Watershed Based Edge Detection in Noisy Images

G.I. Lee, I.-K. Kim, D.-W. Jung, J.-H. Song, W.-G. Kwak, and R.-H. Park (Korea)


Edge detection; Water-flow model; Locally adaptive threshold; Computer vision


We present an edge detection method based on water flow model and gradient information. The water-flow process emphasizes the local edge characteristics and adaptive thresholding to water filled regions leads to locally adaptive edge detection. Experiments show that the proposed method is effective for multimodal edge detection in noisy images compared to conventional derivative-based edge detectors.

Important Links:

Go Back