|
|
 |
| |
|
|
|
A Novel Image Classification Model Based on Contourlet Transform and Dynamic Fuzzy Graph Cuts |
|
|
|
PP: 93S-97S |
|
|
|
Author(s) |
|
|
|
Zhang Guang-Ming,
Cui Zhi-Ming,
|
|
|
|
Abstract |
|
|
| The contourlet transform as a time-frequency and multiresolution analysis tool is often
used in the domain of image processing, meanwhile graph cuts as an increasingly important tool for
solving a number of energy minimization problems in computer vision and other fields. By analyzing
the characters of monitor image, this paper proposes a novel image classification model which is
combining contourlet transform and graph cuts theory. Firstly, the image was decomposed by
contourlet transform to obtain the different subbands coefficients. Then the entropy from certain
subband was calculated, and a fuzzy energy function based on graph cuts theory and dynamic fuzzy
theory was constructed to adjust the threshold of critical entropy. At last a model was constructed to
classify traffic images. It could alert for the ratio of road occupancy to traffic control. |
|
|
|
|
|
|
|
