Salient object detection via effective background prior and novel graph
- PDF / 2,650,095 Bytes
- 17 Pages / 439.37 x 666.142 pts Page_size
- 67 Downloads / 196 Views
Salient object detection via effective background prior and novel graph Yu Pang 1 & Yunhe Wu 2 & Chengdong Wu 1 & Ming Zhang 3 Received: 29 July 2018 / Revised: 27 April 2020 / Accepted: 15 June 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Salient object detection is getting more and more attention in computer vision field. In this paper, we propose a novel and effective framework for salient object detection. Firstly, we develop a robust background-based map by using spatial prior to remove the foreground noises of image boundary regions. The proposed background-based map and Objectness map are integrated to obtain a coarse saliency map. Then, an effective saliency propagation mechanism is utilized to further highlight salient object and suppress background region by defining a novel graph model, each node connects to its more similar neighbors and nodes with low saliency values in the proposed graph. As a result, the coarse saliency map is optimized to the refined saliency map by novel graph based saliency propagation. Finally, we construct a novel integration framework to further integrate two saliency maps for performance improvement. Experiments on three benchmark datasets are tested, experimental results show the superiority of the proposed algorithm than other state-of-the-art methods. Keywords Saliency object detection . Background prior . Saliency propagation . Novel graph structure . Integration mechanism
1 Introduction With the development of image processing technology, we can obtain quickly a large number of image data, this is a challenge for the efficiency of image processing algorithms. * Chengdong Wu [email protected] * Ming Zhang [email protected]
1
Faculty of Robot Science and Engineering, Northeastern University, Shenyang 110169, China
2
College of Information Science and Engineering, Northeastern University, Shenyang 110819, China
3
School of Information Science and Technology, Northeast Normal University, Changchun 130117, China
Multimedia Tools and Applications
Fortunately, visual attention mechanism indicates that human eye can select quickly the most interesting object and ignore useless information in a scene, saliency detection task aims to identify these important regions. In recent years, saliency detection becomes gradually a hot topic in computer vision field, more and more researchers focus on this problem. As important pre-processing procedure, saliency detection contributes to numerous computer vision and image processing tasks, such as compression [10, 38], visual tracking [17], object recognition [5, 22], segmentation [35] and so on. Existing saliency detection methods are divided into two categories: bottom-up methods [4, 25] and top-down methods [13, 29, 33]. Bottom-up methods are driven by visual stimulation, their performances rely on various low-level visual features extraction, e.g., color feature, texture, orientation and so forth. In contrast, top-down methods are driven by specific tasks, supervised learning fr
Data Loading...
