@article{ Huang00,
  author = {Yu-Len Huang and Ruey-Feng Chang},
  title = {Error Concelament Using Adaptive Multilayer Perceptrons (MLPs) for Block-Based Image Coding},
  journal = {Neural Computing \& Applications},
  year = {2000},
  volume = {9},
  number = {2},
  pages = {83-92},
  annote = {Describes error concealment using neural networks. Classifies missing blocks to one of five classes depending on the block intensity gradient; smooth, left, right, up or down. Different networks is used for each class. Arbitrary sizes of blocks can be processed and the missing blocka are iteratively provessed one outer pixel frame at the time using the two outside frames as inputs to the network. Preproceesing is used to create frames for blocks with neighbouring blocks lost. },
  url = {papers/Huang00ErrConcNeurNetw.pdf},
  submitter = {Johan Garcia},
  bibdate = {Sunday, February 24, 2002 at 20:09:02 (CET)}
}

{"_id":"NRMopEtYc4XvMDtsM","bibbaseid":"huang-chang-errorconcelamentusingadaptivemultilayerperceptronsmlpsforblockbasedimagecoding-2000","downloads":0,"creationDate":"2015-12-04T23:32:57.819Z","title":"Error Concelament Using Adaptive Multilayer Perceptrons (MLPs) for Block-Based Image Coding","author_short":["Huang, Y.","Chang, R."],"year":2000,"bibtype":"article","biburl":"http://www.cs.kau.se/cs/prtp/prtp.bib","bibdata":{"annote":"Describes error concealment using neural networks. Classifies missing blocks to one of five classes depending on the block intensity gradient; smooth, left, right, up or down. Different networks is used for each class. Arbitrary sizes of blocks can be processed and the missing blocka are iteratively provessed one outer pixel frame at the time using the two outside frames as inputs to the network. Preproceesing is used to create frames for blocks with neighbouring blocks lost.","author":["Huang, Yu-Len","Chang, Ruey-Feng"],"author_short":["Huang, Y.","Chang, R."],"bibdate":"Sunday, February 24, 2002 at 20:09:02 (CET)","bibtex":"@article{ Huang00,\n author = {Yu-Len Huang and Ruey-Feng Chang},\n title = {Error Concelament Using Adaptive Multilayer Perceptrons (MLPs) for Block-Based Image Coding},\n journal = {Neural Computing \\& Applications},\n year = {2000},\n volume = {9},\n number = {2},\n pages = {83-92},\n annote = {Describes error concealment using neural networks. Classifies missing blocks to one of five classes depending on the block intensity gradient; smooth, left, right, up or down. Different networks is used for each class. Arbitrary sizes of blocks can be processed and the missing blocka are iteratively provessed one outer pixel frame at the time using the two outside frames as inputs to the network. Preproceesing is used to create frames for blocks with neighbouring blocks lost. },\n url = {papers/Huang00ErrConcNeurNetw.pdf},\n submitter = {Johan Garcia},\n bibdate = {Sunday, February 24, 2002 at 20:09:02 (CET)}\n}","bibtype":"article","id":"Huang00","journal":"Neural Computing \\& Applications","key":"Huang00","number":"2","pages":"83-92","submitter":"Johan Garcia","title":"Error Concelament Using Adaptive Multilayer Perceptrons (MLPs) for Block-Based Image Coding","type":"article","url":"papers/Huang00ErrConcNeurNetw.pdf","volume":"9","year":"2000","bibbaseid":"huang-chang-errorconcelamentusingadaptivemultilayerperceptronsmlpsforblockbasedimagecoding-2000","role":"author","urls":{"Paper":"http://www.cs.kau.se/cs/prtp/papers/Huang00ErrConcNeurNetw.pdf"},"downloads":0},"search_terms":["error","concelament","using","adaptive","multilayer","perceptrons","mlps","block","based","image","coding","huang","chang"],"keywords":[],"authorIDs":[],"dataSources":["6WGcSu2Ku7pZzqCcg"]}