An Improved and Optimized Content-Aware Resizing Algorithm for Images with Densely Situated Foreground Objects

An image in general consists of a combination of significant objects in the foreground and not-so-significant objects in the background. Content aware image resizing or seam carving is a process of resizing an image while maintaining the significant objects (the foreground) in proper visual saliency. The standard algorithms, however, often generate unpredictable distortions in images with densely situated foreground objects. The optimized content aware image resizing (OCAIR) algorithm presented herein, uses iterative graph cuts and edge detection to generate an energy map based on the important sections of the image, so that the resized image does not exhibit unpredictable artefacts. An improved energy map generation algorithm is designed here, which not only marks out the important foreground elements quicker than previously available techniques, but also uses that information to quantity the amount of distortion (if any) that might take place after adding or deleting seams by means of calculating a distortion factor. The process being considerably faster than previous algorithms, allows precise modifications to the input parameters to obtain a well-doctored final image.

[1]  Tony F. Chan,et al.  Active contours without edges , 2001, IEEE Trans. Image Process..

[2]  O. P. Verma,et al.  Newtonian Gravitational Edge Detection Using Gravitational Search Algorithm , 2012, 2012 International Conference on Communication Systems and Network Technologies.

[3]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.

[4]  Reiner Lenz,et al.  Modified Gradient Search for Level Set Based Image Segmentation , 2013, IEEE Transactions on Image Processing.

[5]  Tanmoy Dasgupta,et al.  An improved content aware image resizing algorithm based on a novel adaptive seam detection technique , 2015, 2015 International Conference on Advances in Computing, Communications and Informatics (ICACCI).

[6]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[7]  Ariel Shamir,et al.  Seam Carving for Content-Aware Image Resizing , 2007, ACM Trans. Graph..

[8]  Vitoantonio Bevilacqua,et al.  Improving a genetic algorithm segmentation by means of a fast edge detection technique , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[9]  Harry Shum,et al.  To appear in the ACM SIGGRAPH conference proceedings Drag-and-Drop Pasting , 2022 .

[10]  Andrew Blake,et al.  "GrabCut" , 2004, ACM Trans. Graph..