An Efficient Hybrid Sequence of Retargeting Operators to Minimize Structural Deformities in Image

Researchers have proposed several techniques to alter image aspect ratios in order to preserve visual information while retargeting. However, researchers discovered that no single retargeting approach consistently produces the optimum outcomes. As a result, several retargeting operators are combined to form a hybrid sequence that takes use of the current algorithms. In this paper a hybrid set of image retargeting operators is proposed that combines five different retargeting operators. The addition of a switching point among these operators is an attempt to maximize the benefits of each algorithm via the retargeting operation. In addition, an image morphological procedure is used to minimize abnormalities in crucial image regions and increase the effectiveness of the suggested approach. To identify prominent parts in the image, an importance map is created by integrating a saliency map, gradient map, edge map, skin map, shadow map, and object map. These maps are generated using MATLAB's built-in functions and libraries. The suggested technique's effectiveness is tested using a structural similarity index measure (SSIM) that is based on three parameters: luminance comparison, contrast comparison, and structure comparison. The suggested hybrid sequence's superiority is compared with single image retargeting operators using the generated SSIM index. The SSIM index results show that the suggested technique achieves 86.51% similarity between the original and retargeted images.