Abstract
There are many type of facial skin includes normal, dry, oily, combination and sensitive. In this research, it will be concentrated on three types of facial skin only which are oily, dry and combination skin. Users might have oily skin, dry skin or combination skin. This system will help users to identify which category the user’s skin belongs to. So, it is easier for the users to choose the right facial skin product that is suit with the user skin due to various products that can be found in the market nowadays. K-Means clustering technique are used for this project to cluster a partition of skin to identify either the regions of skin are oily or dry and determine either the user skin belongs to oily group, dry group or combination group which is combination of oily and dry skin. The result of this technique obtained 94% of precision rate and 4% of error rate which make the accuracy rate is 94%. This result is compared with observation’s result made earlier which is done experimentally to support the accuracy of the result.
Metadata
Item Type: | Thesis (Degree) |
---|---|
Creators: | Creators Email / ID Num. Nur Saina, Haji Basri 2009971527 |
Contributors: | Contribution Name Email / ID Num. Thesis advisor Dr. Puteri Nor Hashimah, Megat Abdul Rahman UNSPECIFIED |
Subjects: | Q Science > QA Mathematics > Mathematical statistics. Probabilities Q Science > QA Mathematics > Multivariate analysis. Cluster analysis. Longitudinal method T Technology > T Technology (General) > Philosophy. Theory. Classification. Methodology |
Divisions: | Universiti Teknologi MARA, Perak > Tapah Campus > Faculty of Computer and Mathematical Sciences |
Keywords: | Image cluster, texture descriptor, facial skin, observation technique, research methodology |
Date: | 1 July 2012 |
URI: | https://ir.uitm.edu.my/id/eprint/33536 |
Download
33536.pdf
Download (229kB)