DCT information recovery of erroneous image blocks by a neuralpredictor

摘要

In this paper, the problem of DCT information recovery in thentransmission of coded visual data over packet networks is addressed. Thenloss of a packet conveying coded block data leads to the unsuccessfulnreconstruction of the relevant area, with consequent degradation of thenreceived image quality. The proposed method allows recovery of a subsetnof the missing DCT coefficients sufficient to achieve goodnreconstruction quality of the lost block, based on the availablensurrounding information. To this purpose, a neural predictor wasncarefully designed and suitably trained with an appropriate set ofnsynthetic and natural patterns. An extensive testing phase, performed onna large set of images with different frequency characteristics, revealednthat the method provides very good reconstruction capabilities