Object Detection with Saliency Space for Low Depth of Field Images Using Graph Cut Method
H. P. Narkhede, Payal Mahajan, Jyoti Mahajan · 2014
The graph cut based approach has become very popular for interactive segmentation of the object -of-interest from the background. Content-based multimedia application plays an important role on automatic segmentation of images with low depth of field (DOF).Graph cut method, separate the important objects (i.e. interest regions) of a given image from its defocused background using novel cluster ensemble algorithm and minimal graph cut is constructed using object and background seeds which is based on the max-flow method. The multi-scale reblurring model is used to detect the object-of-interest (OOI) in saliency space. A global energy item related with the saliency map adopted to find the global minimum and a local energy term regarding the low DOF images is used to improve the segmentation precision. Various experiments are performed which shows an average segmentation performance of 97.35% on the 117 test images to extract the ROF in an image.