JP5222878B2 - Intra prediction apparatus, encoder, decoder, and program - Google Patents

Description

  The present invention relates to an intra prediction apparatus, an encoder, a decoder, and a program that can perform intra prediction in an arbitrary direction.

  MPEG-4 AVC / H. In the H.264 system, for a pixel block that is encoded without using inter prediction (inter-screen prediction), a prediction image is generated from the pixel values of the encoded adjacent blocks, and a difference from the prediction image is encoded. Intra-prediction coding (intra-screen prediction coding) is employed.

  18 shows a conventional MPEG-4 AVC / H. The block diagram of the encoder 4 in a H.264 system is shown. The encoder 4 includes a rearrangement unit 21, a subtraction unit 22, an orthogonal transformation unit 23, a quantization unit 24, a variable length coding unit 25, an inverse quantization unit 26, and an inverse orthogonal transformation unit 27. A changeover switch 28, an intra prediction unit 29, a frame memory 30, a motion compensation prediction unit 31, and an addition unit 32.

  The rearrangement unit 21 temporarily stores the input video signal, rearranges the order of the frame images, and outputs the frame image necessary for the encoding process to the subtraction unit 22 and the motion compensation prediction unit 31.

  The motion compensation prediction unit 31 performs motion vector detection on the input image supplied from the rearrangement unit 21 using a reference image acquired from the frame memory 30, and performs motion compensation using the obtained motion vector. Then, the predicted image obtained as a result is output to the subtraction unit 22 and the addition unit 32 via the changeover switch 28. The motion vector information is output to the variable length coding unit 25.

  The subtraction unit 22 generates a difference image between the input image signal from the rearrangement unit 21 and the predicted image from the motion compensation prediction unit 31 or the intra prediction unit 29 and outputs the difference image to the orthogonal transformation unit 23.

  The orthogonal transform unit 23 subjects the difference image supplied from the subtraction unit 22 to orthogonal transform (for example, DCT; DiLcrete CoLine TranLform) for each small-area pixel block, and outputs the orthogonal transform coefficient to the quantization unit 24. .

  The quantization unit 24 selects a quantization table for the orthogonal transform coefficient input from the orthogonal transform unit 23, performs quantization processing, and outputs the quantization table to the variable length coding unit 25 and the inverse quantization unit 26.

  The variable length encoding unit 25 scans the quantized orthogonal transform coefficient input from the quantization unit 24 and performs variable length encoding processing to generate a bitstream, and also receives the input from the motion compensation prediction unit 31. The motion vector information is also subjected to variable length coding and output.

  The inverse quantization unit 26 performs an inverse quantization process on the quantized orthogonal transform coefficient input from the quantization unit 24 and outputs the result to the inverse orthogonal transform unit 27.

  The inverse orthogonal transform unit 27 performs inverse orthogonal transform (for example, IDCT; Inverse Le Dicrete CoLine TranLform) on the orthogonal transform coefficient input from the inverse quantization unit 26 and outputs the result to the addition unit 32.

  The addition unit 32 adds the image obtained by the inverse orthogonal transform obtained from the inverse orthogonal transform unit 27 and the prediction image obtained from the motion compensation prediction unit 31 or the intra prediction unit 29 to generate a decoded image, and generates an intra prediction unit. 29 and the frame memory 30.

  The changeover switch 28 switches between motion compensation prediction and intra prediction.

  The intra prediction unit 29 generates a predicted image that is intra-predicted from the image obtained by decoding the already-encoded block (the output image of the addition unit 32), and outputs the prediction image to the subtraction unit 22 and the addition unit 32. Here, the subtraction unit 22 outputs the difference image between the predicted image and the original image to the orthogonal transform unit 23 and encodes it via the quantization unit 24 and the variable length coding unit 25.

  Intra prediction is performed in units of 4 pixels × 4 lines, 8 pixels × 8 lines, or 16 pixels × 16 lines, and a plurality of types of prediction modes (prediction directions) (for example, prediction in units of 4 pixels × 4 lines). Selects the optimum prediction direction from among nine types. FIG. 19 is a diagram illustrating a prediction mode when prediction is performed in units of 4 pixels × 4 lines. In the figure, hatched circles indicate decoded pixels, white circles indicate pixels to be predicted, and arrows indicate prediction directions. (A) Prediction mode 0 is vertical direction prediction, (b) Prediction mode 1 is horizontal direction prediction, (c) Prediction mode 2 is DC prediction, (d) Prediction mode 3 is diagonal lower left direction prediction, ( e) Prediction mode 4 in diagonal lower right direction prediction, (f) prediction mode 5 in vertical right direction prediction, (g) prediction mode 6 in horizontal down direction prediction, and (h) prediction mode 7 in vertical left direction. In the prediction mode 8 of (i), horizontal upward prediction is performed. The above is MPEG-4 AVC / H. This is a technique of intra prediction in the H.264 system.

  In addition, in order to increase the accuracy of intra prediction, a technique of encoding a difference value between adjacent pixels by always referring to adjacent pixels instead of referring to spatially separated pixels (for example, Patent Document 1). And a technique for performing intra prediction based on the value of a pixel adjacent to a prediction block and the value of a pixel that is at least one pixel from the block (see, for example, Patent Document 2). Also known is a technique for generating a value that takes into account the frequency characteristics of adjacent decoded images as a prediction value in intra coding in order to improve the prediction performance for a picture image whose pixel value has a certain change tendency. (For example, see Patent Document 3). Furthermore, a technique is known in which intra prediction is performed by motion compensation prediction, and a motion vector is detected on the decoder side without transmitting a motion vector (see, for example, Patent Document 4).

JP 2009-049969 A JP 2008-271371 A JP 2008-245088 A JP 2007-036551 A

  However, since the conventional intra prediction method having directionality is selected from predetermined patterns, there is a problem that an appropriate intra prediction direction cannot be selected depending on the target image. In addition, since it is necessary to transmit the direction of intra prediction, there is a problem that the amount of transmission information increases. Furthermore, when intra prediction is performed by motion compensation prediction, there is a problem that a huge amount of computation is required for the decoder.

  In order to solve the above problem, the object of the present invention is not to limit the direction of intra prediction to a predetermined pattern, but enables intra prediction in an arbitrary direction, and does not transmit the direction of intra prediction, thereby reducing the bit rate. Another object of the present invention is to provide an intra-prediction device, an encoder, a decoder, and a program that can be reduced and that does not require an enormous operation in the decoder.

  In order to solve the above-mentioned problem, the present invention performs intra prediction by extrapolation of direct reference pixel values. The direction of extrapolation is obtained from indirect reference pixels.

  That is, the intra prediction apparatus according to the present invention is an intra prediction apparatus that generates an image of a prediction block as a prediction image from an image of a decoded reference block adjacent to a prediction block that is a prediction target. Edge estimation straight line generating means for selecting a pixel included in a block as an indirect reference pixel and generating an edge estimation straight line that is a straight line that passes through a plurality of indirect reference pixels and that estimates the edge direction of the image of the prediction block And a direct reference pixel that selects a pixel used for calculation of a pixel value of a predicted image from among the indirect reference pixels as a direct reference pixel, and extracts a direct reference pixel that exists on the edge estimation straight line for each pixel of the prediction block An extraction means; and a non-directional prediction means for generating a prediction image of a prediction block based on the direct reference pixels. .

  In the intra prediction apparatus according to the present invention, the edge estimation straight line generation unit selects pixels included in the reference block as indirect reference pixels, defines straight lines passing through a plurality of indirect reference pixels, and is on the straight lines. Means for calculating an index indicating a difference between a plurality of indirect reference pixels, and coefficient optimization means for obtaining a slope of the edge estimation line by setting a straight line having a minimum difference index of the indirect reference pixels as the edge estimation line. It is characterized by having.

  Further, in the intra prediction apparatus according to the present invention, the coefficient optimization unit may use the index of the difference between the indirect reference pixels as a square sum of the differences between adjacent indirect reference pixels existing on the edge estimation line, or A variance value of indirect reference pixels existing on the edge estimation straight line is used.

Further, the intra prediction apparatus according to the present invention, the direct reference pixel extracting means may comprise a direct reference pixels pixels adjacent to the predicted block of the indirect reference pixel.

  Further, in the intra prediction apparatus according to the present invention, the non-directional prediction means, when there is one direct reference pixel extracted for each pixel of the prediction block, the value of the direct reference pixel or the direct reference pixel When there is a plurality of direct reference pixels extracted for each pixel of the prediction block, using the average value or the weighted average value of pixels adjacent to the direct reference pixel as a prediction value, the direct reference closest to the pixel of the prediction block A pixel, an average value or a weighted average value of the extracted direct reference pixels, or an average value of all the indirect reference pixels and the direct reference pixels on the edge estimation line is used as a prediction value, and a direct extracted for each pixel of the prediction block. If no reference pixel exists, the direct reference pixel closest to the edge estimation line, the average value of predicted pixels in the vicinity of the pixel of the prediction block, or all The average value of the contact reference pixels, characterized in that the predicted values.

  Further, in the intra prediction apparatus according to the present invention, means for determining whether or not the index of the difference between the indirect reference pixels exceeds a predetermined value, and intra-screen prediction means for generating a prediction image from the image of the reference block When the index is less than or equal to the predetermined value, the prediction image generated by the non-directional prediction means is output, and when the index exceeds the predetermined value, the prediction image generated by the in-screen prediction means And a means for outputting the predicted image.

  In the intra prediction apparatus according to the present invention, the intra-screen prediction means generates a predicted image by selecting a prediction direction from the image of the reference block.

  In the intra prediction apparatus according to the present invention, the edge estimation straight line generation unit, the direct reference pixel extraction unit, and the non-direction prediction unit have a plurality of processing modes, and each of the processing modes has an indirect reference pixel. The position, the position of the direct reference pixel, or the derivation method of the coefficient of the edge estimation straight line is different, and the non-directional prediction unit generates a predicted image for each processing mode, and further generates a predicted image for each processing mode. The image processing apparatus includes a determination unit that selects and outputs a predicted image from among the images.

  Furthermore, an encoder according to the present invention includes the intra prediction apparatus described above.

  Furthermore, a decoder according to the present invention includes the above-described intra prediction device.

  Furthermore, the present invention provides a computer constituting an intra prediction apparatus that generates an image of a prediction block as a prediction image from an image of a decoded reference block adjacent to a prediction block to be predicted. Selecting a pixel included in the block as an indirect reference pixel and generating an edge estimation line that is a straight line that passes through a plurality of indirect reference pixels and that estimates the edge direction of the image of the prediction block; ) Selecting a pixel used for calculation of a pixel value of a predicted image from among the indirect reference pixels as a direct reference pixel, and extracting a direct reference pixel existing on the edge estimation line for each pixel of the prediction block; c) generating a prediction image of a prediction block based on the direct reference pixels, and also characterized as a program for executing It is.

  In the present invention, since the direction of intra prediction is not limited, intra prediction suitable for various images can be performed. Further, since the direction of intra prediction is obtained by calculation, it is not necessary to transmit the direction of intra prediction. Furthermore, the amount of calculation for calculating the direction of intra prediction is small.

It is a figure which illustrates the image block which performs intra prediction by this invention. It is a figure which illustrates the edge estimation straight line by this invention. It is a figure which illustrates the edge estimation straight line by this invention. It is a figure which illustrates the edge estimation straight line by this invention. It is a block diagram of the intra prediction apparatus of Example 1 by this invention. It is a figure which illustrates the image block which performs intra prediction by this invention. It is a flowchart which shows operation | movement of the intra prediction apparatus of Example 1 by this invention. It is a block diagram of an encoder provided with the intra prediction apparatus of Example 1 by this invention. It is a block diagram of a decoder provided with the intra prediction apparatus of Example 1 by this invention. It is a block diagram of the intra prediction apparatus of Example 2 by this invention. It is a block diagram of the 1st modification of the intra prediction apparatus of Example 2 by this invention. It is a block diagram of the 2nd modification of the intra prediction apparatus of Example 2 by this invention. It is a block diagram of the intra prediction apparatus of Example 3 by this invention. It is a flowchart which shows operation | movement of the intra prediction apparatus of Example 2 by this invention. It is a flowchart which shows operation | movement of the intra prediction apparatus of Example 3 by this invention. It is a block diagram of the intra prediction apparatus of Example 4 by this invention. It is a figure which illustrates a prediction pixel value curved surface. Conventional MPEG-4 AVC / H. 2 is a block diagram of an encoder in the H.264 scheme. FIG. Conventional MPEG-4 AVC / H. It is a figure which shows the prediction mode in the case of estimating by 4 pixel x4 line unit in a H.264 system.

  The intra prediction apparatus of Example 1 by this invention is demonstrated in detail with reference to drawings.

  In the intra prediction method according to the present invention, first, direct reference pixels and indirect reference pixels are defined. Here, the indirect reference pixel refers to a pixel that has been decoded and is used for obtaining an extrapolation direction. A direct reference pixel is a pixel that has been decoded and is used to calculate a predicted value by extrapolation. The indirect reference pixel and the direct reference pixel may be the same or different.

  FIG. 1 is a diagram illustrating an image block for performing intra prediction according to the present invention. The image block M is an image block (hereinafter referred to as “prediction block”) that is a target of intra prediction. Image blocks A, B, C, and D are image blocks that have been encoded and then decoded, and are image blocks that are referred to in order to predict pixel values of a prediction block (hereinafter referred to as “reference blocks”). . The size of each image block is shown for the case of 4 pixels × 4 lines in the example of FIG. 1, but may be 8 pixels × 8 lines or 16 pixels × 16 lines.

A circle with hatching in FIG. 1A directly indicates a reference pixel. The direct reference pixels include pixels adjacent to the predicted block M is suitable. In the following, as shown in FIG. 1A, all pixels included in the reference blocks A to D are indirect reference pixels, and among the indirect reference pixels, pixels that are in contact with the x-axis or y-axis are directly reference pixels. As will be described, the arrangement of the direct reference pixels and the indirect reference pixels is not limited to this. For example, as shown by the hatched circle in FIG. 1B, two pixels (two lines) in contact with the x-axis and y-axis may be used as direct reference pixels and / or indirect reference pixels, or in FIG. A pixel group having a gap such as a hatched circle may be used as a direct reference pixel and / or an indirect reference pixel. In addition, it is not always necessary to use a pixel at the same position as the indirect reference pixel, and the position of the pixel as the indirect reference pixel may be switched according to the image. For example, four types of pixels, a pixel included in the reference block A, a pixel included in the reference block B, a pixel included in the reference block D, and a pixel included in the reference blocks A, B, and D are switched, and this switching information is transmitted to the decoder. You may make it transmit.

FIG. 2 is a diagram illustrating a straight line for estimating the edge direction according to the present invention (hereinafter referred to as “edge estimated straight line”). Hereinafter, each pixel in the image block is represented by (x, y) coordinates, and the pixel value of the pixel (x, y) is represented by z (x, y). Further, as shown in the figure, the pixel closest to the origin of the prediction block M is set to (0, 0). The edge estimation straight line L passes through the indirect reference pixels (x _i , y _i ) (i = 1 to N, N is the number of indirect reference pixels), and among a plurality of straight lines represented by the formula (1), It is a straight line passing over an indirect reference pixel. The intra prediction apparatus according to the present invention estimates (derived) the slope of the edge estimation line L that minimizes the index indicating the difference between the values of the indirect reference pixels existing on the same edge estimation line L, that is, the coefficients a and b. . Here, the index indicating the difference includes various values such as variance, the sum of the absolute values of the differences, the minimum value of the absolute values of the differences, the product of the sum of squares of the differences or the absolute values of the differences, and the average value per pixel. This is a generic term for the difference values determined by the method. Further, the term “difference” in the following description means an index indicating the difference. If the slope of the edge estimation line L that minimizes the difference between the values of the indirect reference pixels is obtained, it can be estimated that the slope is the slope of the edges of the reference blocks A to D and the prediction block M, and edge estimation is performed. The image of the predicted image M can be predicted using the straight line L.

  For deriving the coefficients a and b that minimize the difference between the values of the indirect reference pixels, for example, optimization means such as a least square method, a linear programming method, a learning method using a neural network, or the like can be used.

The case where the least square method is used as the optimization means will be described with reference to FIG. FIGS. 3A and 3B are diagrams illustrating the edge estimation straight line L. FIG. 3A is an edge when the inclination is 45 degrees, FIG. 3B is an edge when the inclination is 90 degrees, and FIG. 3C is an edge when the inclination is 0 degrees. An estimated straight line L is shown. As is apparent from FIG. 3, the number of edge estimation straight lines L changes according to the inclination, and is 13 when the inclination is 45 degrees (a) and 12 when the inclination is 90 degrees (b). In the case of 0 degree in (c), the number is 8. In FIG. 3, all straight lines passing through two or more indirect reference pixels are used as the edge estimation straight line L, but only some of these straight lines may be used as the edge estimation straight line L (however, reference numerals Need to be shared between the decoder and the decoder). For example, only a straight line passing through two or more indirect reference pixels and passing through a prediction block can be used as the edge estimation straight line L. In this case, using the L ₈ ~L ₁₃ in FIGS. 3 (a) as an edge estimation linear, using L ₅ ~L ₈ in FIG. 3 (b) as an edge estimation linear, FIG 3 (c) the L ₅ ~L ₈ Are used as edge estimation straight lines.

The sum of squared errors is the sum of the squares of the differences of the values of the indirect reference pixels existing on the edge estimation line L _k (k = 1 to α, α is the number of edge estimation lines L). Deriving coefficients a and b that minimizes. Since the number of the edge estimation straight lines L varies depending on the inclination, the average value V _a per straight line of the sum of squared errors P _k is expressed by the equation (2).

ここで、Ｐ_ｋはエッジ推定直線Ｌ_ｋ上に存在する間接参照画素の値の差の平均値である。図３（ａ）の場合を例にとって説明すると、エッジ推定直線Ｌ_１の誤差の二乗Ｐ_１ ^２は｛ｚ（−３，４）−ｚ（−４，３）｝^２、エッジ推定直線Ｌ_２の誤差の二乗Ｐ_２ ^２は［｛ｚ（−２，４）−ｚ（−３，３）｝^２＋｛ｚ（−３，３）−ｚ（−４，２）｝^２］／２、エッジ推定直線Ｌ_３の誤差の二乗Ｐ_３ ^２は［｛ｚ（−１，４）−ｚ（−２，３）｝^２＋｛ｚ（−２，３）−ｚ（−３，２）｝^２＋｛ｚ（−３，２）−ｚ（−４，１）｝^２］／３となる。また、エッジ推定直線Ｌ_９の誤差の二乗Ｐ_９ ^２は［｛ｚ（５，４）−ｚ（４，３）｝^２＋｛ｚ（４，３）−ｚ（３，２）｝^２＋｛ｚ（３，２）−ｚ（２，１）｝^２＋｛ｚ（２，１）−ｚ（−１，−２）｝^２＋｛ｚ（−１，−２）−ｚ（−２，−３）｝^２］／５となる。このようにして得られるエッジ推定直線Ｌ_ｋの誤差の二乗Ｐ_ｋ ^２を合計することで傾きが４５度の場合の誤差の二乗和を算出する。エッジ推定直線が間接参照画素の中心を通らない場合の計算方法は、あらかじめ符号化器と復号器で共通の間接参照画素を計算に用いるように定めておく。例えば、間接参照画素の中心とエッジ推定直線Ｌとの距離が所定の値以下の間接参照画素のみを計算に用いる。

Here, P _k is an average value of the difference between the values of the indirect reference pixels existing on the edge estimation straight line L _k . The case of FIG. 3A will be described as an example. The square P ₁ ² of the error of the edge estimation line L ₁ is {z (−3,4) −z (−4,3)} ² , and the edge estimation line L _2. The square of error P ₂ ² is [{z (−2,4) −z (−3,3)} ² + {z (−3,3) −z (−4,2)} ² ] / 2, The error square P ₃ ² of the edge estimation straight line L ₃ is [{z (−1,4) −z (−2,3)} ² + {z (−2,3) −z (−3,2)}. ² + {z (−3,2) −z (−4,1)} ² ] / 3. Further, the square P ₉ ² of the error of the edge estimation straight line L ₉ is [{z (5,4) −z (4,3)} ² + {z (4,3) −z (3,2)} ² + {Z (3,2) -z (2,1)} ² + {z (2,1) -z (-1, -2)} ² + {z (-1, -2) -z (-2 , -3)} ² ] / 5. By summing the error squares P _k ² of the edge estimation straight line L _k obtained in this way, the sum of squares of errors when the inclination is 45 degrees is calculated. The calculation method when the edge estimation straight line does not pass through the center of the indirect reference pixel is determined in advance so that the indirect reference pixel common to the encoder and decoder is used for the calculation. For example, only indirect reference pixels whose distance between the center of the indirect reference pixels and the edge estimation straight line L is a predetermined value or less are used in the calculation.

In addition to the above, the coefficients a and b are also derived by a method that minimizes the variance of the values of the indirect reference pixels existing on the edge estimation line L _k (k = 1 to α, α is the number of edge estimation lines L). be able to. That is, when the indirect pixel group on the edge estimation line L _k is (x _i , y _i ) (i = 1 to m, m is the number of pixels of the indirect pixels on the edge estimation line L _k ), the variance V _n is _expressed by the equation It is represented by (3). Coefficient average value per linear dispersion V _n minimize a, may be obtained with b. As described above, the coefficients a and b can be derived by a method of minimizing various indexes having different values depending on the inclination.

Once the coefficients a and b are determined, the following formula (1) shows the coordinates (x _j , y _j ) (j = 1 to M, M are prediction pixels) of the pixels of the prediction block M (hereinafter referred to as “prediction pixels”). The edge estimation line L passing through (x _j , y _j ) obtained by substituting (number) is obtained for each prediction pixel, and the value of the direct reference pixel existing on the edge estimation line L is used to obtain the pixel (x _j , y _j ) is calculated.

  Also, instead of deriving the coefficients a and b and substituting them into the equation (1), a plurality of types of edge estimation straight lines L having different inclinations are prepared in advance, and the inclination that minimizes the difference between the values of the indirect reference pixels is prepared. The amount of calculation can be reduced by selecting. That is, a plurality of types of edge estimation straight lines L having different inclinations are prepared in advance, and for each pixel of the prediction block M, a table of indirect reference pixels and direct reference pixels passing on the edge estimation straight line L is prepared in advance for each inclination. Keep it. Then, for each pixel in the prediction block M, for each of a plurality of types of gradients, an average value of variance of indirect reference pixels through which a certain straight line passes is obtained, and the average value of variance obtained for each pixel is further calculated in the prediction block. It is also possible to select the slope that minimizes the average value obtained in step (b).

Calculation of the predicted value will be described with reference to FIG. FIG. 4 shows an edge estimation straight line L ₁ passing through the pixel (0, 0), an edge estimation straight line L ₂ passing through the pixel (1, −2), and an edge estimation straight line L ₃ passing through the pixel (2, −3). It is a figure illustrated. Edge estimated straight line _{L 2} on the present direct reference pixels (2,1), so the edge estimated straight line _{L 2} passes through the center of direct reference pixel (2,1), pixel (1, -2) of Let the predicted value be the pixel value z (2, 1).

Since direct reference pixels on the edge estimated straight line _{L 3} (3, 1) is present, it can be a pixel (2, -3) pixel values predicted value of z (3, 1). However, the edge estimated straight line L ₃ is not a straight line passing through the center of the direct reference pixel (3,1). Thus, when the edge estimation straight line L is not a straight line passing directly through the center of the reference pixel, prediction can be made from the direct reference pixel on the edge estimation straight line L and the direct reference pixel adjacent thereto. For example, an average value of the direct reference pixel value z (3, 1) and the direct reference pixel value z (4, 1) adjacent thereto or a weighted average value weighted according to the distance is set to the pixel (2, −3). ).

On the other hand, on the edge estimated straight line _{L 1} is a direct reference pixels (0,1), (- 1, -2), (- 1, -3) is present. Thus, when there are a plurality of direct reference pixels used for prediction, a predicted value is obtained by a predetermined method. For example, when the direct reference pixels used for prediction with respect to the prediction pixel are equidistant, the average value of the direct reference pixels is used as the prediction value. When the direct reference pixels used for prediction with respect to the prediction pixel are not equidistant, the value of the direct reference pixel closest to the prediction pixel, the average value of the direct reference pixels or the weighted average value corresponding to the distance, or the edge estimation straight line L Various methods such as an average value of all the above indirect reference pixels and direct reference pixels can be considered.

  Further, when the number of direct reference pixels is smaller than the number of direct reference pixels described in the present embodiment, there may be a case where no direct reference pixel exists on the edge estimation line L. In this case, the predicted value is obtained by a predetermined method. For example, various methods are conceivable, such as using the value of the direct reference pixel closest to the edge estimation line L, the average value of the predicted pixels in the vicinity, or the average value of all the direct reference pixels as the predicted value. Further, there may be a case where the edge estimation straight line L does not exist as in the pixel (3, −3) illustrated in FIG. Similarly in this case, the predicted value is obtained by a predetermined method, for example, using the average value of the predicted pixels in the vicinity or the average value of all the direct reference pixels as the predicted value, etc. Various methods are conceivable.

  FIG. 5 is a configuration diagram of the intra prediction apparatus 1 according to the first embodiment of the present invention. The intra prediction apparatus 1 according to the first embodiment includes an indirect reference pixel difference calculation unit 11, a coefficient optimization unit 12, a direct reference pixel extraction unit 13, and a non-directional prediction unit 14.

  The indirect reference pixel difference calculation unit 11 selects an indirect reference pixel from the input decoded image, calculates a difference between a plurality of indirect reference pixels existing on the edge estimation straight line L (indirect reference pixel difference), and calculates a coefficient. The result is output to the optimization unit 12.

  The coefficient optimizing unit 12 derives the coefficients a and b of the edge estimation straight line L expressed by Expression (1) and outputs them directly to the reference pixel extracting unit 13. Alternatively, the inclination of the edge estimation line L that minimizes the difference between the values of the indirect reference pixels is selected from a plurality of edge estimation lines L having different inclinations prepared in advance, and is directly output to the reference pixel extraction unit 13.

The direct reference pixel extraction unit 13 selects a pixel used for calculation of the pixel value of the predicted image among the indirect reference pixels as a direct reference pixel. The position of the direct reference pixel does not always need to be constant, and can be changed according to the inclination of the edge estimation straight line L, for example. Next, from the coefficient (or slope) input from the coefficient optimization unit 12 and the coordinates (x _j , y _j ) (j = 1 to M, M is the number of predicted pixels) of the predicted pixel, for each predicted pixel An edge estimation straight line L passing through the prediction pixel is obtained. Then, a direct reference pixel existing on the edge estimation straight line L is extracted for each prediction pixel, and is output to the non-directional prediction unit 14.

  As described above, when there is one direct reference pixel extracted by the direct reference pixel extraction unit 13, the non-directional prediction unit 14 determines the value of the direct reference pixel when the edge estimation straight line L passes through the center of the direct reference pixel. When the edge estimation straight line L does not directly pass through the center of the reference pixel as a predicted value, an average value or a weighted average value weighted according to the distance between the value of the direct reference pixel and the value of the pixel adjacent to the direct reference pixel Estimated value. When there are a plurality of direct reference pixels extracted by the direct reference pixel extraction unit 13, when the direct reference pixels used for prediction are equidistant with respect to the prediction pixel, the average value of the direct reference pixels is set as the prediction value. When the direct reference pixels used for prediction with respect to the prediction pixel are not equidistant, the value of the direct reference pixel closest to the prediction pixel, the weighted average value according to the distance, or all the indirect reference pixels on the edge estimation line L And the average value of the direct reference pixels is set as the predicted value. Thereafter, processing such as integerization and clipping is performed, and a predicted image after processing is output.

  In this embodiment, the reference pixel (reference block) is adjacent to the prediction block M, but the pixel to be referred to is not limited to the images of the reference blocks A to D as long as it is a decoded pixel. For example, it is possible to further refer to pixels other than the reference blocks A to D, or to refer to only some pixels of the reference blocks A to D. It is also conceivable to divide the prediction block into smaller blocks and perform encoding in units of small blocks. For example, as illustrated in FIG. 6, when the small block E in the prediction block M has already been decoded, the pixel value of the small block E is predicted when a region other than the small block E in the prediction block M is predicted. Can also be referred to. As described above, as long as the pixel referred to as the direct reference pixel and the indirect reference pixel is a decoded pixel, various combinations of the positions and the number of pixels are possible. In this specification, for ease of explanation, reference blocks A to D are used as examples of reference pixels.

  Next, the operation of the intra prediction apparatus 1 according to the first embodiment will be described with reference to FIG.

[Operation of Intra Prediction Device 1 of First Embodiment]
FIG. 7 is a flowchart illustrating the operation of the intra prediction apparatus 1 according to the first embodiment. First, in step S101, the indirect reference pixel difference calculation unit 11 calculates a difference (indirect reference pixel difference) between a plurality of indirect reference pixels existing on the edge estimation straight line L.

  In step S102, the coefficient optimization unit 12 derives the coefficient (or slope) of the edge estimation straight line L using the indirect reference pixel difference.

  In step S103, the direct reference pixel extraction unit 13 extracts direct reference pixels existing on the edge estimation straight line L.

  In step S104, the non-direction prediction unit 14 generates a prediction image of the prediction block M.

  Next, an encoder and a decoder according to the present invention will be described with reference to FIGS.

[Encoder and decoder]
FIG. 8 is a configuration diagram of an encoder including the intra prediction apparatus according to the first embodiment of the present invention. The encoder 2 is a conventional MPEG-4 AVC / H. The intra prediction apparatus 1 according to the first embodiment of the present invention is provided instead of the intra prediction unit in the H.264 encoder (see FIG. 18). The intra prediction apparatus 1 inputs an image obtained by decoding an already-encoded block via the inverse quantization unit 26 and the inverse orthogonal transform unit 27, and as described above, the indirect reference pixel difference calculation unit 11, the coefficient optimization unit 12, and the direct The prediction image generated through the reference pixel extraction unit 13 and the non-directional prediction unit 14 is output to the subtraction unit 22 and the addition unit 32 via the switch 28.

  FIG. 9 is a configuration diagram of a decoder including the intra prediction apparatus according to the first embodiment of the present invention. The decoder 3 switches between a variable length decoding unit 41, an inverse quantization unit 42, an inverse orthogonal transform unit 43, an addition unit 44, an intra prediction device 1, a frame memory 45, and a motion compensation prediction unit 46. A switch 47 and a rearrangement unit 48 are provided. The decoder 3 is a conventional MPEG-4 AVC / H. An intra prediction device 1 according to the present invention is provided instead of the intra prediction unit of the H.264 decoder.

  The variable length decoding unit 41 inputs a bit stream encoded by inter-frame prediction, performs variable length decoding processing, outputs the bit stream to the inverse quantization unit 42, decodes motion vector information, and motion compensation prediction unit Output to 46.

  The inverse quantization unit 42 performs an inverse quantization process on the quantized orthogonal transform coefficient input from the variable length decoding unit 41, acquires an orthogonal transform coefficient of the motion compensated difference image, and performs an inverse orthogonal transform To the unit 43.

  The inverse orthogonal transform unit 43 performs inverse orthogonal transform (for example, IDCT) on the orthogonal transform coefficient of the difference image input from the inverse quantization unit 42, and outputs the obtained difference image to the addition unit 44.

  The addition unit 44 adds the difference image obtained from the inverse orthogonal transform unit 43 and the prediction image input from the motion compensation prediction unit 46 or the prediction image input from the intra prediction device 1 to restore the image, The data is output to the rearrangement unit 48, the intra prediction apparatus 1, and the frame memory 45.

  The motion compensation prediction unit 46 generates a prediction image using the reference image obtained from the frame memory 45 and the motion vector obtained from the variable length decoding unit 41, and outputs the prediction image to the addition unit 34 via the changeover switch 47.

  The rearrangement unit 48 rearranges the frames of the restored decoded image signal input from the addition unit 44 in the order of display frames and outputs the rearranged frames.

  The intra prediction apparatus 1 inputs an image decoded through the inverse quantization unit 42 and the inverse orthogonal transform unit 43, and as described above, the indirect reference pixel difference calculation unit 11, the coefficient optimization unit 12, and the direct reference pixel extraction unit 13 The prediction image generated through the non-directional prediction unit 14 is output to the addition unit 47 via the changeover switch 47.

  When the same optimization means is used in the encoder 2 and the decoder 3, the values of the intra prediction value on the encoder side and the intra prediction value on the decoder side can be made the same. The optimization means may not be a means for strictly obtaining an optimum value as long as the same result is obtained by the encoder and the decoder.

  The intra prediction apparatus 1 for the encoder according to the present invention shown in FIG. 8 and the intra prediction apparatus 1 for the encoder according to the present invention of the decoder shown in FIG. Of course, the device 1 can be replaced.

  Next, the intra prediction apparatus of Example 2 by this invention is demonstrated with reference to drawings. In addition, the same reference number is attached | subjected to the component similar to Example 1, and description is abbreviate | omitted.

  When the difference between the indirect reference pixels existing on the edge estimation straight line L is large, the intra prediction value by this method may not be appropriate. Therefore, when the indirect reference pixel difference is large, the intra prediction value is switched to the intra prediction value using the pixel value of the reference block by a conventional prediction method (for example, MPEG-4 AVC / H.264 method). In this case, since the same difference can be calculated by both the encoder and the decoder, it is not necessary to transmit a flag for switching between the intra prediction value according to the present invention and the conventional intra prediction value to the decoder.

  FIG. 10 is a configuration diagram of the intra prediction apparatus 1 according to the second embodiment of the present invention. The intra prediction apparatus 1 according to the second embodiment includes an indirect reference pixel difference calculation unit 11, a coefficient optimization unit 12, a direct reference pixel extraction unit 13, a non-directional prediction unit 14, an in-screen prediction unit 15, and a switching unit. 16. Compared with the intra prediction apparatus 1 (see FIG. 5) according to the first embodiment, the intra prediction apparatus 1 according to the second embodiment further includes an in-screen prediction unit 15 and a switching unit 16, and an indirect reference pixel difference calculation unit. 11 differs in that a determination signal is output to the switching unit 16.

  After calculating the indirect reference pixel difference, the indirect reference pixel difference calculation unit 11 determines whether the calculated indirect reference pixel difference is equal to or less than a threshold value, and outputs the determination result to the switching unit 16 as a determination signal.

  The intra-screen prediction unit 15 receives the decoded pixel value and performs prediction using the pixel values of the reference blocks A to D by a prediction method of selecting a prediction direction in the past (for example, MPEG-4 AVC / H.264 method). A predicted image of block M is obtained.

  The switching unit 16 outputs the prediction pixel value obtained by the non-directional prediction unit 14 when the indirect reference pixel difference is equal to or smaller than the threshold value based on the determination signal input from the indirect reference pixel difference calculation unit 11, and the indirect reference pixel When the difference is larger than the threshold, switching is performed so that the predicted pixel value obtained by the in-screen prediction unit 15 is output.

  In addition, as a first modification of the intra prediction device 1 according to the second embodiment, the indirect reference pixel difference calculation unit 11 does not perform the determination, but the prediction pixel value obtained by the non-directional prediction unit 14 and the reference blocks A to D. A determination unit that determines whether or not the difference from the pixel value is larger than the threshold value is separately provided. When the difference is larger than the threshold value, the prediction pixel value obtained by the in-screen prediction unit 15 is output by the switching unit 16. You can also.

  FIG. 11 illustrates a second modification of the intra prediction device 1 according to the second embodiment. In this modified example, the original image of the prediction block M and the prediction image obtained by the non-directional prediction unit 14 are input to the determination unit 17, and whether or not the difference between the pixel value of the original image and the pixel value of the prediction image is greater than a threshold value. If it is larger than the threshold, the switching unit 16 outputs the predicted image obtained by the in-screen prediction unit 15. In this case, the determination unit 17 also has a function of transmitting a flag indicating the determination result to the decoder. Further, the determination unit 17 of the decoder does not need to calculate an error, and has only a function of operating the switching unit 16 according to the received determination result.

  FIG. 12 illustrates a third modification of the intra prediction device 1 according to the second embodiment. In this modified example, the original image of the prediction block M, the prediction image obtained by the non-direction prediction unit 14, and the prediction image obtained by the in-screen prediction unit 15 are input to the determination unit 17 and obtained by the non-direction prediction unit 14. It is determined which of the pixel value of the predicted image and the pixel value of the predicted image obtained by the in-screen prediction unit 15 has a smaller difference from the pixel value of the original image, and the switching unit 16 determines the difference from the pixel value of the original image. The smaller predicted image is output. Also in this case, the determination unit 17 has a function of transmitting a flag indicating the determination result to the decoder. Further, the determination unit 17 of the decoder does not need to calculate an error, and has only a function of operating the switching unit 16 according to the received determination result.

  Next, the operation of the intra prediction apparatus 1 according to the second embodiment will be described with reference to FIG.

[Operation of Intra Prediction Device 1 of Embodiment 2]
FIG. 14 is a flowchart illustrating the operation of the intra prediction apparatus 1 according to the second embodiment. In the same manner as the operation of the intra prediction apparatus 1 according to the first embodiment (see FIG. 7), indirect reference pixel difference calculation unit 11 calculates an indirect reference pixel difference in step S201.

  In step S202, it is determined whether or not the indirect reference pixel difference exceeds a threshold value. If it is determined that the indirect reference pixel difference is equal to or smaller than the threshold value, the process proceeds to step S203. If it is determined that the indirect reference pixel difference is greater than the threshold value, the process proceeds to step S206.

  When it is determined in step S202 that the indirect reference pixel difference is equal to or smaller than the threshold value, in the same manner as the operation of the intra prediction apparatus 1 according to the first embodiment (see FIG. 7), the coefficient optimization unit 12 performs the operation in step S203. The coefficient of the edge estimation straight line L is derived using indirect reference pixel difference calculation, and the direct reference pixel passing through the estimated straight line L is extracted by the direct reference pixel extraction unit 13 in step S204. In step S <b> 205, the switching unit 16 outputs a prediction image input from the non-directional prediction unit 14.

  On the other hand, when it is determined in step S202 that the indirect reference pixel difference is larger than the threshold value, in step S206, the switching unit 16 performs another method (for example, the conventional MPEG-4 AVC / H.264 method). The generated predicted image is output. In addition, the prediction image by another system is produced | generated by the prediction part 15 in a screen in parallel with the production | generation of the prediction image using the edge estimation straight line L. FIG.

  According to the intra prediction device 1 of the second embodiment, a prediction image generated using the edge estimation straight line L and a prediction image generated by selecting a conventional direction can be switched, so that a more appropriate prediction image can be obtained. Be able to generate.

  Next, an intra prediction apparatus according to a third embodiment of the present invention will be described with reference to the drawings. In addition, the same reference number is attached | subjected to the component similar to Example 2, and description is abbreviate | omitted.

  The intra prediction value using the edge estimation straight line L may not be appropriate for a gentle image that does not include an edge, such as a gradation image. Therefore, the intra prediction apparatus according to the second embodiment also performs intra prediction using another method, and selects the most appropriate prediction image.

  FIG. 13 is a configuration diagram of the intra prediction apparatus 1 according to the third embodiment of the present invention. The intra prediction apparatus 1 according to the third embodiment includes an indirect reference pixel difference calculation unit 11, a coefficient optimization unit 12, a direct reference pixel extraction unit 13, a non-directional prediction unit 14, an in-screen prediction unit 15, and a switching unit. 16 and a second intra-screen prediction unit 18. The intra prediction apparatus 1 according to the third embodiment is different from the intra prediction apparatus 1 according to the second embodiment (see FIG. 10) in that it further includes a second intra-screen prediction unit 18.

  The second intra-screen prediction unit 18 employs a method capable of intra prediction suitable for an image that does not include an edge. For example, the pixel value of the prediction block M at the position of the coordinates (x, y) can be predicted using the prediction pixel value curved surface W that is a curved surface including pixel values approximating the pixel values of the reference block. FIG. 17 is a diagram illustrating a predicted pixel value curved surface. The curved surface w on the predicted pixel value curved surface W is a curved surface composed of the predicted pixel values of the predicted block M. If the predicted pixel value curved surface W can be defined, the predicted pixel value z (x, y) can be obtained by substituting the coordinates (x, y) of the pixels of the predicted block M.

  Further, the second intra-screen prediction unit 18 determines whether or not the error of the generated predicted image is equal to or less than the threshold value, and outputs the determination result to the switching unit 16 as a determination signal. The error is, for example, the difference between the pixel value of the predicted image and the pixel values of the reference blocks A to D. Further, the pixels for which the error is to be calculated may be all or some of the pixels in the reference blocks A to D. When only some pixels are used, it is preferable to use pixels close to the prediction block M.

  Next, operation | movement of the intra prediction apparatus 1 of Example 3 is demonstrated with reference to FIG.

[Operation of Intra Prediction Device 1 of Embodiment 3]
FIG. 15 is a flowchart illustrating the operation of the intra prediction apparatus 1 according to the third embodiment. Steps S301, S302, S303, S304, and S305 are the same as steps S201, S202, S203, S204, and S205 of the operation of the intra prediction apparatus 1 according to the second embodiment (see FIG. 14), respectively, and thus description thereof is omitted. The case where it is determined in step S302 that the indirect reference pixel difference is larger than the threshold will be described.

  If it is determined in step S302 that the indirect reference pixel difference is greater than the threshold value, first, in step S306, a prediction image is generated by the second intra-screen prediction unit 18, and an error of the prediction image is obtained.

  If it is determined in step S307 that the error of the predicted image generated by the second intra prediction unit 18 is equal to or less than the threshold, the process proceeds to step S308, and if it is determined that the error is greater than the threshold. The process proceeds to step S309.

  In step S308, the switching unit 16 outputs a predicted image generated by the first other method. In addition, the prediction image by another system is produced | generated by the 2nd intra-screen prediction part 18 in parallel with the production | generation of the prediction image using the edge estimation straight line L. FIG.

  In step S309, the switching unit 16 outputs a prediction image generated by a direction selection method for selecting a prediction direction. Note that the predicted image by the direction selection method is generated by the intra-screen prediction unit 15 in parallel with the generation of the predicted image using the edge estimation straight line L.

  In the flowchart of FIG. 15, the error of the predicted image generated using the edge estimation straight line L is determined, and when the error is equal to or smaller than the threshold (step S302—Yes), the prediction generated using the edge estimation straight line L is determined. When the image is output (step S305) and the error is larger than the threshold (step S302-No), the error of the predicted image generated by another method is determined (step S307), but this order is changed. Is also possible. That is, the error of the predicted image generated by the other method is determined first, and when the error is equal to or smaller than the threshold, the predicted image generated by the other method is output. You may determine the error of the estimated image produced | generated using the estimated straight line L. FIG.

  According to the intra prediction device 1 of the third embodiment, a predicted image generated using the edge estimation straight line L, a predicted image generated by the first other method (for example, a method using the predicted pixel value curved surface W), 2 can switch between predicted images generated by other methods (for example, MPEG-4 AVC / H.264 method), so that an appropriate predicted image can be generated even for an image that does not include an edge. become. In addition, when the direction can be estimated even if a gradation region, an edge, or the like is included, it is not necessary to send information on the direction of intra prediction that has been conventionally required.

  Next, an intra prediction apparatus according to a fourth embodiment of the present invention will be described with reference to the drawings. In addition, the same reference number is attached | subjected to the component similar to Example 1, and description is abbreviate | omitted.

  The intra prediction apparatus 1 according to the fourth embodiment uses the position of the indirect reference pixel and the direct reference pixel, the type of the index of the edge estimation straight line L, and optimization means (least square method, linear programming method, neural network). Multiple processing modes for which prediction method is to be calculated from the extracted direct reference pixels and which processing mode to select for each prediction block. Is switched. FIG. 16: is a block diagram which shows the intra prediction apparatus 1 of Example 4 by this invention. Here, an example including a plurality (n) of intra prediction apparatuses 1 according to the first embodiment and the processing mode determination unit 19 is shown, but the intra prediction apparatus 1 includes the intra prediction apparatus 1 according to the second or third embodiment. Of course, it can be set as a structure.

  The processing mode determination unit 19 selects and outputs an optimal prediction image from prediction images generated in a plurality of processing modes. For example, the prediction image with the maximum encoding efficiency or the prediction image with the smallest error between the pixel value of the prediction image and the pixel values of the reference blocks A to D is selected.

  Further, the processing mode determination unit 19 outputs a flag indicating which of the processing modes 1 to n has been selected. However, when the intra prediction apparatus 1 of the present embodiment is used for an encoder and a decoder, it is necessary to transmit a flag if a rule for determining a processing mode is defined between the encoder and the decoder. Absent. For example, when the variance value of the decoded image is large, it is possible to estimate the processing mode from the decoded image, such as selecting a processing mode that uses optimization means with a small influence of error. Further, the predicted pixel value curved surface expression may be separately transmitted at a timing such as for each frame, in addition to selecting from a predetermined one.

  According to the intra prediction apparatus 1 of the fourth embodiment, since a desired predicted image can be selected from among a plurality of processing modes, an optimal predicted image can be generated for each image with respect to various images. Become.

  Here, in order to function as the intra prediction apparatus 1, a computer can be used suitably. The computer that causes the intra prediction apparatus 1 of the first embodiment to function is a control for causing the indirect reference pixel difference calculation unit 11, the coefficient optimization unit 12, the direct reference pixel extraction unit 13, and the non-directional prediction unit 14 to function. Can be realized by a CPU (Central Processing Unit) and a storage unit, and by causing such a computer to execute a predetermined program by the CPU, an indirect reference pixel difference calculation unit 11, a coefficient optimization unit 12, The functions of the direct reference pixel extraction unit 13 and the non-directional prediction unit 14 can be realized.

  Similarly, a computer that functions as the intra prediction apparatus 1 according to the second embodiment includes an indirect reference pixel difference calculation unit 11, a coefficient optimization unit 12, a direct reference pixel extraction unit 13, a non-directional prediction unit 14, and intra-screen prediction. The control unit for causing the unit 15 and the switching unit 16 to function can be realized by a CPU and a storage unit, and by causing the computer to execute a predetermined program by the CPU, the indirect reference pixel difference calculation unit 11 And the function which the coefficient optimization part 12, the direct reference pixel extraction part 13, the non-directional prediction part 14, the in-screen prediction part 15, and the switching part 16 has is realizable.

  Similarly, a computer that functions as the intra prediction apparatus 1 according to the third embodiment includes an indirect reference pixel difference calculation unit 11, a coefficient optimization unit 12, a direct reference pixel extraction unit 13, a non-directional prediction unit 14, and intra-screen prediction. The control unit for causing the unit 15, the switching unit 16, and the second intra-screen prediction unit 18 to function can be realized by the CPU and the storage unit, and by causing the computer to execute a predetermined program by the CPU The indirect reference pixel difference calculation unit 11, the coefficient optimization unit 12, the direct reference pixel extraction unit 13, the non-directional prediction unit 14, the intra-screen prediction unit 15, the switching unit 16, and the second intra-screen prediction unit 18 can be realized.

  Similarly, a computer that functions as the intra prediction apparatus 1 according to the fourth embodiment includes a plurality of indirect reference pixel difference calculation units 11, a plurality of coefficient optimization units 12, a plurality of direct reference pixel extraction units 13, and a plurality of non-directions. A control unit for causing the prediction unit 14 and the processing mode determination unit 19 to function can be realized by a CPU and a storage unit, and by causing the computer to execute a predetermined program, a plurality of coefficient optimization The functions of the conversion unit 12, the plurality of direct reference pixel extraction units 13, the plurality of non-direction prediction units 14, and the processing mode determination unit 19 can be realized.

  Moreover, in order to function as the encoder 2, a computer can be used suitably. Such a computer includes an intra prediction device 1, a rearrangement unit 21, a subtraction unit 22, an orthogonal transformation unit 23, a quantization unit 24, a variable length coding unit 25, an inverse quantization unit 26, A control unit for causing the inverse orthogonal transform unit 27, the changeover switch 28, the frame memory 30, the motion compensation prediction unit 31, and the addition unit 32 to function can be realized by a CPU and a storage unit. By causing the CPU to execute a predetermined program, the intra prediction device 1, the rearrangement unit 21, the subtraction unit 22, the orthogonal transformation unit 23, the quantization unit 24, the variable length coding unit 25, The functions of the inverse quantization unit 26, the inverse orthogonal transform unit 27, the changeover switch 28, the frame memory 30, the motion compensation prediction unit 31, and the addition unit 32 can be realized.

  Furthermore, in order to function as the decoder 3, a computer can be suitably used. Such a computer includes an intra prediction device 1, a variable length decoding unit 41, an inverse quantization unit 42, an inverse orthogonal transform unit 43, an addition unit 44, a frame memory 45, a motion compensation prediction unit 46, A control unit for causing the changeover switch 47 and the rearrangement unit 48 to function can be realized by a CPU and a storage unit, and by causing the computer to execute a predetermined program by the CPU, The variable length decoding unit 41, the inverse quantization unit 42, the inverse orthogonal transform unit 43, the addition unit 44, the frame memory 45, the motion compensation prediction unit 46, the changeover switch 47, and the rearrangement unit 48. Functions can be realized.

  Although the above embodiments have been described as representative examples, it will be apparent to those skilled in the art that many changes and substitutions can be made within the spirit and scope of the invention. Therefore, the present invention should not be construed as being limited by the above-described embodiments, and various modifications and changes can be made without departing from the scope of the claims. For example, the size of each image block is not limited to the above value, and a suitable value can be used according to the image size to which the present invention is applied.

  As described above, according to the present invention, it is possible to perform intra prediction suitable for various images, which is useful for any application for encoding or decoding an image.

DESCRIPTION OF SYMBOLS 1 Intra prediction apparatus 11 Indirect reference pixel difference calculation part 12 Coefficient optimization part 13 Direct reference pixel extraction part 14 Non-directional prediction part 15 In-screen prediction part 16 Switching part 17 Determination part 18 2nd intra-screen prediction part 19 Processing mode determination part 2 Encoder 3 Decoder

Claims (11)

An intra prediction device that generates an image of a prediction block as a prediction image from an image of a decoded reference block adjacent to a prediction block to be predicted,
An edge estimation line that selects pixels included in the reference block as indirect reference pixels and generates an edge estimation line that is a straight line that passes through a plurality of indirect reference pixels and that estimates the edge direction of the image of the prediction block. Generating means;
A direct reference pixel extraction unit that selects a pixel used for calculation of a pixel value of a predicted image from among the indirect reference pixels as a direct reference pixel, and extracts a direct reference pixel existing on the edge estimation line for each pixel of the prediction block When,
Non-directional prediction means for generating a prediction image of a prediction block based on the direct reference pixels;
An intra prediction apparatus comprising: The edge estimation straight line generating unit selects pixels included in the reference block as indirect reference pixels, defines straight lines passing through the plurality of indirect reference pixels, and indicates a difference between the plurality of indirect reference pixels existing on the straight line. Means for calculating the indicator;
Coefficient optimization means for determining the slope of the edge estimation straight line as a straight line that minimizes the index of difference between the indirect reference pixels,
The intra prediction apparatus according to claim 1, comprising:   The coefficient optimizing means uses an index of the difference between the indirect reference pixels as a square sum of differences between adjacent indirect reference pixels existing on the edge estimation line, or an indirect reference pixel existing on the edge estimation line. The intra prediction apparatus according to claim 2, wherein a variance value is used. The direct reference pixel extracting means may comprise a direct reference pixels pixels adjacent to the predicted block of the indirect reference pixels, the intra prediction apparatus according to any one of claims 1 to 3. The non-directional prediction means, when there is one direct reference pixel extracted for each pixel of the prediction block, the value of the direct reference pixel or the average of the direct reference pixel and the pixel adjacent to the direct reference pixel Value or weighted average value as the predicted value,
When there are a plurality of direct reference pixels extracted for each pixel of the prediction block, the direct reference pixel closest to the pixel of the prediction block, the average value or the weighted average value of the extracted direct reference pixels, or the edge estimation line The average value of all the indirect reference pixels and direct reference pixels above is taken as the predicted value,
If there is no direct reference pixel extracted for each pixel of the prediction block, the direct reference pixel closest to the edge estimation line, the average value of the predicted pixels in the vicinity of the pixel of the prediction block, or all The intra prediction apparatus according to claim 1, wherein an average value of direct reference pixels is used as a prediction value. Means for determining whether an index of the difference between the indirect reference pixels exceeds a predetermined value;
In-screen prediction means for generating a predicted image from the image of the reference block;
When the index is less than or equal to the predetermined value, the prediction image generated by the non-directional prediction unit is output, and when the index exceeds the predetermined value, the prediction generated by the intra-screen prediction unit Means for outputting an image;
The intra prediction apparatus according to any one of claims 2 to 5, comprising:   The intra-prediction device according to claim 6, wherein the intra-screen prediction unit generates a prediction image by selecting a prediction direction from the image of the reference block. The edge estimation straight line generation unit, the direct reference pixel extraction unit, and the non-directional prediction unit have a plurality of processing modes, and the position of the indirect reference pixel, the position of the direct reference pixel, or the edge for each processing mode The method for deriving the coefficient of the estimated line is different,
The non-directional prediction means generates a predicted image for each processing mode,
Furthermore, the intra prediction apparatus as described in any one of Claims 1-7 provided with the determination means which selects and outputs a prediction image from the prediction images produced | generated for every said processing mode.   The encoder provided with the intra prediction apparatus as described in any one of Claims 1-8.   A decoder comprising the intra prediction apparatus according to any one of claims 1 to 8. From a decoded reference block image adjacent to a prediction block to be predicted, an image of the prediction block is generated as a predicted image to a computer constituting an intra prediction apparatus.
(A) A pixel included in the reference block is selected as an indirect reference pixel, and an edge estimation line that is a straight line that passes through a plurality of indirect reference pixels and that estimates the edge direction of the image of the prediction block is generated. Steps,
(B) selecting a pixel used for calculating a pixel value of a predicted image from among the indirect reference pixels as a direct reference pixel, and extracting a direct reference pixel existing on the edge estimation line for each pixel of the predicted block; ,
(C) generating a prediction image of a prediction block based on the direct reference pixels;
A program for running

Images