Abstract:

Based on image strip dividing, an effective and fast image retargeting algorithm is proposed for resizing images. First, we construct the image energy map using gradient magnitude of the pixels and calculate the accumulated energy of each column, dividing the image into several strips by integrating similar energy columns. The reduced amount of dimension is decided in inverse proportion to the average energy for each strip. Then we retarget the image combining scaling with cropping in terms of each strip’s reduced ratio. Experiment results show that the proposed algorithm is capable of implementing fast image retargeting and preserving both the local structures and the global visual effect of the image.

Key words: image retargeting, strip dividing, energy map, scaling, cropping