melongex's profile - activity

Which feature detector

http://docs.opencv.org/modules/features2d/doc/feature_detection_and_description.html#ORB : public Feature2D

do you suggest for my images?

here two examples http://imageshack.us/a/img51/8590/8wlo.png

http://imageshack.us/a/img542/5932/n6a3.png

as I said I am merely interested in the rotation and position of these shapes (I have given polygons, which are rotated differently)

No not really, I am dealling with real images and the contours found in this images are not as perfect as in this tutorial. I really think that I need some kind of matching algorithmn for this problem

Hi guys"

I have to following issue:

I have an arbitrary polygon (defined with points A,B,C,D,...) and I want to find the position of this polygon in "real" images: In these images the polygons are not ideal geometric shapes, but physical objects.

For instance the polygon might be a right-angled triangle and I then want to detect "set squares" in an image + its rotation.

Do you know any solutions to this problem? I already tried to do this with SURF and SIFT where I would simply paint an artifical image with my polygon and extracting features of this artifical image, however the extracted feature do not seem to work quite well. Template matching with an artifical image is not working either, because the objects in the real images can be rotated.

Hi!

I am using a grid of filled circles in order to calibrate my camera (get rid of objective distortions) However, since I know the physical distance and size of the grid, it has to be possible to calculate the pixel size of my objective + camera (conversion pixel to millimeter);

is there a direct way to access this information with either the camera matrix or the output of cv::calibrateMatrixValues?

Cheers

push. .

Thanks for your suggestion, however I don't really know how to get The Polygons rotation etc. with findContours. Can I use my contour points as keypoints for SURF an The match these with my polygon?

Hallo,

So in an application I'm writing I have to detect polygons in a scene. The polygons are the contours of some objects (I simply illuminate my objects using a backlight), such that the object is black and the background is grey.

There's always only one object present in each image at a time and I know what polygon I should look for (All points of the polygon and therefore its edges are given)

What I don't know is firstly the location of the object in the image (e.g. its contour) and the rotation of the object.

Is there any way to detect this with opencv. What I tried so far is using some of the FeatureDetectors. The do not seem to work properly with my grayscale images since they don't detect so much interesting features, probably because my images are almost binary.

I'd love to simply use a canny filter on the image to get all the edges and match those with my polygon, but I don't really know how to do this.

Do you have any suggestions?

Thanks a lot for your ideas! I'll try if it is fast enough if I do it like this. Cheers!

I already tried creating a new matrix and putting my dots there, I didn't feel so good about doing this. I already have the dots, so it simply makes no sense for doing a for loop on an almost empty matrix. For small images it is okay of course, but for a 10 megapixel image it is probably not a so good idea. Of course ai could simply write my own line hough, but as this transformation is pretty common, I was hoping for a fast and robust implementation, which I could simply use. I already thought about a simple line regression, but I indeed want to use the linehough first to kill dots, which a too great distance from the 'best' line, and do the line regression afterwards.

So, do you know any solution? How are you detecting lines in large images in real time (for me the opencv implementation is already slow at about 5 megapixel)

Hi guys!

For some measurements I want to perform with OpenCV the canny algorithm in HoughLines seems just be too slow. However, I can also use a much easier approach for finding edges. Namely, I create search rays and then, take a profile across those ray, searching for maxima. The benefit of doing so is that I can set the resolution of the search rays: For rough searching it's just enough to have those rays each few pixels or so.

http://imageshack.us/a/img87/7904/searchrays.png

(Obviously, a plain houghlines would do it for this image, so please no comments about that) As you see in the image, I find points around edges (blue dots). However, now I want to find good candidates for lines for these points. Unfortunately, HoughLines NEEDS a greyscale image as input rather than taking points with x,y coordinates.

Is there any solution to find hough lines for an array of 2d points within opencv (I rather not implement my own hough transformation)?