Improved Seam Carving For Image Resizing Using GrowCut

In Image Processing the resizing of images is very important, especially when we improve the image condition.Image resizing is an important problem in digital image processing. Images often need to be resized spatially in order to ensure that they are tailored to the communications networks over which they travel and to the end-user display devices upon which they will be presented. Image resizing is the process of make the image small in it is size to be suitable to any design or application.One of the most notable image resizing techniques is Seam carving. Seam carving is an image processing operator for content-aware image resizing. However, it sometimes fails to protect important objects in the image. In this thesis, a new interactive image resizing technique is proposed, which combines seam carving and GrowCut image segmentation algorithm to perceptually improve the visual effects after image resizing.GrowCut is utilized to select the objects in the image for retaining or removal. An intuitive and easy user interface is also supported because user just needs to draft some trails inside and outside the object to select the Regions of interest (ROIs) automatically for object removal.The new technique is using for image resizing in both reduction and expansion. By combining Seam Carving and GrowCut (cute segmentation), we protect or remove the region of interest by telling the image energy function to make new map of energy to guide the seam to be an optimal seam and improve the image resizing condition.Simulation results demonstrate that the proposed approach can easily remove or protect the region of interest in the image.