Timeline for Efficiently detect overlaying ellipses in distorted images
Current License: CC BY-SA 4.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 4 at 8:07 | comment | added | hello_darkness | That seems true, indeed. And this rather heuristic path seems at least more efficient than using Hough Transforms (and I may can implement it, somewhat efficiently, myself). Thanks for your suggestion! | |
| Apr 2 at 14:57 | comment | added | whpowell96 | Yeah it seems difficult regardless but being able to enumerate the ellipses seems like a necessary first step. | |
| Apr 2 at 6:50 | comment | added | hello_darkness | @whpowell96 I'll probably try this route, though I'm a not too sure whether using the tangient vector will work out for the parts with many overlaps | |
| Mar 25 at 18:12 | comment | added | whpowell96 | It won't be robust, but I think some sort of greedy algorithm will be able to at least assign every black pixel to an ellipse. Start on a black pixel, find a neightboring black pixel, then find the next pixel in the rough direction of the tangent vector. This will ensure you jump over intersections. Once every black pixel is assigned to an ellipse, then the problem reduces to determining whether one set of points is contained in the convex hull of another set of points. | |
| Mar 25 at 12:09 | comment | added | hello_darkness | This might be a way, but there are some issues: 1) I have no concrete parameters but could try to estimate them from few examples (e.g. the clear distinction between small&big ellipses) 2) The problem itself feels like their should be a clean, efficient way to solve it analytically/algorithmically. Maybe I'm wrong (?) 3) I'm no expert in AI, but you suggest "generate representative images. Hand them alongside the (in this case known) target area to Deep Learning algorithm X" (?). Even if I find a way find a good way to approximate the images: What should X be? Run it for how long? | |
| Mar 25 at 11:49 | comment | added | Royi | Since you can generate those, using Deep Learning would probably be a good path to take. | |
| Mar 25 at 10:07 | comment | added | hello_darkness | @Royi I added few more examples! Yes, they could be generated by code (which, however, does not change the requirement of visually extracting the desired targets) | |
| Mar 25 at 10:04 | history | edited | hello_darkness | CC BY-SA 4.0 |
added 382 characters in body
|
| Mar 23 at 10:50 | comment | added | Royi | Can you share more examples? You current example can be easily generated by code, is this the case? | |
| Mar 10 at 11:25 | history | edited | hello_darkness | CC BY-SA 4.0 |
added 104 characters in body
|
| Mar 10 at 11:07 | history | edited | hello_darkness | CC BY-SA 4.0 |
deleted 65 characters in body
|
| S Mar 10 at 11:06 | review | First questions | |||
| Mar 11 at 10:14 | |||||
| S Mar 10 at 11:06 | history | asked | hello_darkness | CC BY-SA 4.0 |