Feature-Based Retinal Image Registration Using D-Saddle Feature
Retinal image registration is important to assist diagnosis and monitor retinal diseases, such as diabetic retinopathy and glaucoma. However, registering retinal images for various registration applications requires the detection and distribution of feature points on the low-quality region that co...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hindawi Publishing Corporation
2017
|
| Subjects: | |
| Online Access: | http://eprints.usm.my/39118/1/Feature-Based_Retinal_Image_Registration_Using_D-Saddle_Feature.pdf |
| _version_ | 1825833989306318848 |
|---|---|
| author | Ramli, Roziana Idris, Mohd Yamani Idna Hasikin, Khairunnisa Karim, Noor Khairiah A. Wahab, Ainuddin Wahid Abdul Ahmedy, Ismail Ahmedy, Fatimah Kadri, Nahrizul Adib Arof, Hamzah |
| author_facet | Ramli, Roziana Idris, Mohd Yamani Idna Hasikin, Khairunnisa Karim, Noor Khairiah A. Wahab, Ainuddin Wahid Abdul Ahmedy, Ismail Ahmedy, Fatimah Kadri, Nahrizul Adib Arof, Hamzah |
| author_sort | Ramli, Roziana |
| collection | USM |
| description | Retinal image registration is important to assist diagnosis and monitor retinal diseases, such as diabetic retinopathy and glaucoma.
However, registering retinal images for various registration applications requires the detection and distribution of feature points on
the low-quality region that consists of vessels of varying contrast and sizes. A recent feature detector known as Saddle detects feature
points on vessels that are poorly distributed and densely positioned on strong contrast vessels. Therefore, we propose a
multiresolution difference of Gaussian pyramid with Saddle detector (D-Saddle) to detect feature points on the low-quality
region that consists of vessels with varying contrast and sizes. D-Saddle is tested on Fundus Image Registration (FIRE) Dataset that
consists of 134 retinal image pairs. Experimental results show that D-Saddle successfully registered 43% of retinal image pairs with
average registration accuracy of 2.329 pixels while a lower success rate is observed in other four state-of-the-art retinal image
registration methods GDB-ICP (28%), Harris-PIIFD (4%), H-M (16%), and Saddle (16%). Furthermore, the registration accuracy
of D-Saddle has the weakest correlation (Spearman) with the intensity uniformity metric among all methods. Finally, the
paired t-test shows that D-Saddle significantly improved the overall registration accuracy of the original Saddle. |
| first_indexed | 2024-03-06T15:16:00Z |
| format | Article |
| id | usm.eprints-39118 |
| institution | Universiti Sains Malaysia |
| language | English |
| last_indexed | 2024-03-06T15:16:00Z |
| publishDate | 2017 |
| publisher | Hindawi Publishing Corporation |
| record_format | dspace |
| spelling | usm.eprints-391182018-02-22T06:05:15Z http://eprints.usm.my/39118/ Feature-Based Retinal Image Registration Using D-Saddle Feature Ramli, Roziana Idris, Mohd Yamani Idna Hasikin, Khairunnisa Karim, Noor Khairiah A. Wahab, Ainuddin Wahid Abdul Ahmedy, Ismail Ahmedy, Fatimah Kadri, Nahrizul Adib Arof, Hamzah RK1-715 Dentistry Retinal image registration is important to assist diagnosis and monitor retinal diseases, such as diabetic retinopathy and glaucoma. However, registering retinal images for various registration applications requires the detection and distribution of feature points on the low-quality region that consists of vessels of varying contrast and sizes. A recent feature detector known as Saddle detects feature points on vessels that are poorly distributed and densely positioned on strong contrast vessels. Therefore, we propose a multiresolution difference of Gaussian pyramid with Saddle detector (D-Saddle) to detect feature points on the low-quality region that consists of vessels with varying contrast and sizes. D-Saddle is tested on Fundus Image Registration (FIRE) Dataset that consists of 134 retinal image pairs. Experimental results show that D-Saddle successfully registered 43% of retinal image pairs with average registration accuracy of 2.329 pixels while a lower success rate is observed in other four state-of-the-art retinal image registration methods GDB-ICP (28%), Harris-PIIFD (4%), H-M (16%), and Saddle (16%). Furthermore, the registration accuracy of D-Saddle has the weakest correlation (Spearman) with the intensity uniformity metric among all methods. Finally, the paired t-test shows that D-Saddle significantly improved the overall registration accuracy of the original Saddle. Hindawi Publishing Corporation 2017 Article PeerReviewed application/pdf en http://eprints.usm.my/39118/1/Feature-Based_Retinal_Image_Registration_Using_D-Saddle_Feature.pdf Ramli, Roziana and Idris, Mohd Yamani Idna and Hasikin, Khairunnisa and Karim, Noor Khairiah A. and Wahab, Ainuddin Wahid Abdul and Ahmedy, Ismail and Ahmedy, Fatimah and Kadri, Nahrizul Adib and Arof, Hamzah (2017) Feature-Based Retinal Image Registration Using D-Saddle Feature. Journal of Healthcare Engineering, 2017 (148952). pp. 1-15. ISSN 2040-2295 https://doi.org/10.1155/2017/1489524 |
| spellingShingle | RK1-715 Dentistry Ramli, Roziana Idris, Mohd Yamani Idna Hasikin, Khairunnisa Karim, Noor Khairiah A. Wahab, Ainuddin Wahid Abdul Ahmedy, Ismail Ahmedy, Fatimah Kadri, Nahrizul Adib Arof, Hamzah Feature-Based Retinal Image Registration Using D-Saddle Feature |
| title | Feature-Based Retinal Image Registration Using D-Saddle Feature |
| title_full | Feature-Based Retinal Image Registration Using D-Saddle Feature |
| title_fullStr | Feature-Based Retinal Image Registration Using D-Saddle Feature |
| title_full_unstemmed | Feature-Based Retinal Image Registration Using D-Saddle Feature |
| title_short | Feature-Based Retinal Image Registration Using D-Saddle Feature |
| title_sort | feature based retinal image registration using d saddle feature |
| topic | RK1-715 Dentistry |
| url | http://eprints.usm.my/39118/1/Feature-Based_Retinal_Image_Registration_Using_D-Saddle_Feature.pdf |
| work_keys_str_mv | AT ramliroziana featurebasedretinalimageregistrationusingdsaddlefeature AT idrismohdyamaniidna featurebasedretinalimageregistrationusingdsaddlefeature AT hasikinkhairunnisa featurebasedretinalimageregistrationusingdsaddlefeature AT karimnoorkhairiaha featurebasedretinalimageregistrationusingdsaddlefeature AT wahabainuddinwahidabdul featurebasedretinalimageregistrationusingdsaddlefeature AT ahmedyismail featurebasedretinalimageregistrationusingdsaddlefeature AT ahmedyfatimah featurebasedretinalimageregistrationusingdsaddlefeature AT kadrinahrizuladib featurebasedretinalimageregistrationusingdsaddlefeature AT arofhamzah featurebasedretinalimageregistrationusingdsaddlefeature |