<div dir="ltr"><div><div><div><div><div><div><div>Hi,<br></div>So far I have not been able to find a function that does what the description describes. The closest was PadByRadius from ImageRegion, however this pads uniformly in any one dimension. <br>
<br>Therefore I started to think about the steps necessary to recreate the method. This is what I have come up with so far, I would appreciate any feedback you might have. <br><br></div><div>0. Original image of the eye (<a href="http://original.im">original.im</a>)<br>
</div>1. Threshold image to create binary image (<a href="http://binary.im">binary.im</a>) (0 = outside/background, 1 = inside field of view)<br></div>2. Dilate binary image by 1 pixel (<a href="http://dilated.im">dilated.im</a>)<br>
</div>3. Subtract binary image from dilated image (<a href="http://dilated.im">dilated.im</a> - <a href="http://binary.im">binary.im</a>) this creates a one pixel width boarder around the field of view FOV (<a href="http://boarder.im">boarder.im</a>)<br>
</div>4. For every pixel (i) = 1 in the boarder image find the 8 neighborhood pixels in both the binary image (<a href="http://binary.im">binary.im</a>) and the original (<a href="http://original.im">original.im</a>) and mutliply each corresponding neighborhood pixel together. This leaves only the grey-values of the pixels inside the FOV and removes the grey-values of the background.<br>
</div>5. Find the average of these multiplied 8 pixels and assign this average to the current pixel (i),<br></div><div>6. Repeat for remaining pixels inside the boarder<br><br></div><div>The whole process is then repeated until a large enough boarder is created around the FOV of the eye. <br>
<br></div><div>Can this be done more efficiently or simply?<br><br></div><div>I look forward to thoughts on this,<br></div><div>much appreciated,<br></div><div>Jesse<br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">
On Wed, Jun 5, 2013 at 1:19 PM, Jesse Ross-Jones <span dir="ltr"><<a href="mailto:jesse.rj@gmail.com" target="_blank">jesse.rj@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr"><div>Dear ITK Users and Devs,<br><br></div><div>I am working on vessel segmentation, using the STARE and DRIVE image datasets, and in order to deal with the image boundary I am trying to implement the following:<br>
</div><div><div><br>The preprocessing algorithm starts with a region of interest<br>(ROI) determined by the camera’s aperture and iteratively grows<br>this ROI. Each step of the algorithm consists in the following.<br>First, the set of pixels of the exterior border of the ROI is de-<br>
termined, i.e., pixels that are outside the ROI and are neighbors<br>(using four-neighborhood) to pixels inside it. Then, each pixel<br>value of this set is replaced with the mean value of its neighbors<br>(this time using eight-neighborhood) inside the ROI. Finally, the<br>
ROI is expanded by inclusion of this altered set of pixels. This<br>process is repeated and can be seen as artificially increasing the<br>ROI, as shown in Fig. 1(b).<br><br></div><div>from <a href="http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=01677727&tag=1" target="_blank">http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=01677727&tag=1</a><br>
<br></div><div>Is there something similar already implemented? I am considering instead to use the Wrap Pad Image Filter: <a href="http://www.itk.org/Doxygen/html/classitk_1_1WrapPadImageFilter.html" target="_blank">http://www.itk.org/Doxygen/html/classitk_1_1WrapPadImageFilter.html</a><br>
</div><div>to pad the region around the eye. Would this be a better possibility?<br><br></div><div>Much Appreciated,<br></div><div>Jesse<br></div><div><div dir="ltr" style="font-size:19.3939px;font-family:serif"><img alt=""><br>
</div></div></div></div>
</blockquote></div><br></div>