D.W. Opitz (USA)
Machine learning, feature extraction, neural networks, image processing
Extracting features from digital images is a time consuming but important task. Therefore, much research has been expended in trying to automate this process. This work, however, has not been met with much success. Machine learning offers an exciting alternative to fully automated feature extraction; however, much of this research is still in its infancy. In this paper we provide an empirical comparison of standard learning techniques with the notion of ensembles and show the utility of ensembles. An ensemble consists of a set of individually trained classifiers (such as neural networks or decision trees) whose predictions are combined when classifying novel instances. Previous research has shown that an ensemble is often more accurate than any of the single classifiers in the ensemble. Our results confirm this finding on the task of extracting features from digital imagery.
Important Links:
Go Back
