Local binary pattern (LBP) is wide adopted for economical image feature description and simplicity. to explain the color photos, it's required to combine the LBPs from each channel of the image. The quality technique of binary combination is to simply concatenate the LBPs from each channel, but it'll increase the dimensionality of the pattern. Thus on deal with this drawback, this paper proposes a unique technique for image description with multichannel decoded LBPs. we tend to introduce adder- and decoderbased two schemas for the mixture of the LBPs from over one channel. Image retrieval experiments area unit performed to observe the effectiveness of the proposed approaches and compared with the present ways in which of multichannel techniques. The experiments square measure performed over twelve benchmark natural scene and color texture image databases, like Corel-1k, MIT-VisTex, USPTex, coloured Brodatz, and so on. It’s determined that the introduced multichannel adder- and decoder-based LBPs considerably improve the retrieval performance over every info and outdo the opposite multichannel-based approaches in terms of the average retrieval preciseness and average retrieval rate.