TY  - CONF
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-57349084669&doi=10.1109%2fITSIM.2008.4632023&partnerID=40&md5=344dab3c788c37ec219d52483d07eb2b
A1  - Irijanti, E.
A1  - Yap, V.V.
A1  - Nayan, M.Y.
VL  - 4
Y1  - 2008///
SN  - 9781424423286
N2  - Selection of the best wavelet from various wavelet families for image compression is challenging problem. There are many wavelets that can be used to transform an image in a wavelet-based codec. However, it is necessary to use only 'one' wavelet to compress an image. The most appropriate wavelet will give a good compressed image; otherwise the wrong selection will produce a low quality image. This paper applies artificial neural network (ANN) as a method to solve this problem instead of manual selection as in a conventional wavelet-based codec. The results show that the neural network based on image characteristics can be used as a solution to solve the problem. The input variables to the ANN are two image features, namely image gradient (IAM) and spatial frequency SF) from three colour components (red, green and blue) and the output the ANN is the wavelet type. © 2008 IEEE.
N1  - cited By 2; Conference of International Symposium on Information Technology 2008, ITSim ; Conference Date: 26 August 2008 Through 29 August 2008; Conference Code:74115
KW  - Backpropagation; Digital image storage; Image classification; Image compression; Information technology
KW  -  Artificial neural networks; Challenging problems; Compressed images; Image characteristics; Image features; Image gradients; Input variables; Low qualities; Red 
KW  -  green and blues; Selection of the bests; Spatial frequencies; Wavelet selections
KW  -  Neural networks
TI  - Neural network for the best wavelet selection on colour image compression
ID  - scholars347
CY  - Kuala Lumpur
AV  - none
ER  -