Local Descriptor for Retinal Fundus Image Registration

A feature-based retinal image registration (RIR) technique aligns multiple fundus images and composed of pre-processing, feature point extraction, feature descriptor, matching and geometrical transformation. Challenges in RIR include difference in scaling, intensity and rotation between images. The...

Full description

Bibliographic Details
Main Authors: Roziana Ramli, Mohd Yamani Idna Idris, Khairunnisa Hasikin, Noor Khairiah A. Karim, Ainuddin Wahid Abdul Wahab, Ismail Ahmedy, Fatimah Ahmedy, Hamzah Arof
Format: Article
Language:English
Published: 2020
Subjects:
Online Access:https://eprints.ums.edu.my/id/eprint/25621/1/Local%20descriptor%20for%20retinal%20fundus%20image%20registration.pdf
_version_ 1825713906670108672
author Roziana Ramli
Mohd Yamani Idna Idris
Khairunnisa Hasikin
Noor Khairiah A. Karim
Ainuddin Wahid Abdul Wahab
Ismail Ahmedy
Fatimah Ahmedy
Hamzah Arof
author_facet Roziana Ramli
Mohd Yamani Idna Idris
Khairunnisa Hasikin
Noor Khairiah A. Karim
Ainuddin Wahid Abdul Wahab
Ismail Ahmedy
Fatimah Ahmedy
Hamzah Arof
author_sort Roziana Ramli
collection UMS
description A feature-based retinal image registration (RIR) technique aligns multiple fundus images and composed of pre-processing, feature point extraction, feature descriptor, matching and geometrical transformation. Challenges in RIR include difference in scaling, intensity and rotation between images. The scale and intensity differences can be minimised with consistent imaging setup and image enhancement during the pre-processing, respectively. The rotation can be addressed with feature descriptor method that robust to varying rotation. Therefore, a feature descriptor method is proposed based on statistical properties (FiSP) to describe the circular region surrounding the feature point. From the experiments on public Fundus Image Registration dataset, FiSP established 99.227% average correct matches for rotations between 0° and 180°. Then, FiSP is paired with Harris corner, scale-invariant feature transform (SIFT), speeded-up robust feature (SURF), Ghassabi's and D-Saddle feature point extraction methods to assess its registration performance and compare with the existing feature-based RIR techniques, namely generalised dual-bootstrap iterative closet point (GDB-ICP), Harris-partial intensity invariant feature descriptor (PIIFD), Ghassabi's-SIFT, H-M 16, H-M 17 and D-Saddle-histogram of oriented gradients (HOG). The combination of SIFT-FiSP registered 64.179% of the image pairs and significantly outperformed other techniques with mean difference between 25.373 and 60.448% (p = <;0.001*).
first_indexed 2024-03-06T03:03:50Z
format Article
id ums.eprints-25621
institution Universiti Malaysia Sabah
language English
last_indexed 2024-03-06T03:03:50Z
publishDate 2020
record_format dspace
spelling ums.eprints-256212020-07-20T01:32:42Z https://eprints.ums.edu.my/id/eprint/25621/ Local Descriptor for Retinal Fundus Image Registration Roziana Ramli Mohd Yamani Idna Idris Khairunnisa Hasikin Noor Khairiah A. Karim Ainuddin Wahid Abdul Wahab Ismail Ahmedy Fatimah Ahmedy Hamzah Arof R Medicine (General) RE Ophthalmology A feature-based retinal image registration (RIR) technique aligns multiple fundus images and composed of pre-processing, feature point extraction, feature descriptor, matching and geometrical transformation. Challenges in RIR include difference in scaling, intensity and rotation between images. The scale and intensity differences can be minimised with consistent imaging setup and image enhancement during the pre-processing, respectively. The rotation can be addressed with feature descriptor method that robust to varying rotation. Therefore, a feature descriptor method is proposed based on statistical properties (FiSP) to describe the circular region surrounding the feature point. From the experiments on public Fundus Image Registration dataset, FiSP established 99.227% average correct matches for rotations between 0° and 180°. Then, FiSP is paired with Harris corner, scale-invariant feature transform (SIFT), speeded-up robust feature (SURF), Ghassabi's and D-Saddle feature point extraction methods to assess its registration performance and compare with the existing feature-based RIR techniques, namely generalised dual-bootstrap iterative closet point (GDB-ICP), Harris-partial intensity invariant feature descriptor (PIIFD), Ghassabi's-SIFT, H-M 16, H-M 17 and D-Saddle-histogram of oriented gradients (HOG). The combination of SIFT-FiSP registered 64.179% of the image pairs and significantly outperformed other techniques with mean difference between 25.373 and 60.448% (p = <;0.001*). 2020 Article PeerReviewed text en https://eprints.ums.edu.my/id/eprint/25621/1/Local%20descriptor%20for%20retinal%20fundus%20image%20registration.pdf Roziana Ramli and Mohd Yamani Idna Idris and Khairunnisa Hasikin and Noor Khairiah A. Karim and Ainuddin Wahid Abdul Wahab and Ismail Ahmedy and Fatimah Ahmedy and Hamzah Arof (2020) Local Descriptor for Retinal Fundus Image Registration. IET Computer Vision, 14 (4). pp. 144-153.
spellingShingle R Medicine (General)
RE Ophthalmology
Roziana Ramli
Mohd Yamani Idna Idris
Khairunnisa Hasikin
Noor Khairiah A. Karim
Ainuddin Wahid Abdul Wahab
Ismail Ahmedy
Fatimah Ahmedy
Hamzah Arof
Local Descriptor for Retinal Fundus Image Registration
title Local Descriptor for Retinal Fundus Image Registration
title_full Local Descriptor for Retinal Fundus Image Registration
title_fullStr Local Descriptor for Retinal Fundus Image Registration
title_full_unstemmed Local Descriptor for Retinal Fundus Image Registration
title_short Local Descriptor for Retinal Fundus Image Registration
title_sort local descriptor for retinal fundus image registration
topic R Medicine (General)
RE Ophthalmology
url https://eprints.ums.edu.my/id/eprint/25621/1/Local%20descriptor%20for%20retinal%20fundus%20image%20registration.pdf
work_keys_str_mv AT rozianaramli localdescriptorforretinalfundusimageregistration
AT mohdyamaniidnaidris localdescriptorforretinalfundusimageregistration
AT khairunnisahasikin localdescriptorforretinalfundusimageregistration
AT noorkhairiahakarim localdescriptorforretinalfundusimageregistration
AT ainuddinwahidabdulwahab localdescriptorforretinalfundusimageregistration
AT ismailahmedy localdescriptorforretinalfundusimageregistration
AT fatimahahmedy localdescriptorforretinalfundusimageregistration
AT hamzaharof localdescriptorforretinalfundusimageregistration