This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
Looking at a binary image, which is in fact an image of 0 and 1 values - black and white regions, you can define contours as the edges around the white regions inside the complete image. Contours can help to define the largest blobs/areas in the binary image, which can then be used to calculate parameters of these regions like area, bounding box, centerpoint, ...
These parameters in turn can be used to pass to higher level algorithms like for example object detection or object recognition.
Contours are mainly used for segmenting / retrieving information that was gathered from an image, using specific operators.
This paper is used for implementing contour functionality in OpenCV.
