Title :
Accuracy enhancement in environment sound recognition using ZC features and MPEG-7 with modified K-NN classifier feature
Author :
AlQahtani, M.O. ; Almazyad, Abdulaziz S.
Author_Institution :
Center of Excellence in Inf. Assurance, King Saud Univ., Riyadh, Saudi Arabia
Abstract :
In this paper, we modify the K-NN classifier feature for environment recognition from audio particularly for forensic application. We compute the distance between the first frame from the testing file with all frames from the training file, instead of only the corresponding frames, then we take the average. We investigated the effect of temporal zero crossing feature and some selected MPEG-7 audio low level descriptors on environment sound recognition. Experimental results show that higher recognition accuracy is achieved by using the modified K-NN classifier and confirm that the accuracy is increased when the size of the training file is decreased.
Keywords :
audio coding; computer forensics; forensic science; pattern classification; speech recognition; video coding; MPEG-7 audio low level descriptors; ZC features; accuracy enhancement; corresponding frames; environment sound recognition; forensic application; modified K-NN classifier feature; recognition accuracy; temporal zero crossing feature; testing file; training file; Transform coding; Audio Power (AP); Audio Spectrum Centroid (ASC); Audio Spectrum Envelop (ASE); Audio Spectrum Spread (ASS); Audio Waveform (AWF); K-Nearest Neighbors (k-NN); Moving Picture Experts Group (MPEG); Zero Crossing (ZC);
Conference_Titel :
Applications of Digital Information and Web Technologies (ICADIWT), 2011 Fourth International Conference on the
Conference_Location :
Stevens Point, WI
Print_ISBN :
978-1-4244-9824-6
DOI :
10.1109/ICADIWT.2011.6041406
