Landmark matching based retinal image alignment by enforcing sparsity in correspondence matrix

• Published in: Medical image analysis Vol. 18; no. 6; pp. 903 - 913
• Main Authors: Zheng, Yuanjie, Daniel, Ebenezer, Hunter, Allan A., Xiao, Rui, Gao, Jianbin, Li, Hongsheng, Maguire, Maureen G., Brainard, David H., Gee, James C.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.08.2014
• Subjects: Algorithms; Humans; Image alignment; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Pattern Recognition, Automated - methods; Point matching; Programming, Linear; Reproducibility of Results; Retina - anatomy & histology; Retinal image; Retinoscopy - methods; Sensitivity and Specificity; Sparsity; Subtraction Technique; Transformation; Image alignment; Transformation; Retinal image; Sparsity; Point matching
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2013.09.009

[Display omitted] •A novel landmark matching formulation by enforcing sparsity in the correspondence matrix.•Joint estimation of correspondences & transformation model+softassign strategy+combinatorial optimization.•A set of reinforced self-similarities descriptors to better characterize retinal image. Retinal image alignment is fundamental to many applications in diagnosis of eye diseases. In this paper, we address the problem of landmark matching based retinal image alignment. We propose a novel landmark matching formulation by enforcing sparsity in the correspondence matrix and offer its solutions based on linear programming. The proposed formulation not only enables a joint estimation of the landmark correspondences and a predefined transformation model but also combines the benefits of the softassign strategy (Chui and Rangarajan, 2003) and the combinatorial optimization of linear programming. We also introduced a set of reinforced self-similarities descriptors which can better characterize local photometric and geometric properties of the retinal image. Theoretical analysis and experimental results with both fundus color images and angiogram images show the superior performances of our algorithms to several state-of-the-art techniques.