Device and method for allocating code amounts to intra prediction modes

An image decoding device for decoding blocks obtained by dividing an image of a frame included in a video, the image decoding device comprising:

• a decoder circuitry configured to acquire identification information of an intra prediction mode by decoding encoded data generated by an image encoding device;
• a most probable modes (MPM) generator circuitry configured to generate most probable modes including a plurality of intra prediction modes applied to peripheral blocks of a target block of intra prediction;
• a feature amount calculator circuitry configured to calculate feature amounts of each of a plurality of reference pixel groups adjacent to the target block, wherein the plurality of reference pixel groups includes a plurality of reference pixel;
• an allocation changer circuitry configured to rearrange an order of the most probable modes among the most probable modes by comparing each of the feature amounts of the plurality of reference pixel groups, and change allocation of code amounts to each of the most probable mode;
• and
• a mode specifier circuitry configured to specify the intra prediction mode based on the decoded identification information and the most probable modes.

An image decoding method for decoding blocks obtained by dividing an image of a frame included in a video, the image decoding method comprising:

• acquiring identification information of an intra prediction mode by decoding encoded data generated by an image encoding device;
• generating most probable modes including a plurality of intra prediction modes applied to peripheral blocks of a target block of intra prediction;
• calculating feature amounts of each of a plurality of reference pixel groups adjacent to the target block, wherein the plurality of reference pixel groups includes a plurality of reference pixel;
• rearranging an order of the most probable modes among the most probable modes by comparing each of the feature amounts of the plurality of reference pixel groups, and change allocation of code amounts to each of the most probable mode;
• and
• specifying the intra prediction mode based on the decoded identification information and the most probable modes.