A survey on matching strategies for boundary image comparison and evaluation. (July 2021)

• Record Type: Journal Article
• Authors: Lopez-Molina, C.; Marco-Detchart, C.; Bustince, H.; De Baets, B.
• Is Part Of: Pattern recognition. Volume 115(2021)
• Publication Date: 2021-07
• Subjects: Boundary image -- Boundary evaluation -- Linear feature matching -- Image comparison; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2021.107883
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We analyze boundary evaluation methods, focusing on those based on (a) boundary matching and (b) analysis of confusion matrices, which are the current most popular trend. We review literature in boundary matching, describing the constraints and goals holding for different tasks. We propose a taxonomy for boundary matching methods in the context of boundary evaluation. We design a novel experimental framework to compare the potential differences between boundary image comparison methods (in general), and boundary matching strategies (in particular). We carry out extensive experiments to quantitatively and qualitatively analyze the similarities and divergences in the results by different boundary matching methods. Abstract: Most of the strategies for boundary image evaluation involve the comparison of computer-generated images with ground truth solutions. While this can be done in different manners, recent years have seen a dominance of techniques based on the use of confusion matrices. That is, techniques that, at the evaluation stage, interpret boundary detection as a classification problem. These techniques require a correspondence between the boundary pixels in the candidate image and those in the ground truth; that correspondence is further used to create the confusion matrix, from which evaluation statistics can be computed. The correspondence between boundary images faces different challenges, mainly related to the matching of potentially displaced boundaries. Interestingly, boundary image comparison relates to many other fields of study in literature, from object tracking to biometrical identification. In this work, we survey all existing strategies for boundary matching, we propose a taxonomy to embrace them all, and perform a usability-driven quantitative analysis of their behaviour.