Detecting complex circular object in 2D images!!!!!!!

2 次查看（过去 30 天）
Fathin
Fathin 2011-7-19
Hi, I have few images, 2D images taken by DSLR camera. The images are so complex for me as I am a beginner. The circular object is small and the poor background contrast. I want to detect almost circular or ellipse object in those image. How do I detect that object in MATLAB? There are few circular objects there to detect. The circular object quite small and some of it is partially hidden by other object. (Detecting apple fruit from tree)
This is the link for real image: http://www.flickr.com/photos/64698236@N03/5891874966/in/photostream/
Here is my testing image(this 1 might be easier): http://www.flickr.com/photos/64698236@N03/5892060924/in/photostream/
I already tried Detecting Circles<http://blogs.mathworks.com/pick/2008/05/23/detecting-circles-in-an-image/> in an Image but the results seem not accurate or does not works well for that image. I did change the radius for the circle. Do I miss any steps? I also tried other commands such as blobs demo, Hough_transform for circle detection.
This is the result image for circle detection; http://www.flickr.com/photos/64698236@N03/5931398663/in/photostream
Someone please help me.Thank you in advance.

请先登录，再回答此问题。

回答（1 个）

Sean de Wolski
Sean de Wolski 2011-7-20
I would recommend trying to better segment the image before applying any circle detection algorithms.
Perhaps a standard deviation filter since the fruit appear to be smoother that their surroundings
doc stdfilt
  2 个评论
Fathin
Fathin 2011-7-20
Thank you for your comment,
what do you mean by standard deviation filter? Is it Gaussian Filter?
Sean de Wolski
Sean de Wolski 2011-7-20
No. It measures the standard deviation of every neighborhood of a certain size in an image. Things with low standard deviation are more homogeneous. In your image, the fruit are pretty homogeneous.

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Semantic Segmentation 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by