| Paper: | TA-L2.6 |
| Session: | Image and Video Segmentation III: Edge or Color Segmentation |
| Time: | Tuesday, September 18, 11:50 - 12:10 |
| Presentation: |
Lecture
|
| Title: |
COORDINATE LOGIC TRANSFORMS AND THEIR USE IN THE DETECTION OF EDGES WITHIN BINARY AND GRAYSCALE IMAGES |
| Authors: |
Ethan Danahy; Tufts University | | |
| | Karen Panetta; Tufts University | | |
| | Sos Agaian; University of Texas at San Antonio | | |
| Abstract: |
This paper introduces coordinate logic (CL) transforms as an alternative method for calculating coordinate logic (CL) filters. Additionally, a new measure and detection technique are introduced, enhancing the capabilities of the basic CL transform for the application of detecting edges within 2D signals (images). Applicable to binary and grayscale images, computer simulations demonstrate the success of this improved procedure on two classes of signals: synthetic (edge maps are know) and natural (edge maps are unknown). Results are evaluated quantitatively (via Pratt’s figure of merit) and compared visually to two common edge detection techniques. |
