Abstract
This paper presents a comparative study on five feature selectionheuristics applied to a retinal image database called DRIVE. Features are chosenfrom a feature vector (encoding local information, but as well informationfrom structures and shapes available in the image) constructed for each pixel inthe field of view (FOV) of the image. After selecting the most discriminatory features,an AdaBoost classifier is applied for training. The results of classificationsare used to compare the effectiveness of the five feature selection methods.
Lingua originale | English |
---|---|
Pagine | 655-662 |
Numero di pagine | 8 |
Stato di pubblicazione | Published - 2009 |
All Science Journal Classification (ASJC) codes
- ???subjectarea.asjc.2600.2614???
- ???subjectarea.asjc.1700.1700???