Lec4: Feature Extraction Methods for the classification of images
Coverage of Keynote lecture on "Feature Extraction Methods for the classification of images" . Following Topics were discussed:
1:10 Purpose of extracting texture features E.G. Calculating Standard Deviation of all the image pixels will help the computer to decide if the surface is smooth or rough.
7:22 Different texture feature extraction methods available.
7:51 List of First Order Statistics.
8:13 Creating Gray Level Co-occurence Matrix (GLCM) which is a Second Order Statistic.
18:00 Fourteen Different Haralick's texture parameters extracted from GLCM.
26:21 Application of GLCM to determine the orientation of lines in an image and to determine if the image is homogenous.
43.24 Local Binary Pattern (LBP)
47:07 Limitation of LBP.
48:34 Designing a rotational invariant LBP.