Image resizing is one of the basic operations for image and video display, transmission, and analysis, etc. Traditional image resizing methods usually utilize an interpolation algorithm to zoom in or zoom out an image, e.g., the nearest neighborhood interpolation, the bilinear interpolation and the cubic interpolation are most widely used ones. The interpolation based methods don't pay attention to the content of an image and so they may make the image be seen unnaturally. The famous seam carving based image resizing method reserves the high energy part of an image during the image resizing process, so it is content aware and has better performance than the interpolation based methods. But drawbacks still exist in the former seam carving based image resizing methods proposed by researchers. In this paper, a new improved seam carving image resizing method is proposed. In the proposed improved method, a new combined energy is utilized which can better describe the content of the image. Optimization methods are also provided to accelerate the algorithms. Experimental results show that the proposed method is effective.
展开▼
