JP2011130050A5

JP2011130050A5 -

Info

Publication number: JP2011130050A5
Application number: JP2009284780A
Authority: JP
Filing date: 2009-12-16
Publication date: 2013-01-31
Anticipated expiration: 2029-12-16

Description

本発明に係る画像符号化装置は、所定の画素数から成る符号化ブロックを単位として画像データを圧縮符号化する符号化手段であって、前記符号化ブロックごとに量子化ステップサイズを設定して量子化可能な量子化手段を含む符号化手段と、前記量子化手段により量子化された画像データを逆量子化して復号化し、参照画像データを生成する局所復号化手段と、前記符号化ブロックを構成する複数のサブブロックごとに、前記参照画像データを参照して動きベクトルを検出する動きベクトル検出手段と、前記サブブロックごとの画像データと、前記動きベクトル検出手段が検出した動きベクトル位置における前記参照画像データとの画素差分情報に基づいて、前記符号化ブロックに含まれる動物体と背景部分との境界を検出する動物体境界検出手段と、前記量子化手段において設定される量子化ステップサイズを制御する量子化制御手段であって、前記動物体境界検出手段の検出結果に応じて、符号化対象の前記符号化ブロックに前記境界が検出された場合に設定される量子化ステップサイズを、前記境界が検出されない場合に設定される量子化ステップサイズより小さくする量子化制御手段とを有することを特徴とする。 The image coding apparatus according to the present invention is a coding means for compressing and encoding the image data coding blocks having a predetermined number of pixels as a unit, it sets the quantization step size for each of the coding blocks a local decoding means and the encoding means including quantizing possible quantization means, decodes by inverse quantizing the image data quantized by the quantization means generates the reference image data Te, the coded block for each of the plurality of sub-blocks constituting a motion vector detection means for detecting a moving-out vector by referring to the reference image data, the image data for each of the sub-block, the motion vector a position where the motion vector detecting means detects based on the pixel difference information between the reference image data in the animal body boundary for detecting the boundary between the moving object and the background portion included in the coded blocks Means out, a quantization control means for controlling the quantization step size set in the quantization means, in accordance with a detection result of the moving object boundary detection means, wherein the coded block to be coded the quantization step size to be set when the boundary is detected, the boundary is characterized by having a quantization control means to be smaller than the set is the quantization step size if not detected.

このような動作により、符号化ストリームとローカルデコード画像データが作成される。なお、直交変換部１８及び量子化部２０が符号化手段を構成し、逆量子化部２４及び逆直交変換部２６が局所復号化手段を構成する。 By such an operation, an encoded stream and local decoded image data are created. Note that the orthogonal transform unit 18 and the quantization unit 20 constitute an encoding unit, and the inverse quantization unit 24 and the inverse orthogonal transform unit 26 constitute a local decoding unit.

Claims

A encoding means for compressing and encoding the image data coding blocks having a predetermined number of pixels as a unit, including quantization possible quantization means sets the quantization step size for each of the coding blocks Encoding means;
A local decoding means for decoding and dequantizing image data quantized, to generate reference image data by the quantization means,
For each of the plurality of sub-blocks constituting the coded block, the motion vector detecting means for detecting a motion-out vector by referring to the reference image data,
Based on pixel difference information between the image data for each sub-block and the reference image data at the motion vector position detected by the motion vector detecting means, the boundary between the moving object and the background portion included in the encoded block is determined. Moving body boundary detection means for detecting;
A quantization control means for controlling a quantization step size set in the quantization means, wherein the boundary is detected in the coding block to be coded according to a detection result of the moving object boundary detection means. And a quantization control unit configured to make a quantization step size set when the boundary is not detected smaller than a quantization step size set when the boundary is not detected .

The motion vector detecting means, calculates the pixel difference absolute value sum of the reference image data in the image data and the motion vector position of each of the sub-blocks as the pixel difference information,
The moving object boundary detection means includes both a pixel difference absolute value sum smaller than a certain threshold and a pixel difference absolute value sum larger than another threshold among the pixel difference absolute value sums for each sub-block. when the image coding apparatus according to claim 1, characterized in that determining that there is the boundary to the coding block.

The motion vector detecting means, before each sub-block of Kifu Goka block, and the detected motion vector for each neighboring block located around the front Kifu Goka block, the sub-blocks as the pixel difference information or calculating the image data of the surrounding blocks, the pixel difference absolute value sum of the reference image data in the motion vector positions respectively corresponding,
The moving object boundary detecting means, previous pixel difference absolute value sum for each hexa subblock, and among the pixel difference absolute value sum for each of the surrounding blocks, a pixel difference absolute value sum is smaller than a certain threshold, than another threshold the image coding apparatus according to claim 1 in which both also large pixel difference absolute value sum when present, and judging that there is the boundary to the coding block.

The motion vector detection means calculates the pixel difference absolute value sum of the image data for each sub-block as the pixel difference information and the reference image data at the motion vector position,
The moving object boundary detection means has a number of sub-blocks indicating a sum of absolute pixel differences smaller than a certain threshold among the sum of absolute pixel differences for each of the sub-blocks, and another threshold When the number of sub-blocks showing a sum of absolute pixel differences greater than is greater than or equal to b, it is determined that the encoded block has the boundary
The image coding apparatus according to claim 1.

The motion vector detection means detects the motion vector for each sub-block of the encoded block and for each peripheral block located around the encoded block, and the sub-block or the image of the peripheral block is used as the pixel difference information. Calculating a pixel difference absolute value sum of the data and the reference image data at each corresponding motion vector position;
The moving object boundary detection means has a number of sub-blocks indicating a pixel difference absolute value sum smaller than a certain threshold among the pixel difference absolute value sum for each sub-block and the pixel difference absolute value sum for each peripheral block. When the number of sub-blocks that is greater than or equal to a and has a sum of absolute pixel differences greater than another threshold is greater than or equal to b, it is determined that the encoded block has the boundary.
The image coding apparatus according to claim 1.