Super-pixel and Neighborhood Based Contour Detection

Devshri Swami and Bahrti Chaurasia

Scope College of Engineering, Bhopal, M. P., INDIA.


This paper proposes an increased modularity created contour detection algorithm based on super-pixel and neighborhood pixels. Given an over segmented image that entails of many small regions, our algorithm automatically combines those neighboring regions that produce the largest increase in modularity index. When the modularity of the segmented image is increased, the method stops merging and produces the final segmented image based on super-pixel and neighborhood region pixels. To preserve the repetitive patterns in a homogeneous region, we propose a feature on the basis of the histogram of states of image gradients and use it together with the color feature to characterize the similarity of two regions. By building the similarity matrix in an adaptive manner, the over segmentation problem can be successfully avoided.

Keywords :Clustering, community detection, image Contouring, modularity, contour detection.

map1 map2 map3