Rohana, Abdul Karim and Nur Amira Adila, Abd Mobin and Nurul Wahidah, Arshad and Nor Farizan, Zakaria and M. Zabri, Abu Bakar (2019) Early rubeosis iridis detection using feature extraction process. In: 5th International Conference on Electrical, Control and Computer Engineering (INECCE 2019) , 29-30 July 2019 , Swiss Garden Kuantan. pp. 1-10.. (Unpublished)
![[img]](http://umpir.ump.edu.my/style/images/fileicons/other.png) |
Pdf
27. Early rubeosis iridis detection using feature extraction process.pdf Restricted to Repository staff only Download (562kB) | Request a copy |
|
|
Pdf
27.1 Early rubeosis iridis detection using feature extraction process.pdf Download (88kB) | Preview |
Abstract
Iris analytical studies the relationship between human health and changes in the anatomy of the iris. One of the related to the changes of the anat-omy of the iris is diabetic. Diabetic illness can be determine from the iris of hu-man eyes because it’s affects the eyes. Latest advance technologies are intro-duced in the image processing that helps automate detection of diabetic in iris based on the analysis of feature extractions. This analysis not only helps diagnose the disease, besides its helps detect the disease. Various features are detected on iris such as texture, colour, histogram and shape features. In this paper, the dataset of iris image is use to detect and recognise the rubeosis iridis. To detect and rec-ognize the rubeosis iridis, it needs to extract the detail of the image of iris using image processing methods. In this paper, the proposed method is to detect and recognize the rubeosis iridis by using the feature extraction and the datasets used for this project is Warsaw Biobase. Finally, the rubeosis iridis is detected and the proposed method is discussed on this paper. The results obtained from the exper-iment show that the normal and abnormal iris image can be classified using orig-inal and small size of iris image. In this experiment, abnormal original are greater than 1200000 pixel while for small size are less than 35000 pixel. Normal origi-nal size which is less than 1200000 pixel and for small are less than 25000 pixel. By considering this result, the proposed method can be extended to the iris mon-itoring system.
| Item Type: | Conference or Workshop Item (Lecture) |
|---|---|
| Uncontrolled Keywords: | Features extraction; Blood vessels; Diabetic |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Faculty/Division: | Faculty of Electrical & Electronic Engineering |
| Depositing User: | Pn. Hazlinda Abd Rahman |
| Date Deposited: | 23 Dec 2019 08:23 |
| Last Modified: | 23 Dec 2019 08:23 |
| URI: | http://umpir.ump.edu.my/id/eprint/26666 |
| Download Statistic: | View Download Statistics |
Actions (login required)
 |
View Item |