frog classification using machine learning techniques

16
Frog classi cation using machine learning techniques Chenn-Jung Huang a* , Yi-Ju Yang b , Dian-Xiu Yang a , You- Jia Chen a a Department of Computer and Information Science b Institute of Ecology and Environmental Education Expert Systems with Applications 36 (2009) 3737–3743, ELSEVIER Presenter Chia-Cheng Chen 1

Upload: aviv

Post on 25-Feb-2016

49 views

Category:

Documents


2 download

DESCRIPTION

Frog classification using machine learning techniques. Chenn-Jung Huang a* , Yi-Ju Yang b , Dian-Xiu Yang a , You-Jia Chen a a Department of Computer and Information Science b Institute of Ecology and Environmental Education - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Frog classification using machine learning  techniques

Frog classification using machine learning techniques

Chenn-Jung Huang a*, Yi-Ju Yang b, Dian-Xiu Yang a, You-Jia Chen a

aDepartment of Computer and Information SciencebInstitute of Ecology and Environmental Education

Expert Systems with Applications 36 (2009) 3737–3743, ELSEVIER

Presenter Chia-Cheng Chen 1

Page 2: Frog classification using machine learning  techniques

Introduction

Architecture of on-line frog sound identification system

Experimental results

Conclusion and feature work

Outline

2

Page 3: Frog classification using machine learning  techniques

An automatic frog sound identification system is developed in this work.

Three features, spectral centroid, signal bandwidth and threshold-crossing rate, are extracted to serve as the parameters for the frog sound classification.

Introduction

3

Page 4: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system

4

Page 5: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

5

Page 6: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

6

Signal preprocessing

◦Resampled at 8 kHz frequency and saved as 8-bit mono format

◦Normalized to the same level

Page 7: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

7

Syllable segmentation

1. Amplitude matrix S(a, t), initially n=1

2. Find an and tn, such that S(an, tn)=max{S(|a|, t)}

3. If |an| <= athreshold, stop the segmentation process. The athreshold

is the empirical threshold.

4. Store the amplitude trajectories corresponding to the nth

syllable in function An(τ), where τ=tn-ɛ,…,tn,…, tn+ɛ and is

the empirical threshold of the syllable.

Page 8: Frog classification using machine learning  techniques

Feature extraction

◦ Spectral centroid

◦ Signal bandwidth

Architecture of on-line frog sound identification system(Cont.)

8

Page 9: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

9

◦ Threshold-crossing rate

Page 10: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

10

Classification

◦ kth nearest neighboring (KNN)

◦ Support vector machines (SVM)

Page 11: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

11

kth nearest neighboring (KNN)

The kNN method is a simple yet effective method for classification

in the areas of pattern recognition, machine learning, data mining,

and information retrieval.

Page 12: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

12

Support vector machines (SVM)

Lagrangian Multiplier Method:

Page 13: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

13

Page 14: Frog classification using machine learning  techniques

Architecture of on-line frog sound identification system(Cont.)

14

Page 15: Frog classification using machine learning  techniques

Experimental results

15

Page 16: Frog classification using machine learning  techniques

An automatic frog sound identification system is proposed in this work to provide the public to consult online.

The sound samples are first properly segmented into syllables.

Conclusion and feature work

16