Content-aware image retargeting technique and iterated function system: frameworks, applications, and possible future advancements

The objective of the image retargeting operator is to reduce the size of the image in a content-aware fashion. To strengthen the process of single operator-based image retargeting different hybrid sequences are being investigated. The visual quality of fractal images can be improved by utilizing efficient retargeting operators. In this paper, three frameworks are developed and combined to minimize the deficiencies of image retargeting operations. In framework-1, a seam diversion-based image retargeting technique is developed to overcome the drawbacks of the existing seam carving (SC) technique. In framework-2, the technique developed in framework-1 is combined with other retargeting operators to produce an efficient hybrid sequence of image retargeting operators. The objective of framework-2 is to overcome the limitation of single operator based image retargeting. In framework-3, an improved affine transformation function is developed that incorporates hybrid sequence of operator proposed in the framework-2. The main objective of framework-3 is to replace the scaling operator from the existing affine transformation. The structural similarity index measure (SSIM) metric is used to measure the performance of framework-1. The comparative analysis reveals that the performance of the retargeting technique proposed in framework-1 is 7% higher as compared to the existing SC technique. The superiority of framework-2 and 3 is analyzed using an optimization function in which Image Euclidian Distance (IMED) and Dominant Color Descriptor (DCD) techniques are incorporated. Finally, the performance analysis of framework-2 and 3 discloses their higher efficacy as compared to existing state-of-the-art.