Identifying wings

Here’s a challenge for you all. We’ve been investigating the role of a tumor supressor (lgl or lethal giant larvae) and its interactions with some microRNA’s. We’ve created flies where the tumor supressor levels are knocked down in the wing which produces various defects (the most severe of which being holes in the wing). I’ve tried unsuccessfully to get CP to measure the wing sizes and perhaps it is due to low contrast images. However, I thought I’d post them here just to see if anyone wanted to take a crack at it. If CP could identify the wings the next step would be to identify the holes and spots.

Hi Scott,

I’m attaching a pipeline which should get you started. Tophat filtering is used to render the image amenable for wing detection and some illumination correction is applied to make segmentation based on the veins easier (some tweaking is still needed, though).

2011_10_06.cp (7.9 KB)