Title :
Higher order geometrical image features representation for action recognition
Author :
Sjarif, Nilam Nur Amir ; Shamsuddin, Siti Mariyam ; Hashim, Siti Zaiton Mohd ; Ralescu, Anca L.
Author_Institution :
Fac. of Comput., Univ. Teknol. Malaysia, Skudai, Malaysia
Abstract :
Higher order image features based on Hu moment invariants have been used successfully in a variety of image analysis tasks. This study presents the application of an invariant to unequal rescaling of the image in constructing image features suitable for action recognition. These features are computed for video images and can be used for classification. Experimental results suggest that this approach is effective and more accurate when compared with traditional geometric invariants.
Keywords :
image classification; image representation; transforms; video signal processing; Hu moment invariants; action recognition; higher order geometrical image feature representation; image analysis tasks; image classification; video images; Character recognition; Equations; Feature extraction; Image recognition; Mathematical model; Standards; action recognition; feature set; improve scale invariant; moment invariant;
Conference_Titel :
Soft Computing and Pattern Recognition (SoCPaR), 2013 International Conference of
Conference_Location :
Hanoi
Print_ISBN :
978-1-4799-3399-0
DOI :
10.1109/SOCPAR.2013.7054140
