I have images as showing below (Figure 1) and I would like to fit two ellipses to the inner and outer shape.

I have some experience of openCV, but really having problems detecting contours in the images below. I am able to detect some of the black circles shown, but these are dirt on the sensor, and I really want to detect the tracks in the images, as outlined in red in the second image.

I want to detect corners from a image with boxes, although i created the chessboard edge lines with the EDlines algorithm. Now, I have some problems to join them to create perfect boxes. Could you help me maybe?

I have two images. One is reference image (Fig.1) and another one is test images (Fig.2).
I like to make changes to make test images closer to reference image in terms of image size and rotation. So that I can compare the two image for differences.

Fig.1