure that you update ImageJ to the last version; the shift between the lines and ovals reminds me of a problem prior to version 1.48e.
> This is perfect for the lines, thank you. However, I put in a macro to mark the end and quarter points of the line segment when I hit "draw" but this still puts them too large and often off-centered on the image rather than the overlay. I'm terrible with making macros, but I figure the general idea is already there. Is there any way to get it to go to the overlay?
>
> This is the macro I am using:
>
> // The macro "quarters" draws dots along a line selection
> // at the beginning, 1/4 way point, 1/2 way point, 3/4 point, and ending of the line.
>
>
> macro "quarters [q]" {
>
>
> getLine(x1,y1,x5,y5,width);
>
>
> // Make oval size and line width a function of line length
> if(x1==x5) {
> orient = "v";
> length = abs(y5-y1);
> ovwit = length/35;
> }
>
>
> if(y1==y5) {
> orient = "h";
> length = abs(x5-x1);
> ovwit = length/55;
> }
>
>
> y2 = y1 + ((y5-y1)/4);
> y3 = y1 + ((y5-y1)/2);
> y4 = y5 - ((y5-y1)/4);
>
>
> x2 = x1 + ((x5-x1)/4);
> x3 = x1 + ((x5-x1)/2);
> x4 = x5 - ((x5-x1)/4);
>
>
> // Adjust coordinates for the fillOval function
> adj = ovwit/2;
> x1 = x1 - adj;
> x2 = x2 - adj;
> x3 = x3 - adj;
> x4 = x4 - adj;
> x5 = x5 - adj;
>
>
> y1 = y1 - adj;
> y2 = y2 - adj;
> y3 = y3 - adj;
> y4 = y4 - adj;
> y5 = y5 - adj;
>
>
> setTool("brush");
> setColor("magenta");
> setLineWidth(3);
> fillOval(x1,y1,ovwit,ovwit);
> fillOval(x2,y2,ovwit,ovwit);
> fillOval(x3,y3,ovwit,ovwit);
> fillOval(x4,y4,ovwit,ovwit);
> fillOval(x5,y5,ovwit,ovwit);
> setLineWidth(1);
> setTool("line");
>
>
> }
>
>
> Date: Mon, 25 Nov 2013 06:42:13 -0800
> From:
[hidden email]
> To:
[hidden email]
> Subject: Re: Measuring images in screen pixels
>
>
>
> Hi Elkee,
>
>
> you can use an overlay instead of drawing. The overlay can have a line thickness of "zero", which means as thin as possible at any magnification.
>
> Image>Overlay>Add Selection. Hold the alt key to set the properties of the overlay.
>
>
> Another (similar) approach is saving the lines to the ROI Manager instead of drawing them, and using 'show all' in the Roi Manager.
>
>
> To save the image with an overlay or rois, save it as tiff or zip.
>
>
> Michael
>
> ________________________________________________________________
>
> On Nov 25, 2013, at 02:16, Elkee wrote:
>
>
>> I guess I did not explain this well. I tried to get a real picture to show
>
>> you what I am talking about. I am dealing with ratios, so I do not care
>
>> whether or not I am measuring in screen pixels or image pixels, though I
>
>> think screen pixels would allow for finer measurements. However, I find
>
>> when I draw the segment and do not click "draw" on the I have very fine
>
>> lines and the endpoints and center point are also a great size and re
>
>> integrated with the line as the center. I assume this is because the tool
>
>> is using screen pixels. However, when I click to have the segment drawn on
>
>> the image the line and points I am measuring from get much thicker because
>
>> they fill a pixel on the image. This works fine with the larger bacteria,
>
>> but with the smaller bacteria my image becomes almost completely filled in
>
>> with lines (as you can see in the picture), making it difficult to see the
>
>> bacteria. However, if I could base my line segments off of screen pixels,
>
>> as is done before I click "draw" to the right of the smaller measured
>
>> bacteria, I can see what I am doing much better and be far more accurate
>
>> when I am doing my 2nd, 3rd, and 4th measurement. Plus, I would never have
>
>> an off set as seen in the larger bacteria.
>
>>
>
>> Maybe there is a way to use multiple segment tools and measure rather then
>
>> draw on the actual image? Or select to measure and mark screen pixels
>
>> rather than image pixels?
>
>>
>
>> I hope this clears it up slightly better. Thank you for your help
>
> --
>
> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>
>
>
>
>
>
>
>
>
>
>
> If you reply to this email, your message will be added to the discussion below:
>
http://imagej.1557.x6.nabble.com/Measuring-images-in-screen-pixels-tp5005667p5005691.html>
>
>
> To unsubscribe from Measuring images in screen pixels, click here.
>
> NAML
>
>
>
>
> --
> View this message in context:
http://imagej.1557.x6.nabble.com/Measuring-images-in-screen-pixels-tp5005667p5005698.html> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html