Rev | Line | |
---|
[77] | 1 | |
---|
| 2 | Implementation of the segmentation algorithm described in: |
---|
| 3 | |
---|
| 4 | Efficient Graph-Based Image Segmentation |
---|
| 5 | Pedro F. Felzenszwalb and Daniel P. Huttenlocher |
---|
| 6 | International Journal of Computer Vision, 59(2) September 2004. |
---|
| 7 | |
---|
| 8 | The program takes a color image (PPM format) and produces a segmentation |
---|
| 9 | with a random color assigned to each region. |
---|
| 10 | |
---|
| 11 | 1) Type "make" to compile "segment". |
---|
| 12 | |
---|
| 13 | 2) Run "segment sigma k min input output". |
---|
| 14 | |
---|
| 15 | The parameters are: (see the paper for details) |
---|
| 16 | |
---|
| 17 | sigma: Used to smooth the input image before segmenting it. |
---|
| 18 | k: Value for the threshold function. |
---|
| 19 | min: Minimum component size enforced by post-processing. |
---|
| 20 | input: Input image. |
---|
| 21 | output: Output image. |
---|
| 22 | |
---|
| 23 | Typical parameters are sigma = 0.5, k = 500, min = 20. |
---|
| 24 | Larger values for k result in larger components in the result. |
---|
| 25 | |
---|
Note: See
TracBrowser
for help on using the repository browser.