VIDEO IMAGE ENCODING METHOD AND DECODING METHOD

摘要

Quantisation conversion coefficient information is generated by converting and quantising a prediction error image which indicates the difference between an input image containing a pixel signal and a prediction image for the input image. A local decoded image is generated by inverse quantisation and inverse conversion of the quantisation conversion coefficient information. Filter coefficient information is generated indicating the filter coefficients of one or more filters that are applied to a pixel region in the local decoded image. Processing units include one or more encoding blocks representing pixel regions, and filter application information is generated for each processing unit. The filter application information indicates whether or not filter processing is to be applied to the encoding blocks, and also indicates one or more candidates for filters to be applied if a filter is to be applied. The filter application information is generated based on block division information that indicates how the input image to be encoded is divided into encoding blocks. When a filter is to be applied to the processing units that include one or more encoded blocks, if two or more filters are present, the filters to be applied are switched in accordance with the filter application information and also in accordance with an index calculated from the pixel signal contained in the processing units to which the filters are to be applied.