Abstract

This Letter proposes a rapid method for automatic salient object detection inspired by the idea that an image consists of redundant information and novelty fluctuations. We believe object detection can be achieved by removing the nonsalient parts and focusing on the salient object. Considering the relation between the composition of the image and the aim of object detection, we constructed what we believe is a more reliable saliency map to evaluate the image composition. The local energy feature is combined with a simple biologically inspired model (color, intensity, orientation) to strengthen the integrity of the object in the saliency map. We estimated the entropy of the object via the maximum entropy method. Then, we removed pixels of minimal intensity from the original image and compute the entropy of the resulting images, correlating this entropy with the object entropy. Our experimental results show that the algorithm outperforms the state-of-the-art methods and is more suitable in real-time applications.

© 2013 Optical Society of America

Full Article  |  PDF Article

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Figures (4)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Tables (2)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Equations (8)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Metrics

You do not have subscription access to this journal. Article level metrics are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription