| Paper: | TA-L2.4 |
| Session: | Image and Video Segmentation III: Edge or Color Segmentation |
| Time: | Tuesday, September 18, 10:50 - 11:10 |
| Presentation: |
Lecture
|
| Title: |
MULTI-RESOLUTION LOCAL HISTOGRAM ANALYSIS FOR EDGE DETECTION |
| Authors: |
Amar Aggoun; Brunel University | | |
| | Magid Khallil; Brunel University | | |
| Abstract: |
The objectives of this paper is to present a novel multi resolution edge extraction algorithm, based on processing of the local histograms of small non-overlapping blocks of the output of the first derivative of a narrow 2D Gaussian filter. The proposed edge detection algorithm starts by convolving the image with a narrow 2D Gaussian smoothing filter to minimise the edge displacement, and increase the resolution and detectability. Processing of the local histogram of small non-overlapping blocks of the edge map is carried out to perform an additional noise rejection operation and automatically determine the local thresholds. It is shown that the proposed edge extraction algorithm provides the best trade off between noise rejection and accurate edge localisation and resolution. |
