JP4590337B2 - Moving picture coding apparatus and moving picture coding method - Google Patents

Description

  The present invention relates to a moving image encoding apparatus and a moving image encoding method, and more particularly to a technique suitable for use in a camera-integrated moving image recording apparatus that compresses and records a moving image.

  2. Description of the Related Art Conventionally, for example, a digital video camera is well known as a camera-integrated moving image recording apparatus that captures a subject and compresses and records a moving image obtained thereby. In recent years, recording media have been changed from conventional magnetic tapes to disk media, semiconductor memories, and the like due to high convenience such as random accessibility. However, in general, a disk medium or the like has a small storage capacity and needs to be compressed and encoded more efficiently.

  In addition, with the expectation of high image quality, development of digital video cameras that handle high-definition video with a large amount of information is being carried out. Thus, more efficient compression coding is desired from another viewpoint. . Currently, MPEG is often used as a standard method for highly efficient compression coding.

  Furthermore, in recent years, there has been an increasing need for encoding at a low bit rate for portable terminals and further improvement in the recordable time on a recording medium, and higher efficiency encoding has been studied. One of them is H. It is H.264, and a large amount of calculation is required for encoding and decoding as compared with conventional encoding methods such as MPEG2 and MPEG4, but it is known that higher encoding efficiency is realized.

  H. In H.264, motion prediction / compensation processing is performed with high pixel accuracy such as 1/4 pixel accuracy in order to increase encoding efficiency. In this motion prediction / compensation processing, image data with integer pixel accuracy is read from the frame memory, interpolation processing is performed to generate pixel values with decimal pixel accuracy, and motion prediction / compensation is performed using the generated image data with decimal pixel accuracy. Processing is performed. The interpolation process is performed by high-order filtering in order to generate a more accurate interpolation signal. H. The H.264 calculation method is disclosed in Patent Document 1, for example.

JP 2004-56827 A

  However, as described above, since interpolation processing is performed by high-order filtering, the frequency of reading out image data from the frame memory becomes very high. As a result, there is a problem that a large-scale and expensive frame memory having a wide bandwidth is required and power consumption is increased.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to reduce power consumption by reducing a bandwidth necessary for motion detection.

Moving picture coding apparatus of the present invention, a motion vector detecting means for detecting a motion vector between the see image and sign-encoding target image in sub-pixel units, according to the motion vector detected by said motion vector detecting means A motion image encoding device having motion compensation means for performing motion compensation, wherein the motion vector detection means reads out and stores a pixel value at an integer pixel position of a reference image stored in a frame memory; A first interpolation pixel generation unit that interpolates and generates a pixel value at a decimal pixel position by a filter having a predetermined number of taps using a pixel value at an integer pixel position stored in the storage unit; Second, the pixel value at the decimal pixel position is generated by interpolation using a filter having a smaller number of taps than the first interpolation pixel generation means using the pixel value at the integer pixel position. A pixel interpolating unit, a pixel value of the decimal pixel position is generated by one of said first and second pixel interpolating unit is selected as the pixel value for forming a reference image of the decimal pixel precision interpolation pixel Selection means and reference image forming means for forming a reference image with decimal pixel precision used for motion vector detection in the decimal pixel unit from the pixel value selected by the interpolation pixel selection means, and the storage means the integer pixel using the pixel value of the stored integer pixel position from the motion vector detection in units, the detection of the motion vector with fractional pixel units by changing stepwise detection accuracy a row Umono, wherein when the motion vector detecting means performs motion vector detection in units of sub-pixel, the interpolation pixel selecting means, only the pixel value of an integer pixel position stored in the storage unit When the interpolation process by the first interpolation pixel generation unit cannot be performed, the pixel value is generated by the second interpolation pixel generation unit instead of the pixel value at the decimal pixel position generated by the first interpolation pixel generation unit. Switching is performed so that the pixel value at the decimal pixel position is selected .

Moving picture coding method of the present invention, a motion vector detection step of detecting a motion vector between the see image and sign-encoding target image in sub-pixel units, according to the motion vector detected by said motion vector detection step A motion compensation method for performing motion compensation, wherein the motion vector detection step reads and stores a pixel value at an integer pixel position of a reference image stored in a frame memory; A first interpolation pixel generation step of interpolating and generating a pixel value at a decimal pixel position using a filter with a predetermined number of taps using the pixel value at an integer pixel position stored in the storage step; Using the pixel value at the integer pixel position, the pixel value at the decimal pixel position is generated by interpolation with a filter having a smaller number of taps than in the first interpolation pixel generation step. A second interpolation pixel generating step, pixel values of the fractional pixel position is generated by one of said first and second interpolation pixel generation process, selected as pixel values for forming a reference image of the sub-pixel accuracy a interpolation pixel selecting step, from the pixel values selected by the interpolation pixel selecting step and the reference image forming step of forming a reference image of the sub-pixel accuracy to be used in the motion vector detection in the sub-pixel units, the from the motion vector detection integer pixel using the pixel value of the stored integer pixel positions in units in the storage step, a detection of motion vectors in units of sub-pixel is changed stepwise detection accuracy in line cormorants method Te, the motion when the vector detection step detecting the motion vector in units of sub-pixel, the interpolation pixel selecting step, said integer pixel position stored in the storing step When the interpolation process by the first interpolation pixel generation process cannot be performed with only the pixel value of the second interpolation pixel generation, the pixel value of the decimal pixel position generated by the first interpolation pixel generation process is replaced with the pixel value of the second interpolation pixel generation process. Switching is made so that the pixel value at the decimal pixel position generated in the process is selected .

Computer program of the present invention, a motion vector detection step of detecting a motion vector between the see image and sign-encoding target image in sub-pixel units, motion compensation according to the motion vector detected by said motion vector detection step A motion compensation method for performing a computer program for causing a computer to execute each step of the moving image encoding method, wherein the motion vector detection step uses a pixel value at an integer pixel position of a reference image stored in a frame memory. A storage step of reading and storing, and a first interpolation pixel generation step of interpolating and generating a pixel value of a decimal pixel position by a filter having a predetermined number of taps using the pixel value of the integer pixel position stored in the storage step; , using the pixel value of the stored integer pixel positions in the storing step, less than the first interpolation pixel generation step A second interpolation pixel generation process for interpolation generates the pixel value of the sub-pel position by the number of taps of the filter have the pixel value of the decimal pixel position is generated by one of said first and second interpolation pixel generation process An interpolation pixel selection step for selecting as a pixel value for forming a reference image with decimal pixel accuracy , and a decimal pixel accuracy used in motion vector detection in the decimal pixel unit from the pixel value selected by the interpolation pixel selection step A reference image forming step for forming a reference image , and the detection accuracy is changed step by step from motion vector detection in units of integer pixels using pixel values at integer pixel positions stored in the storage step. the detection of a motion vector in units of sub-pixel a line cormorants method, when performing motion vector detection the motion vector detection step in units of sub-pixel, the auxiliary Pixel selection step, if the only pixel values of integer pixel positions stored can not be interpolated by the first interpolation pixel generation process in said storage step, the fraction generated by the first interpolation pixel generation process Instead of the pixel value at the pixel position , the computer is caused to perform switching so as to select the pixel value at the decimal pixel position generated in the second interpolation pixel generation step .

  The recording medium of the present invention records the computer program described above.

  ADVANTAGE OF THE INVENTION According to this invention, the moving image encoding apparatus which can achieve power saving can be provided, without significantly reducing encoding efficiency.

(First embodiment)
Embodiments of the present invention will be described below with reference to the drawings.
FIG. 2 is a block diagram illustrating a configuration example of a moving image encoding apparatus according to the present invention.
In FIG. 2, the moving image coding apparatus according to the present embodiment includes an imaging unit 101 including a camera unit such as a lens and a CCD, a motion vector detection circuit 102, a motion compensation circuit 108, a subtractor 103, and a DCT (orthogonal transform) circuit 104. Have

  Further, the quantization circuit 105, the variable length coding circuit 106, the recording unit 107, the recording medium 109, the code amount control circuit 110, the inverse quantization circuit 111, the IDCT (inverse orthogonal transform) circuit 112, the adder 113, and the display unit 117. A frame memory 118 is provided.

  The operation of each component including the code amount control circuit 110 is controlled by a system controller (not shown). In other words, this system controller controls the operation of the entire apparatus, and controls the operation of the entire apparatus in accordance with an instruction from a user as required from an operation unit (not shown).

  Image signals obtained by imaging by the imaging unit 101 are sequentially stored in the frame memory 118 in the order of the first frame, the second frame, the third frame,. From the frame memory 118, for example, image data is extracted in the order of encoding, such as the third frame, the first frame, the second frame,.

  The encoding performed in the present embodiment includes “intra encoding” in which only image data in a frame is encoded, and “inter encoding” in which encoding is performed including inter-frame prediction. In inter-coding, a P picture that performs prediction with one reference frame for a motion compensation unit (MC block) and a B picture that performs prediction with up to two reference frames for an MC block. is there. A picture to be subjected to intra coding is called an I picture. The reason why the order of frames to be encoded differs from the order of input frames is to enable temporal prediction (backward prediction) with future frames.

An outline of processing performed when intra coding is performed will be described with reference to a flowchart of FIG.
Image data of a block serving as a coding unit is read from the frame memory 118 (step S501), and orthogonal transformation is performed by the orthogonal transformation circuit 104 (step S502). The transform coefficient that is the output of the orthogonal transform circuit 104 is quantized in the quantization circuit 105 (step S503). The quantized transform coefficient that is the output of the quantization circuit 105 is subjected to variable length coding in the variable length coding circuit 106 (step S504), and then a recording signal to the recording medium is generated by the recording unit 107 (step S504). In step S505, recording is performed on the recording medium 109 (step S506).

  In each process described above, the quantization coefficient in the quantization circuit 105 is calculated by the code amount control circuit 110 from the feedback of the code amount generated by the variable length coding circuit 106. Further, the quantized transform coefficient that is the output of the quantization circuit 105 is inversely quantized by the inverse quantization circuit 111, and is subjected to inverse orthogonal transformation processing in the inverse orthogonal transformation circuit 112 to become a decoded image signal. The image signal is stored in the frame memory 118.

  On the other hand, when inter coding is performed, image data of a block serving as a coding unit is read from the frame memory 118 and input to the motion vector detection circuit 102. At the same time, the motion vector detection circuit 102 reads a reference image from the frame memory 118 and detects a motion vector from the encoded image and the reference image.

  The motion compensation circuit 108 performs motion compensation according to the motion vector to generate a predicted image. The difference between the encoded image and the predicted image is calculated by the subtracter 103 to generate a difference image. The generated difference image is output to the orthogonal transform circuit 104, and other processing is the same as in the case of the above-described intra coding.

Next, the operation of the motion vector detection circuit 102 will be described in detail.
FIG. 1 is a block diagram illustrating a configuration example of a motion vector detection circuit 102 included in the video encoding device according to the first embodiment.

  As shown in FIG. 1, the motion vector detection circuit 102 performs vector detection with integer pixel accuracy, an encoding block buffer 201 that stores image data of an encoding block, a detection range buffer 202 that stores image data of a reference image. A single pixel accuracy vector detection circuit 203 is provided.

  Further, a half-pixel accuracy vector detection circuit 208 that performs vector detection with half-pixel accuracy, and a first half-pixel accuracy image generation circuit 205 that generates a half-pixel accuracy image used when performing vector detection with half-pixel accuracy. Have

Further includes a second half-pixel precision image generation circuit 206 for generating a half-pixel accuracy image in a different manner from the first half-pixel precision image generation circuit 205. Here, for example, the first half-pixel accuracy image generation circuit 205 is a half-pixel accuracy image generation circuit using a 6-tap filter, and the second half-pixel accuracy image generation circuit 206 is a half-pixel accuracy image generation circuit using a 2-tap filter. is there. In addition to these, the first switch 204 and the second switch 207 are included.

For example, an example of a processing procedure when detecting a motion vector of an encoded block of 16 horizontal pixels and 16 vertical pixels will be described with reference to the flowchart of FIG.
First, an encoding block of 16 horizontal pixels and 16 vertical pixels is read from the frame memory 118 and stored in the encoding block buffer 201 (step S601). On the other hand, if the motion vector detection range is ± 4 pixels in both horizontal and vertical directions, a reference image of 24 horizontal pixels and 24 vertical pixels is read from the frame memory 118 and stored in the detection range buffer 202 (step). S602).

  The single pixel precision vector detection circuit 203 performs block matching on the block of the reference image stored in the detection range buffer 202 by comparing the block having the strongest correlation with the coding block stored in the coding block buffer 201. A single pixel precision motion vector is detected from the coordinates of the block having the strongest correlation (step S603).

  Next, a reference image with half-pixel accuracy is generated (step S604). As will be described later, this half-pixel precision reference image is generated by interpolation processing from single-pixel precision image data. Then, this half-pixel precision reference image is output to the half-pixel precision vector detection circuit 208.

  In addition, the single-pixel precision motion vector generated in step S604 is also output to the half-pixel precision vector detection circuit 208. In this embodiment, the half-pixel accuracy vector detection circuit 208 performs vector detection with half-pixel accuracy only in the peripheral range of the motion vector detected with single-pixel accuracy (step S605). In step S606, motion compensation is performed according to the motion vector detected in step S605.

In this embodiment, a characteristic motion vector detection operation will be described with reference to FIG.
FIG. 3 is an image diagram for explaining the motion vector detection operation. Reference numeral 202 in FIG. 3 denotes a detection range buffer, which stores reference image data in a range larger by 8 pixels both horizontally and vertically than the encoded block. The size of the encoded block is indicated by a dotted line in the figure.

  The dotted line indicates the position of the motion vector (0, 0). The range of single pixel precision vector detection is represented by black circles in the figure. Note that the black circle represents the pixel position at the upper left of the block. Calculates the sum of absolute differences of each pixel between the reference image block with the black circle at the top left of the block and the encoded block, and detects a single-pixel motion vector from the block position where the absolute difference sum is minimum. To do. For example, an arrow 301 shown in FIG. 3 is detected as a single pixel precision motion vector.

Next, vector detection with half-pixel accuracy is performed within a range of ± 0.5 pixels both horizontally and vertically with the above-described single-pixel accuracy motion vector as the center. In the figure, it is represented by a white circle. Block matching is similarly performed at the white circle points, and, for example, an arrow 302 is detected as a half-pixel motion vector.

  Incidentally, a half-pixel precision reference image is required for half-pixel precision motion vector detection. The half-pixel precision reference image is generated by interpolation processing from single-pixel precision image data.

  The first half-pixel accuracy image generation circuit 205 is a circuit that interpolates and generates half-pixel accuracy image data from single-pixel accuracy image data using a 6-tap filter, and the second half-pixel accuracy image generation circuit 206 has 2 taps. This circuit interpolates and generates half-pixel precision image data from single-pixel precision image data using a filter. The half-pixel precision vector detection circuit 208 uses the half-pixel precision reference image generated by the first half-pixel precision image generation circuit 205 or the second half-pixel precision image generation circuit 206 to move the half-pixel precision motion. Vector detection processing is performed. For example, H.M. In H.264, half-pixel accuracy image data is generated by a 6-tap filter.

  Therefore, three pixels of single pixel accuracy are required before and after the pixel to be interpolated. For example, when the single-pixel precision motion vector is (−3, −2) as shown in FIG. 4, the pixels stored in the detection range buffer 202 are used to generate the pixels indicated by white circles in the figure. In addition to the above, pixels with single pixel accuracy indicated by triangles are required. However, when half-pixel accuracy image data is interpolated with a 2-tap filter, interpolation is possible with only the pixels stored in the detection range buffer 202 even in the case of FIG.

  Therefore, when interpolation processing using a 6-tap filter is performed according to the result of a motion vector with single pixel accuracy, if the interpolation processing cannot be performed using only the pixels stored in the detection range buffer 202, the first switch 204 is used. , And the second switch 207 are switched so as to be connected to the second half-pixel accuracy image generation circuit 206, and if not, the second switch 207 is switched so as to be connected to the first half-pixel accuracy image generation circuit 205. To.

  In this embodiment, a half-pixel precision motion vector is detected, but the same processing can be performed even when a higher-precision motion vector detection such as quarter-pixel precision is performed thereafter.

(Second Embodiment)
The moving picture coding apparatus according to the second embodiment of the present invention is as shown in FIG. 2, except that the operation of the motion vector detection circuit 102 is different, the moving picture coding apparatus according to the first embodiment described above. The same operation is performed.

  FIG. 7 is a block diagram showing a configuration of the motion vector detection circuit 102 included in the moving picture coding apparatus according to the second embodiment of the present invention. A data bus usage rate detection circuit 701 is added to FIG. 1 used in the description of the first embodiment.

  The data bus usage rate detection circuit 701 monitors the state of the data bus in the frame memory 118 and calculates the data bus usage rate. The data bus utilization rate is a control signal for the first switch 204 and the second switch 207, and as described in the first embodiment, according to the result of the single pixel precision motion vector, A reference image with half-pixel accuracy is generated.

  For that purpose, in addition to the image data stored in the detection range buffer 202, when it is necessary to read out the image data from the frame memory 118, and when the usage rate of the data bus is larger than a certain threshold value. The switch control is performed as follows. In this case, the first switch 204 and the second switch 207 are switched so as to be connected to the second half-pixel accuracy image generation circuit 206. Otherwise, the first half-pixel accuracy image generation circuit 205 is switched. Switch to connect to.

  When it is necessary to read image data from the frame memory 118 for the interpolation process, the first half-pixel accuracy image generation circuit 205 of the 6-tap filter reads the image data from the frame memory 118 and executes the interpolation process. .

(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 on which the program is recorded are included in the present invention.

  In addition, the present invention can be implemented as, for example, a system, apparatus, method, program, storage medium, or the like. 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.

  In the present invention, a software program (in the embodiment, a program corresponding to the flowcharts shown in FIGS. 5 and 6) for realizing the functions of the above-described embodiments may be directly applied to the system or apparatus. Alternatively, it may be achieved by supplying from a remote location and the computer of the system or apparatus reads and executes 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 floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, MO, CD-ROM, CD-R, and CD-RW. Moreover, a magnetic tape, a non-volatile memory card, ROM, DVD (DVD-ROM, DVD-R), etc. are also included.

  As another program supply method, the program can be supplied by connecting to a homepage on the Internet using a browser of a client computer and downloading the computer program of the present invention from the homepage. Alternatively, it can be supplied by downloading a compressed file including an automatic installation function 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, the present invention includes 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.

  In addition, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and key information for decryption is downloaded from a homepage via the Internet to users who have cleared predetermined conditions. Let 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. In addition, 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 also be realized by the processing.

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

It is a block diagram which shows the structure of the motion vector detection circuit with which the moving image encoder which concerns on the 1st Embodiment of this invention is provided. It is a block diagram which shows the structure of the moving image encoder which concerns on this invention. It is a schematic diagram for demonstrating the operation | movement which detects a motion vector in steps. It is a schematic diagram for demonstrating a pixel required when interpolating with a 6 tap filter. It is a flowchart explaining the outline of an encoding process. It is a flowchart explaining an example of the process sequence in the case of detecting a motion vector. It is a block diagram which shows the structure of the motion vector detection circuit with which the moving image encoder which concerns on the 2nd Embodiment of this invention is provided.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Image pick-up part 102 Motion vector detection circuit 103 Subtractor 104 DCT (orthogonal transformation) circuit 105 Quantization circuit 106 Variable length coding circuit (VLC)
107 recording unit 108 motion compensation circuit 109 recording medium 110 code amount control circuit 111 inverse quantization circuit (IQ)
112 IDCT (Inverse Orthogonal Transform) Circuit 113 Adder 117 Display Unit 117
118 Frame Memory 201 Encoding Block Buffer 202 Detection Range Buffer 203 Single Pixel Accuracy Vector Detection Circuit 204 First Switch 205 First Half Pixel Accuracy Image Generation Circuit (6-tap Filter)
206 Second half-pixel precision image generation circuit (2-tap filter)
207 Second switch 208 Half-pixel accuracy vector detection circuit

Claims (7)

Video with the motion vector detection means for detecting a motion vector between the coding target picture and see the image at sub-pixel basis, and motion compensation means for performing motion compensation according to the motion vector detected by said motion vector detecting means An image encoding device comprising:
The motion vector detecting means includes
Storage means for reading out and storing pixel values at integer pixel positions of the reference image stored in the frame memory;
First interpolated pixel generation means for interpolating and generating pixel values at decimal pixel positions by a filter with a predetermined number of taps using pixel values at integer pixel positions stored in the storage means ;
Second interpolation pixel generation means for interpolating and generating pixel values at decimal pixel positions using a filter having a smaller number of taps than the first interpolation pixel generation means, using pixel values at integer pixel positions stored in the storage means When,
Interpolation pixel selection means for selecting a pixel value at a decimal pixel position generated by any one of the first and second interpolation pixel generation means as a pixel value for forming a reference image with decimal pixel precision ;
Reference image forming means for forming a reference image with decimal pixel precision used for motion vector detection in the decimal pixel unit from the pixel value selected by the interpolation pixel selection means ;
From the motion vector detection integer pixel using the pixel value of the stored integer pixel positions in units in the storage means, the detection of a motion vector in units of sub-pixel is changed stepwise detection accuracy in line Umono There,
When the motion vector detection means performs motion vector detection in units of decimal pixels,
The interpolation pixel selection means is generated by the first interpolation pixel generation means when interpolation processing by the first interpolation pixel generation means cannot be performed only with the pixel value at the integer pixel position stored in the storage means. A moving picture coding apparatus that switches to select a pixel value at a decimal pixel position generated by the second interpolation pixel generation means instead of the pixel value at a decimal pixel position . The motion vector detection means performs a coarse search of motion vectors in integer pixel units, and then performs a dense search of motion vectors in decimal pixel units around the motion vectors detected by the coarse search, and detects by the coarse search. An interpolation pixel for forming a reference image with decimal pixel precision used in the dense search is selected from the first and second interpolation pixel generation means by the interpolation pixel selection means in accordance with the motion vector thus determined . The moving picture encoding apparatus according to claim 1, wherein: The motion vector detection means performs a coarse search of motion vectors in units of integer pixels, and then performs a dense search of motion vectors in units of decimal pixels around the motion vectors detected by the coarse search, and the data in the frame memory According to the bus usage rate, an interpolation pixel for forming a reference image with decimal pixel precision used in the dense search is selected from the first and second interpolation pixel generation means by the interpolation pixel selection means. The moving picture encoding apparatus according to claim 1, wherein: Reference image data of integer pixel positions read from the frame memory for the coarse search is stored in a storage medium,
The moving picture encoding apparatus according to claim 2, wherein the reference image in decimal pixel units used in the dense search is generated using only image data stored in the storage medium. A motion vector detection step of detecting a motion vector between the sign-encoding target image and see the image at sub-pixel basis, and a motion compensation step of performing motion compensation according to the motion vector detected by said motion vector detection step A video encoding method comprising:
The motion vector detection step includes
A storage step of reading out and storing the pixel value of the integer pixel position of the reference image stored in the frame memory;
A first interpolation pixel generation step of interpolating and generating a pixel value at a decimal pixel position by a filter having a predetermined number of taps using a pixel value at an integer pixel position stored in the storage step ;
A second interpolation pixel generation step of interpolating and generating a pixel value at a decimal pixel position using a filter having a smaller number of taps than in the first interpolation pixel generation step, using the pixel value at the integer pixel position stored in the storage step When,
An interpolation pixel selection step of selecting a pixel value at a decimal pixel position generated by one of the first and second interpolation pixel generation steps as a pixel value for forming a reference image with decimal pixel accuracy ;
A reference image forming step of forming a reference image of decimal pixel accuracy used in motion vector detection in the decimal pixel unit from the pixel value selected by the interpolation pixel selection step ;
From the motion vector detection integer pixel using the pixel value of the stored integer pixel positions in units in said storage step, the detection of a motion vector in units of sub-pixel is changed stepwise detection accuracy in line Cormorants method There,
When the motion vector detection step performs motion vector detection in units of decimal pixels,
The interpolation pixel selection step is generated by the first interpolation pixel generation step when the interpolation processing by the first interpolation pixel generation step cannot be performed only with the pixel value at the integer pixel position stored in the storage step. A moving picture encoding method, wherein switching is performed so that the pixel value at the decimal pixel position generated in the second interpolation pixel generation step is selected instead of the pixel value at the decimal pixel position . Video with the motion vector detection step of detecting a motion vector between the coding target picture and see the image at sub-pixel basis, and a motion compensation step of performing motion compensation according to the motion vector detected by said motion vector detection step A computer program for causing a computer to execute each step of the image encoding method,
The motion vector detection step includes
A storage step of reading out and storing the pixel value of the integer pixel position of the reference image stored in the frame memory;
A first interpolation pixel generation step of interpolating and generating a pixel value at a decimal pixel position by a filter having a predetermined number of taps using a pixel value at an integer pixel position stored in the storage step ;
A second interpolation pixel generation step of interpolating and generating a pixel value at a decimal pixel position using a filter having a smaller number of taps than in the first interpolation pixel generation step, using the pixel value at the integer pixel position stored in the storage step When,
An interpolation pixel selection step of selecting a pixel value at a decimal pixel position generated by one of the first and second interpolation pixel generation steps as a pixel value for forming a reference image with decimal pixel accuracy ;
A reference image forming step of forming a reference image of decimal pixel accuracy used in motion vector detection in the decimal pixel unit from the pixel value selected by the interpolation pixel selection step ;
From the motion vector detection integer pixel using the pixel value of the stored integer pixel positions in units in said storage step, the detection of a motion vector in units of sub-pixel is changed stepwise detection accuracy in line Cormorants method There,
When the motion vector detection step performs motion vector detection in units of decimal pixels,
The interpolation pixel selection step is generated by the first interpolation pixel generation step when the interpolation processing by the first interpolation pixel generation step cannot be performed only with the pixel value at the integer pixel position stored in the storage step. A computer program for causing a computer to perform switching so as to select a pixel value at a decimal pixel position generated in the second interpolation pixel generation step instead of a pixel value at a decimal pixel position . A computer-readable recording medium on which the computer program according to claim 6 is recorded .

Images