Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/476244Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.author | Ali Shojaeipour (P58894) | - |
| dc.date.accessioned | 2023-10-06T09:15:09Z | - |
| dc.date.available | 2023-10-06T09:15:09Z | - |
| dc.date.issued | 2014 | - |
| dc.identifier.other | ukmvital:78291 | - |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/476244 | - |
| dc.description | Master of Information Technology,Recently pattern recognition approach becomes an interest in medical diagnosis has significant contribution to the growth of medical field. In future some factors are required to be improved, such as accuracy, speed, rate of recognition and computational cost. In order to use the pattern recognition in medical field the special medical knowledge and experiences are required. According to the increasing consumption of sugar materials in human life and growing trend of the machine life, the prevalence of diabetes is on the rise and observed all patients of this disease with regardless of this effects mostly decrease or loss their visions. For automatic diagnosis of diabetic retinopathy (DR) and determine diabetic eye from healthy eye we need to extract some features from the retinopathy images. Inasmuch there are various possible characteristics can be achieved from the retina photography images it is important to find out which features are most effective in diabetic detection. In this study we used some methods to detect and observe the retinal vessels, optic disc and then exudates in diabetic retinopathy color images. Optic disc contain entry and exit of optic nerve and blood vessels to the retina, so localization and segmentation of this area is one of the most important steps in automatic analysis system. To determine the optic disc Gaussian filter was used to enhancing image, and separate vessels, with brightness intensity distribution then wavelet transform was used to extract vessels and after that according to 7 features which obtained from expert, the location of optic disc was determined and after extract this area, exudates were determined. Finally images are classified with Adaboost classifier. The Adaboost algorithm combines the weak learners, and utilizing their advantages, improves the performance of system. The dataset was used in this study taken from Khatam-Al-Anbia ophthalmology subspecialty hospital of IRAN and the obtained results are compared with pathology expert. This comparison shows the system obtained promising results and cause of using weak learners, the proposed approach has low computing complication. | - |
| dc.language.iso | may | - |
| dc.publisher | UKM, Bangi | - |
| dc.relation | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | - |
| dc.rights | UKM | - |
| dc.subject | Universiti Kebangsaan Malaysia -- Dissertations | - |
| dc.subject | Dissertations, Academic -- Malaysia | - |
| dc.subject | Diabetic retinopathy | - |
| dc.subject | Pattern recognition systems | - |
| dc.title | Automatic diagnosis of diabetic disease via retinopathy images | - |
| dc.type | theses | - |
| dc.format.pages | 104 | - |
| dc.identifier.callno | TK7882.P3S548 2014 3 tesis | - |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_78291+SOURCE1+SOURCE1.0.PDF Restricted Access | 4.13 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.