Loughborough University
Leicestershire, UK
LE11 3TU
+44 (0)1509 263171
Loughborough University

Loughborough University Institutional Repository

Please use this identifier to cite or link to this item: https://dspace.lboro.ac.uk/2134/20269

Title: Wrinkle detection using hessian line tracking
Authors: Ng, Choon-Ching
Yap, Moi Hoon
Costen, Nicholas
Li, Baihua
Keywords: Wrinkle detection
Hessian filter
Line tracking
Bosphorus dataset
Jaccard similarity index
Issue Date: 2015
Publisher: © IEEE
Citation: NG, C.-C. ... et al, 2015. Wrinkle detection using hessian line tracking. IEEE Access, 3, pp.1079-1088
Abstract: Wrinkles play an important role in face-based analysis. They have been widely used in applications such as facial retouching, facial expression recognition and face age estimation. Although a few techniques for wrinkle analysis have been explored in the literature, poor detection limits the accuracy and reliability of wrinkle segmentation. Therefore, an automated wrinkle detection method is crucial to maintain consistency and reduce human error. In this paper, we propose Hessian line tracking (HLT) to overcome the identified problem. The proposed HLT is composed of Hessian seeding and directional line tracking. It is an extension of a Hessian filter; however it significantly increases the accuracy of wrinkle localization when compared with existing methods. In the experimental phase, three coders were instructed to annotate wrinkles manually. To assess the manual annotation, both intra- and inter-reliability were measured, with an accuracy of 94% or above. Experimental results show that the proposed method is capable of tracking hidden pixels; thus it increases connectivity of detection in between wrinkles, allowing some fine wrinkles to be detected. In comparison to the stateof- the-art methods such as the Cula method, Frangi filter, and hybrid Hessian filter, the proposed HLT yields better results, with an accuracy of 84%. This work demonstrates that HLT is a remarkably strong detector of forehead wrinkles in 2D images.
Description: This is the accepted manuscript version of the paper. © 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Version: Accepted for publication
DOI: 10.1109/ACCESS.2015.2455871
URI: https://dspace.lboro.ac.uk/2134/20269
Publisher Link: http://dx.doi.org/10.1109/ACCESS.2015.2455871
Appears in Collections:Published Articles (Computer Science)

Files associated with this item:

File Description SizeFormat
Chong-IEEE Access- 2015-accepted.pdfAccepted version1.63 MBAdobe PDFView/Open


SFX Query

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.