Affiliations: [a] Department of Mathematics, Quaid-I-Azam University, Islamabad, Pakistan
| [b] Department of Computer Science, Bahria University, Islamabad, Pakistan
| [c] Department of Mathematics, Preston University, Islamabad, Pakistan
Correspondence:
[*]
Corresponding author. Mahwish Bano, Department of Mathematics, Quaid-I-Azam University, Islamabad, Pakistan. Tel.: +92 519263351; E-mail: [email protected].
Abstract: In continuation of the previous work, we now introduced a new algorithm for text embedding based on scan pattern with XOR and texture synthesis. In this work we have considered natural images and the scan pattern and XOR function in three standalone steps have been used to embed text into patching regions from a sample natural image/video transformed and copied to the output and then stitched together along optimal seams to generate a new output. The size of the patch is not chosen before, but instead a graph cut technique is used to determine the optimal patch region for any given offset between the input and the output texture. The graph cut technique for seam optimization is applicable in two and three dimensions. This graph cut technique of patching into original picture is very much applicable in image processing. We present a new idea to embed text into patching seam using scan pattern with XOR and then placed in the original image. Although this technique is expensive in terms of image area usage but it is a new idea for text embedding and has to be optimized in the sense of original picture usages. We show results for regular, random and the natural images. We also demonstrate how this method can be used interactively.
Keywords: Text embedding, image based randering, image processing, graph cut technique, scan pattern with XOR
