Koninklijke Philips Corporation N.v.

Arnaud Gesnot

Jorge Caviedes

Carolina Miro Sorolla

A method and an arrangement for processing data in a digital input image include a decoder configured to decode encoded data. A detector is configured to detect the decoded data and a correction device is configured to correct the detected data. A predicted maximum frequency is calculated based on first and second data sets transformed by first and second transform, respectively. A real maximum frequency is calculated based on data transformed by an overall transform of a data set corresponding to the concatenation of the first and second data sets. A grid is determined by comparing the real and predicted maximum frequencies for a portion of the digital input image.

Video data processing method by calculating a DCT of the concatenated data set

The present invention relates to a device for processing blocks of coded data included in a sequence of coded digital images (ES) according to a block-based coding technique, said device comprising a post-processing circuit DFD (). Such a device effectively corrects the blocking artifacts while minimizing the bandwidth used, the decoder directly transmitting decoded data blocks (B) to the post-processing circuit. In addition, such a device introduces no or little drift during the decoding since the decoder continues to use reference images (RB) during a motion compensation step. This is because the reference images are no longer the decoded images but the completely processed images as the post-processing circuit is situated in the decoding loop.

Growing community of inventors

Montigny le Bretonneux, France

Arnaud Gesnot