JP2008301270A - Moving image encoding device and moving image encoding method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To determine an MB type of favorable encoding efficiency simply with high accuracy. <P>SOLUTION: A motion vector is searched for by use of a current image frame, an inter-frame prediction evaluation value indicating the encoding efficiency is produced based on an inter-frame prediction error and the motion vector weighted based on a quantization parameter (QP), thus raising a search accuracy for the motion vector searched for at a low bit rate, and the motion vector is reflected in producing the evaluation value, thereby determining the MB type of favorable encoding efficiency simply with high accuracy. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a moving image encoding apparatus, a moving image encoding method, a program, and a recording medium, and more particularly, used to encode a moving image using a spatial correlation within a frame and a temporal correlation between frames. And a suitable technique.

  As a method for compressing and coding moving image data in units of blocks using the correlation between frames or within a frame, MPEG4 or H.264 is used. The H.264 system is known. In these methods, high-quality compression is possible by using an inter-frame prediction method when temporal correlation between screens is high, and by using an intra-frame prediction method when correlation within a screen is high. In particular H. In the H.264 system, a plurality of prediction units or prediction modes are prepared in the inter-frame or intra-frame prediction system as compared with the MPEG4 system, and the compression encoding system is more complicated.

  The inter-frame prediction method is effective when the movement of the subject is relatively small, and an image with a high degree of correlation is cut out from a temporally previous or subsequent frame and a difference from the current macroblock is encoded. At this time, a motion vector indicating which part has been cut out is also encoded.

  On the other hand, when the correlation in the screen is high, the intra-frame prediction method is effective. The intra-frame prediction scheme is a prediction scheme that calculates a prediction error using a correlation with a macroblock adjacent to the current macroblock or a correlation within a macroblock. Therefore, even in a situation where the motion of the subject is large and it is difficult to obtain a motion vector with good coding efficiency by the inter-frame prediction method, the coding can be performed relatively efficiently.

  When these two types of prediction methods are used, a means for determining which method is used to improve the coding efficiency is required. When the processing time is not limited (for example, when already recorded moving image data is compressed), an optimal method may be selected after encoding is performed once in all methods, units, and modes. .

  However, when encoding is performed in real time while shooting with a video camera, the processing contents are complicated by the above method, and mounting using the above method is difficult in view of processing time and mounting cost on a semiconductor. Therefore, a means for selecting an inter-frame coding method or an intra-frame coding method at high speed by a simple calculation instead of the above method is required.

  Thus, a simplified intra-frame prediction / inter-frame prediction encoding determination method is proposed in Non-Patent Document 1. In the method described in Non-Patent Document 1 (hereinafter referred to as “model 3 method”), the encoding efficiency at the time of interframe predictive encoding is estimated from the prediction residual of motion compensation. Then, the coding efficiency at the time of intra-frame coding is estimated from the MB (macroblock) intra-frame prediction residual, and the motion compensation prediction residual is compared with the intra-MB prediction residual. A predictive coding method is determined. This processing method will be described with reference to the flowchart of FIG.

  In step S201 of FIG. 2, the mean square value (VAR) of the prediction residual in the 16 × 16 pixel inter-frame prediction method of the input MB (macroblock) is calculated. Specifically, as shown in the following equation (1), the difference between the input MB signal A [i, j] and the predicted MB signal F [x + i, y + j] referenced by the motion vector (x, y). The mean square value (VAR) of is calculated.

An absolute value may be used instead of the square value.
In the next step S202, a variance value (VAROR) is calculated from the following equation (2) as a prediction residual of the input MB.

  In the next step S203, the mean square value (VAR) calculated in step S201 and the variance value (VAROR) calculated in step S202 are compared from the characteristic diagram shown in FIG. As a result of this comparison, if VAR> THR and VAROR <VAR, the MB mode is set to intraframe prediction encoding (step S204). On the other hand, as a result of the comparison in step S203, in other cases, the MB mode is set to interframe predictive coding (step S205). Here, THR indicates a threshold value, for example, THR = 64.

  Next, in step S206, the input MB is encoded with the designated MB type. Specifically, the prediction error is frequency-converted into a designated prediction encoding method, and entropy encoding is performed.

  On the other hand, Patent Document 1 discloses a determination method obtained by improving Model 3 described in Non-Patent Document 1.

JP-A-10-23434 MPEG Video Simulation Model 3 (ISO / IEC JTC1 / SC2 / WG11 N0010 MPEG90 / 041 July 1990)

  In the above-described conventional example, the inter-frame prediction method searches for a motion vector using a frame that has been encoded and reconstructed (local decoding) and a frame that is the current encoding target, and calculates a predicted value. . The MB evaluation value of the inter-frame prediction method used for the MB type determination is calculated using a motion vector and a prediction value.

  When a motion vector search is performed using such an encoded reconstructed frame, there is a problem that a motion vector searched at a low bit rate has a high probability of being erroneously determined. Therefore, there is a problem that an MB type with good coding efficiency cannot be determined with high accuracy.

  An object of the present invention is to make it possible to easily and accurately determine an MB type having good coding efficiency in view of the above-described problems.

  A moving picture coding apparatus according to the present invention performs coding on a coding target frame using a prediction error by an inter-frame prediction method or an intra-frame prediction method, and generates moving picture coded data. A motion search means for searching for a motion vector using the encoding target frame and a current image frame as a reference frame, a motion vector searched for by the motion search means, and an encoded reconstruction An inter-frame prediction error calculating unit that calculates an inter-frame prediction value from the frame, and calculates an inter-frame prediction error using the calculated inter-frame prediction value and a pixel of the encoding target frame; and the encoding target An intra-frame prediction value is calculated using the reconstructed pixels that have been encoded in the frame, the calculated intra-frame prediction value, and the encoding target frame An intra-frame prediction error calculating means for calculating an intra-frame prediction error using an element, and an intra-frame prediction evaluation value indicating coding efficiency based on the intra-frame prediction error calculated by the intra-frame prediction error calculating means. A frame indicating encoding efficiency based on the inter-frame prediction error calculated by the inter-frame prediction error calculating means and a motion vector weighted according to a quantization parameter indicating a quantization step level. An evaluation value calculating means for calculating an inter prediction evaluation value, and comparing the intra-frame prediction evaluation value calculated by the evaluation value calculating means with the inter-frame prediction evaluation value, And a prediction selection means for selecting one of them.

  The moving picture coding method of the present invention is a moving picture coding method in which coding is performed on a coding target frame using a prediction error by an inter-frame prediction method or an intra-frame prediction method, and moving image coded data is generated. A motion search step of searching for a motion vector using the encoding target frame and a current image frame as a reference frame, a motion vector searched in the motion search step, and an encoded reconstructed frame An inter-frame prediction value is calculated from the inter-frame prediction value using the calculated inter-frame prediction value and the pixel of the encoding target frame, and the encoding target frame. An intra-frame prediction value is calculated using the encoded reconstructed pixels, and the calculated intra-frame prediction value and the image of the encoding target frame are calculated. And calculating an intra-frame prediction error value indicating coding efficiency based on the intra-frame prediction error calculation step for calculating the intra-frame prediction error using the intra-frame prediction error and the intra-frame prediction error calculated in the intra-frame prediction error calculation step. In addition, the inter-frame prediction evaluation indicating the coding efficiency based on the inter-frame prediction error calculated in the inter-frame prediction error calculating step and the motion vector weighted according to the quantization parameter indicating the quantization step level. An evaluation value calculation step for calculating a value, and the intra-frame prediction evaluation value generated in the evaluation value calculation step and the inter-frame prediction evaluation value are compared, and either the inter-frame prediction method or the intra-frame prediction method is selected. And a prediction selection step.

  The program of the present invention is a program that causes a computer to execute coding on a frame to be coded using a prediction error by an inter-frame prediction method or an intra-frame prediction method, and generate moving image coded data. A motion search step of searching for a motion vector using the encoding target frame and the current image frame as a reference frame, a motion vector searched in the motion search step, and an encoded reconstructed frame An inter-frame prediction error calculating step of calculating an inter-frame prediction value, calculating an inter-frame prediction error using the calculated inter-frame prediction value and a pixel of the encoding target frame; and a code of the encoding target frame An intra-frame prediction value is calculated using the converted reconstructed pixels, the calculated intra-frame prediction value, and the code An intra-frame prediction error calculation step for calculating an intra-frame prediction error using pixels of the target frame, and an intra-frame prediction evaluation indicating coding efficiency based on the intra-frame prediction error calculated in the intra-frame prediction error calculation step The coding efficiency is calculated based on the inter-frame prediction error calculated in the inter-frame prediction error calculation step and the motion vector weighted according to the quantization parameter indicating the quantization step level. An evaluation value calculation step for calculating an inter-frame prediction evaluation value, an intra-frame prediction evaluation value calculated in the evaluation value calculation step and an inter-frame prediction evaluation value are compared, and the inter-frame prediction method or the intra-frame prediction method The computer is caused to execute a prediction selection step for selecting one of them.

  The recording medium of the present invention is characterized by recording the program described above.

According to the present invention, a motion vector is searched using the current image frame, and an inter-frame prediction indicating coding efficiency is performed based on an inter-frame prediction error and a motion vector weighted according to a quantization parameter. An evaluation value was calculated. Thereby, the search accuracy of the motion vector searched at the low bit rate can be increased, and the motion vector can be reflected in the calculation of the evaluation value. Therefore, it is possible to easily and accurately determine an MB type with good encoding efficiency.

(First embodiment)
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram illustrating a basic configuration example of a moving image encoding apparatus according to the present embodiment.
1 includes an intra-frame prediction unit 301, an inter-frame prediction unit 302, a motion search unit 303, a first subtractor 304, a second subtractor 305, and a selector 306. . Furthermore, a first reference frame buffer 307, a second reference frame buffer 308, a reference pixel buffer 309, an encoding processing unit 310, a local decoding processing unit 316, and an MB prediction evaluation unit 311 are included.

A procedure for selecting and encoding one prediction method from two prediction methods of an intra-frame prediction method and an inter-frame prediction method in MB units with reference to the block diagram shown in FIG. 1 and the flowchart shown in FIG. Will be described.
First, the encoding target frame in the moving image sequence is input to the motion search unit 303, the first subtracter 304, the second subtracter 305, and the first reference frame buffer 307. The encoding target frame is divided into MB units and sequentially input.

  First, in step S <b> 301, the motion search unit 303 searches for a motion vector using the input encoding target frame and the reference frame read from the first reference frame buffer 307. The searched motion vector is output to the inter-frame prediction unit 302, the MB prediction evaluation unit 311, and the encoding processing unit 310.

  Next, in step S302, the intra-frame prediction unit 301 and the first subtracter 304 function as an intra-frame prediction error calculation unit, and calculate an intra-frame prediction error. Further, the inter-frame prediction unit 302 and the second subtracter 305 function as an inter-frame prediction error calculation unit, and calculate an inter-frame prediction error. As a specific procedure, first, the intra-frame prediction unit 301 uses one or more reference pixels in the same frame read from the reference pixel buffer 309 to calculate an intra-frame prediction value of the encoding target pixel. To the first subtracter 304. Note that the calculation of the intra-frame prediction value in the intra-frame prediction unit 301 is performed in units of rectangular areas.

  On the other hand, the inter-frame prediction unit 302 uses the reference frame read from the second reference frame buffer 308 and the motion vector output from the motion search unit 303 to predict the inter-frame prediction value of the MB to be encoded. Is calculated and output to the second subtractor 305.

  The first subtracter 304 receives the intra-frame prediction value output from the intra-frame prediction unit 301 and the pixel value of the encoding target MB. Then, a difference value between the intra-frame prediction value and the pixel value is output to the selector 306 and the MB prediction evaluation unit 311 as an intra-frame prediction error. The second subtracter 305 receives the inter-frame prediction value output from the inter-frame prediction unit 302 and the pixel value of the encoding target MB. Then, a difference value between the inter-frame prediction value and the pixel value is output to the selector 306 and the MB prediction evaluation unit 311 as an inter-frame prediction error.

  Next, in step S303, the MB prediction evaluation unit 311 performs evaluation for determining one of the intra-frame prediction method and the inter-frame prediction method when generating a moving image encoded stream. Then, a signal indicating the MB type is output to the selector 306. The MB prediction evaluation unit 311 receives a quantization parameter (QP), a quantization parameter threshold value (QPth), and a search range (f_code). Furthermore, the motion vector output from the motion search unit 303 is input, and the prediction error output from the first subtractor 305 and the second subtractor 305 is input. A specific configuration of the MB prediction evaluation unit 311 will be described later.

  The selector 306 determines the intra-frame prediction error output from the first subtractor 304 or the inter-frame prediction output from the second subtractor 305 according to the signal indicating the MB type output from the MB prediction evaluation unit 311. One of the errors is selected and output to the encoding processing unit 310.

  In step S304, the encoding processing unit 310 performs encoding processing using the prediction error output from the selector 306, and generates moving image encoded data. On the other hand, the output of the encoding processing unit 310 is input to the local decoding processing unit 316. The local decoding processing unit 316 decodes the encoded image data and outputs the decoded image data to the reference frame buffer 308 and the reference pixel buffer 309 as a local decoded image.

  Next, functions of the three buffers used for the above-described processing will be described. The first reference frame buffer 307 stores frames in the moving image sequence. In addition, the motion search unit 303 has a function of reading the current image frame stored in the first reference frame buffer 307 as a reference frame in order to search for a motion vector. Note that the frame stored in the first reference frame buffer 307 is a frame in a state where the frame of the moving image sequence is not processed. However, if the frame is not a reconstructed frame that has been subjected to the encoding process and the local decoding process, it may be a frame that has been subjected to preprocessing such as changing the resolution or gradation.

  On the other hand, the second reference frame buffer 308 stores the reconstructed frame locally decoded by the local decoding processing unit 316. The inter-frame prediction unit 302 has a function of reading out a frame stored in the second reference frame buffer 308 as a reference frame in order to calculate a prediction value.

  Further, the reference pixel buffer 309 stores the reconstructed pixels that are locally decoded by the local decoding processing unit 316. The intra-frame prediction unit 301 has a function of reading out pixels stored in the reference pixel buffer 309 as reference pixels in order to calculate a prediction value.

  Next, an example of a specific configuration and processing procedure of the MB prediction evaluation unit 311 will be described with reference to the flowchart of FIG. FIG. 4 is a flowchart illustrating an example of a detailed procedure of the process in step S303 in FIG. The MB prediction evaluation unit 311 includes a weighting coefficient calculation unit 312, a first SAD calculator 313, a second SAD calculator 314, and an MB type determination unit 315.

  First, in step S401, the weighting coefficient calculation unit 312 determines the weighting coefficient (α) from the quantization parameter (QP) and the motion search range (f_code), and outputs the weighting coefficient (α) to the MB type determination unit 315. Note that the quantization parameter (QP) is a parameter representing the quantization step level, and the bit rate becomes lower as the quantization parameter (QP) increases. The motion search range (f_code) is a range for searching for a motion vector. When the search range is 16 × 16, the motion search range (f_code) is defined as 1. In the case of 32 × 32, the motion search range (f_code) is defined as 2, and in the case of 64 × 64, the motion search range (f_code) is defined as 3 or the like. A detailed calculation method of the weighting coefficient in the weighting coefficient calculation unit 312 will be described later.

  Next, in step S402, an absolute difference sum (SAD) of MBs is calculated. Specifically, the first SAD calculator 313 calculates an absolute difference sum (SAD) of MBs from the intra-frame prediction error output from the first subtractor 304 and outputs it to the MB type determination unit 315. . In addition, the second SAD calculator 314 calculates an absolute difference sum (SAD) of MBs from the inter-frame prediction error output from the second subtractor 305 and outputs it to the MB type determination unit 315.

  The MB type determination unit 315 receives the weighting coefficient output from the weighting coefficient calculation unit 312 and the motion vector output from the motion search unit 303. Further, the sum of absolute differences (SAD) of MBs output from the first SAD calculator 313 and the second SAD calculator 314 is input. Therefore, next, in step S403, the MB type determination unit 315 functions as an evaluation value calculation unit, and uses the input signal to evaluate each coding efficiency of the intra-frame prediction method and the inter-frame prediction method. Calculate the value. In step S404, the MB type determination unit 315 functions as a prediction selection unit, and outputs a prediction method having a small evaluation value to the selector 306 as an MB type.

  Next, an evaluation value calculation method in the MB type determination unit 315 will be described. Note that the evaluation value described below is a numerical model that simulates or approximates the coding efficiency of each prediction method with the code amount, coding distortion, and the like.

The intra-frame prediction evaluation value (COST _INTRA ) is obtained by directly calculating the sum of absolute differences (SAD _INTRA ) of the _intra- frame prediction error output from the first SAD calculator 313 as shown in the following equation (4). Used as

On the other hand, the calculation method of the inter-frame prediction evaluation value (COST _INTER ) is shown in the following formula (5).

In the equation (5), the inter-frame prediction evaluation value includes the sum of absolute differences (SAD _INTER ) of the inter-frame prediction error output from the second SAD calculator 314 and the motion vector output from the motion search unit 303. (MV). Note that α shown in Expression (5) is a weighting coefficient, and is determined by the quantization coefficient (QP) and the motion search range (f_code).

The weighting coefficient α decreases as the motion search range (f_code) increases, and increases as the motion search range decreases. Therefore, the weighting coefficient α represents the relationship between the motion vector and the encoding efficiency. Here, the sum of absolute differences (SAD _INTER ) approximates the entropy code amount of the inter-frame prediction error. However, at a low bit rate, the entropy code amount of inter-frame prediction error tends to decrease. On the other hand, the motion vector does not depend on the bit rate, and a unique code amount is generated according to the motion vector value. Therefore, when the quantization parameter (QP) becomes larger than the threshold value, the bit rate becomes low, so the weighting coefficient α is increased.

  In the present embodiment, since the original encoded current image frame is used for motion search, encoding distortion does not cause a reduction in accuracy of motion detection in motion vector search. Therefore, the accuracy of the motion vector is very high compared to the case where motion search is performed using a reconstructed frame obtained by locally decoding an encoded frame.

  However, in a moving image sequence that pans at high speed, the motion search range is related to the motion vector search accuracy. In a high-speed panning video sequence, when the motion search range is high, the accuracy of the motion vector is high and the coding efficiency is increased. On the other hand, when the motion search range is a narrow range, the motion vector search accuracy is lowered. Therefore, the motion vector weighting in the inter-frame prediction evaluation value may be changed according to the motion vector search range. In this case, when the motion vector search range (f_code) is small, the motion vector weighting is increased. As a result, the inter-frame prediction evaluation value becomes large and the intra-frame prediction method is selected, and it is possible to suppress image quality deterioration detected by the viewer as block noise or flicker.

  As described above, in this embodiment, a motion vector is searched using the current image frame, and encoding is performed based on the inter-frame prediction error and the motion vector weighted based on the quantization parameter (QP). An inter-frame prediction evaluation value indicating efficiency is calculated. Thereby, the search accuracy of the motion vector searched at the low bit rate can be increased, and the motion vector can be reflected in the calculation of the evaluation value. Therefore, it is possible to determine an MB type with good coding efficiency with high accuracy.

  Further, since the weighting of the motion vector in the inter-frame prediction evaluation value is changed according to the motion vector search range, it is possible to suppress visual degradation without reducing the encoding efficiency.

(Other embodiments according to the present invention)
Each means constituting the moving picture coding apparatus and each step of the moving picture coding method in the embodiment of the present invention described above can be realized by operating a program stored in a RAM or ROM of a computer. This program and a computer-readable recording medium recording the program are included in the present invention.

  Further, the present invention can be implemented as, for example, a system, apparatus, method, program, or recording medium. Specifically, the present invention may be applied to a system including a plurality of devices. The present invention may be applied to an apparatus composed of a single device.

  Note that the present invention includes a case where a software program (in the embodiment, a program corresponding to the flowcharts shown in FIGS. 3 and 4) for realizing the functions of the above-described embodiments is directly or remotely supplied to a system or apparatus. . This includes the case where the system or the computer of the apparatus is also achieved by reading and executing the supplied program code.

  Accordingly, since the functions of the present invention are implemented by computer, the program code installed in the computer also implements the present invention. In other words, the present invention includes a computer program itself for realizing the functional processing of the present invention.

  In that case, as long as it has the function of a program, it may be in the form of object code, a program executed by an interpreter, script data supplied to the OS, and the like.

  Examples of the recording medium for supplying the program include a flexible disk, a hard disk, an optical disk, and a magneto-optical disk. Further, there are MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card, ROM, DVD (DVD-ROM, DVD-R) and the like.

  As another program supply method, there is a method of connecting to a homepage on the Internet using a browser of a client computer. The computer program itself of the present invention or a compressed file including an automatic installation function can be downloaded from the homepage by downloading it to a recording medium such as a hard disk.

  It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the present invention.

  As another method, the program of the present invention is encrypted, stored in a recording medium such as a CD-ROM, distributed to users, and encrypted from a homepage via the Internet to users who have cleared predetermined conditions. Download the key information to be solved. It is also possible to execute the encrypted program by using the key information and install the program on a computer.

  Further, the functions of the above-described embodiments are realized by the computer executing the read program. Furthermore, based on the instructions of the program, an OS or the like running on the computer performs part or all of the actual processing, and the functions of the above-described embodiments can be realized by the processing.

  As another method, the program read from the recording medium is first written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer. Then, based on the instructions of the program, the CPU or the like provided in the function expansion board or function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are also realized by the processing.

It is a block diagram which shows the function structural example of the moving image encoder in embodiment of this invention. It is a flowchart which shows an example of the process sequence which selects and selects a prediction method in the past. In the embodiment of the present invention, it is a flowchart which shows an example of a processing procedure which chooses a prediction method and performs coding. It is a flowchart which shows an example of the detailed procedure of the process of step S303 of FIG. It is a figure which shows MB type determination characteristic.

Explanation of symbols

DESCRIPTION OF SYMBOLS 300 Moving image encoder 301 Intra-frame prediction part 302 Inter-frame prediction part 303 Motion search part 304 1st subtractor 305 2nd subtractor 306 Selector 307 1st reference frame buffer 308 2nd reference frame buffer 309 Reference pixel buffer 310 Encoding processing unit 311 MB prediction evaluation unit 312 Weighting coefficient calculation unit 313 First SAD calculator 314 Second SAD calculator 315 MB type determination unit 316 Local decoding processing unit

Claims (6)

A moving picture coding apparatus that performs coding on a coding target frame using a prediction error by an inter-frame prediction method or an intra-frame prediction method, and generates moving image coded data,
Motion search means for searching for a motion vector using the encoding target frame and a current image frame as a reference frame;
An inter-frame prediction value is calculated from the motion vector searched by the motion search means and the encoded reconstructed frame, and a frame is generated using the calculated inter-frame prediction value and the pixel of the encoding target frame. An inter-frame prediction error calculating means for calculating an inter-prediction error;
An intra-frame prediction value is calculated using the encoded reconstructed pixel of the encoding target frame, and an intra-frame prediction error is calculated using the calculated intra-frame prediction value and the pixel of the encoding target frame. Intra-frame prediction error calculating means
Based on the intra-frame prediction error calculated by the intra-frame prediction error calculating means, an intra-frame prediction evaluation value indicating coding efficiency is calculated, and the inter-frame prediction error calculated by the inter-frame prediction error calculating means Evaluation value calculating means for calculating an inter-frame prediction evaluation value indicating coding efficiency based on a motion vector weighted according to a quantization parameter representing a quantization step level;
A prediction selection unit that compares the intra-frame prediction evaluation value calculated by the evaluation value calculation unit with the inter-frame prediction evaluation value, and selects either the inter-frame prediction method or the intra-frame prediction method. A moving image encoding device.   2. The moving picture encoding apparatus according to claim 1, wherein the evaluation value calculating unit weights the motion vector according to a motion search range and calculates the inter-frame prediction evaluation value. A video encoding method that performs encoding on a frame to be encoded using a prediction error by an inter-frame prediction method or an intra-frame prediction method, and generates moving image encoded data,
A motion search step of searching for a motion vector using the encoding target frame and a current image frame as a reference frame;
An inter-frame prediction value is calculated from the motion vector searched in the motion search step and the encoded reconstructed frame, and the inter-frame prediction value is calculated using the calculated inter-frame prediction value and the pixel of the encoding target frame. An inter-frame prediction error calculation step for calculating a prediction error;
An intra-frame prediction value is calculated using the encoded reconstructed pixel of the encoding target frame, and an intra-frame prediction error is calculated using the calculated intra-frame prediction value and the pixel of the encoding target frame. An intra-frame prediction error calculating step,
Based on the intra-frame prediction error calculated in the intra-frame prediction error calculation step, an intra-frame prediction evaluation value indicating coding efficiency is calculated, and the inter-frame prediction error calculated in the inter-frame prediction error calculation step and a quantum An evaluation value calculation step for calculating an inter-frame prediction evaluation value indicating coding efficiency based on a motion vector weighted according to a quantization parameter representing a coding step level;
A prediction selection step of selecting either the inter-frame prediction method or the intra-frame prediction method by comparing the intra-frame prediction evaluation value and the inter-frame prediction evaluation value generated in the evaluation value calculation step. A video encoding method.   4. The moving picture encoding method according to claim 3, wherein, in the evaluation value calculating step, the inter-frame prediction evaluation value is calculated by weighting the motion vector according to a motion search range. A program for performing encoding on a frame to be encoded using a prediction error by an inter-frame prediction method or an intra-frame prediction method, and causing a computer to generate moving image encoded data,
A motion search step of searching for a motion vector using the encoding target frame and a current image frame as a reference frame;
An inter-frame prediction value is calculated from the motion vector searched in the motion search step and the encoded reconstructed frame, and the inter-frame prediction value is calculated using the calculated inter-frame prediction value and the pixel of the encoding target frame. An inter-frame prediction error calculation step for calculating a prediction error;
An intra-frame prediction value is calculated using the encoded reconstructed pixel of the encoding target frame, and an intra-frame prediction error is calculated using the calculated intra-frame prediction value and the pixel of the encoding target frame. An intra-frame prediction error calculating step,
Based on the intra-frame prediction error calculated in the intra-frame prediction error calculation step, an intra-frame prediction evaluation value indicating coding efficiency is calculated, and the inter-frame prediction error calculated in the inter-frame prediction error calculation step and a quantum An evaluation value calculation step for calculating an inter-frame prediction evaluation value indicating coding efficiency based on a motion vector weighted according to a quantization parameter representing a coding step level;
Comparing the intra-frame prediction evaluation value calculated in the evaluation value calculation step with the inter-frame prediction evaluation value, and causing the computer to execute a prediction selection step of selecting either the inter-frame prediction method or the intra-frame prediction method. A program characterized by that.   A computer-readable recording medium on which the program according to claim 5 is recorded.

Images