Infant pain recognition system with GLCM features and GANN under unstructed lighting condition

Author

Mansor, Muhammad Naufal ; Rejab, Mohd Nazri

Author_Institution

Sch. of Mehatronic Eng., Univ. Malaysia Perlis, Arau, Malaysia

fYear

2013

fDate

Nov. 29 2013-Dec. 1 2013

Firstpage

243

Lastpage

248

Abstract

This paper discussed the crucial demand regarding the scheme to translate the silence voice from the newborn. The infant can´t afford to express their feeling of pain by voice. Hence, we proudly present an infant pain recognition system to overcome this matter. We employed the Single Scale Retinex (SSR) to remove the illumination level. Secondly, Gray-Level Co-occurrence Matrix (GLCM) was adopted as the feature extraction. We determine the condition of the infants (pain/no pain) with Hybrid Genetic Algorithm Neural Network (GANN) and Linear Discriminant Analysis (LDA). Several examples were conducted to evaluate the performance of the proposed method under different illumination levels.

Keywords

feature extraction; genetic algorithms; image colour analysis; lighting; matrix algebra; medical computing; neural nets; paediatrics; GANN; GLCM; LDA; SSR; feature extraction; gray-level co-occurrence matrix; hybrid genetic algorithm neural network; illumination level removal; infant pain recognition system; infants condition; linear discriminant analysis; newborn; silence voice; single scale retinex; unstructed lighting condition; Accuracy; Conferences; Databases; Feature extraction; Gray-scale; Lighting; Pain; GANN; GLCM; Infant Pain; SSR;

fLanguage

English

Publisher

ieee

Conference_Titel

Control System, Computing and Engineering (ICCSCE), 2013 IEEE International Conference on

Conference_Location

Mindeb

Print_ISBN

978-1-4799-1506-4

Type

conf

DOI

10.1109/ICCSCE.2013.6719967

Filename

6719967