Modified Texture Features from Histogram and Gray Level Co-occurence Matrix of Facial Data for Ethnicity Detection
Abdul Rehman, Gulraiz Khan, Aiman Siddiqi, Abdullah Aman Khan, Usman Ghani Khan · 2018
Ethnicity detection from facial images is very important information and it has numerous applications in the area of visual imagery. There are many texture features used for this method such as Local Binary Pattern, Harlick features etc. Most of the manuscripts are utilizing combination of other features with texture features and then classification algorithms. We have modified the texture features with the inclusion of harlick features of gray level co-occurrence matrix for ethnicity detection. These features are extracted from the segmented skin image. Also for texture features, effect of Bagging along with Random trees is highlighted to carry out efficient classification. We have performed evaluation of past texture features and our proposed texture feature method.