The Open University of Israel Department of Mathematics and Computer Science Identification of feeding strikes by larval fish from continuous high-speed digital video
Download 1.38 Mb.
|
4.1.3 Video clip extractionThe result of this stage of our processing is a defined area around each detected mouth. We extract 121x121 pixels centered on the mouth’s central pixel for 21 frames from the compressed video (for the Amatitlania nigrofasciata and Hemichromis bimaculatus) or 241x241 pixels for 41 frames from the original hi-res video (for the slower eating Sparus aurata). The choice of spatial dimensions allows coverage of entire heads, along with sufficient margins for possible food floating around the fish. Temporal dimension were empirically determined to be long enough to span feeding. Figure 5 depicts frames from an example pose-normalized video volume of a feeding fish. 
Figure 5: Example of a pose-normalized video volume of a feeding fish Extracted spatio-temporal volume in canonical views (horizontal, right-facing views) of a feeding fish. The prey is marked by a red circle, and enters the mouth at 60 ms. The mouth is closed at 120 ms. Directory: lists lists -> June 2012 Table of Contents lists -> Patterns and Impacts of Korea's Cultural Exports: Focused on East Asia Jeong Gon Kim Se Young Ahn lists -> 2013 Academic Year Exchange Student Application Guidelines lists -> Finding of inquest lists -> Itu consultation on the Role of Governments Submission by a cross-section of African stakeholders Preamble lists -> Artist – title – length – description – quality – songs – place/date – source lists -> Course Title: English Phonetics and Phonology Course Number: 2231222 Meeting Time lists -> Gwr/Western Region lists -> Australian contribution to the cwg-internet Ref cl-13/68 Introduction lists -> The Word Within the Word – List #1 Download 1.38 Mb. Share with your friends:
|