Abstract

The importance of video key frame extraction with the growing use of multimedia in day-to-day life is becoming vital. However, when it comes to video summarization. The key frame extraction plays an important role. Thepadea's Sorted Ternary Block Truncation Coding (TSTBTC) is used for color feature extraction from each video frame. Key frame extraction using TSTBTC is proposed for video summarization using RGB colorspace. The paper attempts performance comparison of eight diverse color spaces including RGB in TSTBTC algorithm for video key frame extraction. The experimentation is done using a test bed of 15 videos with diverse eight color spaces (KLUV, RGB, XYZ, YIQ, YUV, YCbCr and Kekre's bi-orthogonal color spaces LXY and L'X'Y') and five assorted similarity measures (Euclidean Distance, Sorensen Distance, Canberra, Square Chord Distance, Mean Square Error). The experimental analysis shows for TSTBTC KLUV gives better completeness indicating better video key frame experimentation. Along with this Euclidean distance gives best performance of completeness in all color space.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call