Abstract
Image resizing techniques based on image content have been extensively studied. Most of these approaches evaluate the importance of each pixel to preserve more critical regions and discard less critical regions for achieving a target image size. This study proposes an image resizing algorithm based on strengthened edges. First, we created two feature maps: an improved gradient map and an enhanced saliency map. We then computed a strengthened edge map on the basis of these two feature maps. Finally, the target image was constructed using a seam carving method on the basis of the strengthened edge map. More accurate results can be obtained using the proposed approach rather than previous methods.
Acknowledgment
The authors would like to thank the Ministry of Science and Technology of Taiwan for financially supporting this research under Contract No. MOST 104-2221-E-239-025-.
Disclosure statement
No potential conflict of interest was reported by the authors.