Error concealment using adaptive multilayer perceptrons (MLPs) for block-based image coding

Citation
Yl. Huang et Rf. Chang, Error concealment using adaptive multilayer perceptrons (MLPs) for block-based image coding, NEURAL C AP, 9(2), 2000, pp. 83-92
Citations number
22
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
NEURAL COMPUTING & APPLICATIONS
ISSN journal
09410643 → ACNP
Volume
9
Issue
2
Year of publication
2000
Pages
83 - 92
Database
ISI
SICI code
0941-0643(2000)9:2<83:ECUAMP>2.0.ZU;2-X
Abstract
Image coding algorithms such as Vector Quantisation (VQ), JPEG and MPEG hav e been widely used for encoding image and video. These compression systems utilise block-based coding techniques to achieve a higher compression ratio . However, a cell loss or a random bit error during network transmission wi ll permeate into the whole block, and then generate several damaged blocks. Therefore, an efficient Error Concealment (EC) scheme is essential for dim inishing the impact of damaged blocks in a compressed image. In this paper, a novel adaptive EC algorithm is proposed to conceal the error for block-b ased image coding systems by using neural network techniques in the spatial domain. In the proposed algorithm, only the intra-frame information is use d for reconstructing the image with damaged blocks. The information of pixe ls surrounding a damaged block is used to recover the errors using the neur al network models. Computer simulation results show that the visual quality and the PSNR evaluation of a reconstructed image are significantly improve d using the proposed EC algorithm.