Interactive image segmentation with a regression based ensemble learning paradigm

• Published in: Frontiers of information technology & electronic engineering Vol. 18; no. 7; pp. 1002 - 1020
• Main Authors: Zhang, Jin, Tang, Zhao-hui, Gui, Wei-hua, Chen, Qing, Liu, Jin-ping
• Format: Journal Article
• Language: English
• Published: Hangzhou Zhejiang University Press 01.07.2017; Springer Nature B.V
• Subjects: Clustering; Communications Engineering; Computer Hardware; Computer Science; Computer Systems Organization and Communication Networks; Electrical Engineering; Electronics and Microelectronics; Ensemble learning; Image segmentation; Instrumentation; Networks; Regression models; Robustness (mathematics); Semi-supervised learning; Support vector machines; Thin plates; Support vector regression (SVR); Multivariate adaptive regression splines (MARS); Semi-supervised learning; Interactive image segmentation; Thin-plate spline regression (TPSR); Ensemble learning; TP391.4
• ISSN: 2095-9184; 2095-9230
• DOI: 10.1631/FITEE.1601401

To achieve fine segmentation of complex natural images, people often resort to an interactive segmentation paradigm, since fully automatic methods often fail to obtain a result consistent with the ground truth. However, when the foreground and background share some similar areas in color, the fine segmentation result of conventional interactive methods usually relies on the increase o f manual labels. This paper presents a novel interactive image segmentation method via a regression-based ensemble model with semi-supervised learning. The task is formulated as a non-linear problem integrating two complementary spline regressors and strengthening the robustness of each regressor via semi-supervised learning. First, two spline regressors with a complementary nature are constructed based on multivariate adaptive regression splines （MARS） and smooth thin plate spline regression （TPSR）. Then, a regressor boosting method based on a clustering hypothesis and semi-supervised learning is proposed to assist the training of MARS and TPSR by using the region segmentation information contained in unlabeled pixels. Next, a support vector regression （SVR） based decision fusion model is adopted to integrate the results of MARS and TPSR. Finally, the GraphCut is introduced and combined with the SVR ensemble results to achieve image segmentation. Extensive experimental results on benchmark datasets of BSDS500 and Pascal VOC have demonstrated the effectiveness of our method, and the com- parison with experiment results has validated that the proposed method is comparable with the state-of-the-art methods for in- teractive natural image segmentation.