JP2009141775A - Method and device for coding - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To restrain discrepancy in a search range of vector detection also in a boundary of divided blocks when motion detection for extracting candidate vector by each relatively large block wherein one screen is divided into plurality. <P>SOLUTION: When motion detection for extracting candidate vector by each relatively large block wherein one screen is divided into plurality, a macro block search range located in an area near a boundary of a divided first region is set again as a second region in motion vector search for each of the macro block and is corrected, thus preventing troubles to generate large difference in accuracy of motion vector detection in an area near divided block boundary. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an encoding apparatus and an encoding method, and more particularly to a technique suitable for use in encoding performed when image data such as a still image or a moving image is recorded on a medium.

  In recent years, video cameras using disk media as storage media have begun to be commercialized. Video cameras using disk media are small and easy to carry, and are expected to become increasingly popular in the future. In such a device, at the time of recording, video is stored in a recording medium with a video stream that has been compression-encoded such as MPEG.

  For example, frame or inter-field coding using motion compensation represented by MPEG (herein referred to as “inter-frame coding”) will be described. In interframe coding, motion vector detection (search) is performed in units of macroblocks or 8 × 8 pixel blocks (here, “macroblock units”), and the frame to be encoded is selected. Centered on the front and rear frames.

  At this time, the motion vector search is uniformly performed for all the macroblocks existing in the target frame regardless of the amount of motion between frames regarding the image data in the actual macroblock. As a method for obtaining a motion vector, a block matching method is known.

  The block matching method refers to a method of comparing two images in units of blocks such as a reference frame / field and an input frame / field and searching for motion in units of blocks based on the degree of matching of the video signals. By using the block matching method, a motion vector is searched for each block from the reference frame / field and the input frame / field, and motion compensated prediction is performed using the searched motion vector. This block matching method is described in Patent Document 1, for example.

FIG. 7 is a configuration diagram illustrating an example of a conventional encoding device that performs motion vector detection. Hereinafter, the encoding process of the conventional encoding apparatus will be described with reference to FIG.
The conventional encoding device shown in FIG. 7 includes a reference frame holding unit 701, a global vector search unit 702, a search range determination unit 703, a motion vector search unit 704, and an encoding unit 705.

  The reference frame holding unit 701 stores the input image data (frame) as a reference frame, and divides the frame into a plurality of first regions. And it is used for the search of the global vector (global vector) which is a motion vector for every 1st area | region, and the motion vector for every macroblock.

  The global vector search unit 702 obtains a global vector between the input frame and the reference frame held in the reference frame holding unit 701 for each of the first areas. The method for obtaining the global vector is not particularly limited, but a block matching method is used here for explanation.

FIG. 8 is a conceptual diagram showing the process of global vector search by the block matching method. FIG. 9 shows a vector search range of macroblocks included in the first area.
In the block matching method, as shown in FIG. 8, two images such as a reference frame and an input frame are compared in units of blocks, and a motion is searched in units of blocks based on the degree of matching of video signals.

  The search range determination unit 703 determines the search range in the motion vector search for each macroblock using the global vector for each divided region obtained by the global vector search unit 702 as shown in FIG.

  The motion vector search unit 704 searches for a block having the highest degree of matching with the video signal of the current macro block within the search range from the reference frame, and searches for a final motion vector for the current macro block. The encoding unit 705 performs encoding using the motion vector of each macroblock obtained by the motion vector search unit 704, and outputs encoded image data.

JP 2001-61152 A

  In the prior art configuration, candidate vectors are extracted for each relatively large block obtained by dividing one screen into a plurality of blocks, so that when a person or the like is captured in large size, the subject straddles the boundary between the divided blocks. There is.

  In such a case, since the motion vector search area is different for each divided block, a large difference occurs in the accuracy of motion vector detection near the divided block boundary. As a result, the noise seems to be riding, and the user may feel uncomfortable.

  In the present invention, in view of the above-described problems, when motion detection is performed for extracting candidate vectors for each relatively large block obtained by dividing one screen into a plurality of blocks, a difference in search range of vector detection is also detected at the boundary of the divided blocks. The purpose is to be able to suppress it.

  The encoding apparatus according to the present invention includes a reference frame holding unit that divides input image data into a plurality of first areas and holds a reference frame used for searching for a motion vector for each divided first area. , A global vector search means for finding a global vector between the input image data and a reference frame held in the reference frame holding means for each of the first areas, and a divided area obtained by the global vector search means Search range determining means for determining a search range in motion vector search for each macroblock using a global vector for each macroblock, and a video signal of the current macroblock within the search range determined by the search range determining means Is searched for from the reference frame held in the reference frame holding means, and the current macroblock is found. A motion vector search means for searching for the final motion vector of the image, and encoding that performs encoding using the motion vector of each macroblock obtained by the motion vector search means to generate and output encoded image data And a search range correction unit that resets and corrects a macroblock search range located near the boundary of the divided first region as a second region in the motion vector search for each macroblock. It is characterized by that.

  The encoding method of the present invention divides input image data into a plurality of first regions, and holds a reference frame used for searching for a motion vector for each divided first region; , A global vector search step for obtaining a global vector between the input image data and the reference frame held in the reference frame holding step for each of the first regions, and a divided region obtained in the global vector search step A search range determining step for determining a search range in a motion vector search for each macroblock using a global vector for each macroblock, and a matching degree with a video signal of a current macroblock within the search range determined in the search range determining step Is searched from the reference frame held in the reference frame holding step, and the current macroblock is determined. A motion vector search step for searching for a final motion vector, and a code for performing encoding using the motion vector of each macroblock obtained in the motion vector search step to generate and output encoded image data And a search range correction step of resetting and correcting a macroblock search range located near the boundary of the divided first region as a second region in the motion vector search for each macroblock. It is characterized by having.

  The program of the present invention divides input image data into a plurality of first areas, and holds a reference frame used for searching for a motion vector for each divided first area; A global vector search step for obtaining a global vector between the input image data and the reference frame held in the reference frame holding step for each of the first regions, and for each divided region obtained in the global vector search step. A search range determination step for determining a search range in a motion vector search for each macroblock using a global vector, and a video signal of the current macroblock having the highest matching degree within the search range determined in the search range determination step. A high block is searched from the reference frame held in the reference frame holding step, and the current macro block is searched. A motion vector search step for searching for a final motion vector, and a code for performing encoding using the motion vector of each macroblock obtained in the motion vector search step to generate and output encoded image data And a search range correction step of resetting and correcting a macroblock search range located near the boundary of the divided first region as a second region in the motion vector search for each macroblock. A program characterized by being executed by a computer.

  According to the present invention, it is possible to suppress the difference in the search range of vector detection even at the boundary of the divided blocks. In addition, the vector detection accuracy can be kept uniform even near the boundary. As a result, it is possible to configure an encoding device that can reduce the noise caused by different motion vector search areas.

(First embodiment)
FIG. 1 is a block diagram illustrating a configuration example of the encoding apparatus according to the present embodiment. The motion vector search range correction method according to this embodiment will be described below with reference to FIG.
In the encoding apparatus of this embodiment, the reference frame holding unit 101, the search range determination unit 103, the motion vector search unit 104, and the encoding unit 105 are the same as those described with reference to FIG.

  That is, the reference frame holding unit 101 divides input image data into a plurality of first areas, and holds a reference frame used for searching for a motion vector for each divided first area.

  The global vector search unit 102 divides the input storage frame (input image data) into a plurality of first regions as in the conventional case. Then, a global vector with the reference frame held in the reference frame holding unit 101 is obtained for each of the divided first areas. In the present embodiment, in order to simplify the description, a configuration in which the global vector search unit 102 detects a motion vector in units of macroblocks when one screen is divided into 3 × 3 will be described. The search range correction unit 106 sets the search range again for the macroblock located near the boundary of the first region. Details of the setting will be described later.

The search range determination unit 103 determines a search range in the motion vector search for each macroblock using the global vector for each divided region obtained by the global vector search unit 102.
The motion vector search unit 104 searches the reference frame held in the reference frame holding unit 101 for a block having the highest matching degree with the video signal of the current macro block within the search range determined by the search range determining unit 103. To do. Then, the final motion vector for the current macroblock is searched.
The encoding unit 105 performs encoding using the motion vector of each macroblock obtained by the motion vector search unit 104, and generates and outputs encoded image data.

Next, the processing of the search range correction unit 106 in the encoding device of the present embodiment will be described using FIG.
The search range correction unit 106 resets and corrects the macroblock search range positioned near the boundary of the divided first region as the second region in the motion vector search for each macroblock. In the present embodiment, ± n macroblocks in the x direction are set as the application range for the search range correction in the x direction for the boundary in the y direction, and ± m macroblocks are set in the y direction for the boundary in the x direction.

  3A and 3B are conceptual diagrams showing the vector search range of each macroblock after the search range correction for each macroblock to which the search range correction is applied in the search range correction unit 106. FIG. is there.

The correction process of the search range correction unit 106 will be described with reference to FIG.
The search range correction unit 106 sets, as the search range, the average value of the global vectors of the first regions adjacent to each other with respect to each macro block to which the search range correction is applied. In FIG. 3A, the motion vector search range of the first region is “GV (0,0)... GV (2,2)”, and each macroblock to which the search range correction is applied. The motion vector search range is “GV (a)... GV (p)”.

  For example, the corrected search range GV (a) is obtained by “(GV (0,0) + GV (1,0)) / 2”, and GV (c) is “(GV (0,0) + GV (0,1)) / 2 ”. For the vertical and horizontal boundaries, search range correction application areas such as `` GV (d) '' and `` GV (f) '' set the average value of the search range of adjacent application areas. To do. For example, “GV (d) is obtained by (GV (c) + GV (e)) / 2”. As a result, the search range of each macroblock after the search range correction is as shown in FIG.

  As described above, by correcting the search range, the calculation of the global vector is detected for the first region, and the search range near the division boundary of the first region is corrected. As a result, it is possible to suppress a sudden change in the motion vector at the boundary of the divided areas as shown in FIG. . Thereby, a disc video camera suitable for the user can be provided.

(Second Embodiment)
Next, as a second embodiment, a description will be given of a configuration in which the search range correction application area is determined using the difference value of the global vectors of the adjacent first areas in the processing of the search range correction unit 106. The configuration of this embodiment is the same as that of the first embodiment.

  FIG. 4A and FIG. 4B are flowcharts illustrating an example of a processing procedure of the search range correction unit 106 in the present embodiment. Hereinafter, the determination process of the application area of the search range correction in this embodiment will be described with reference to FIG.

First, an example of a procedure for obtaining a search range correction application region for the boundary in the y direction will be described with reference to the flowchart of FIG.
In step S411, a global vector difference T (y) in the y direction is obtained for each first region adjacent to the division boundary.

  Next, it progresses to step S412 and it is judged whether the difference T (y) calculated | required by step S411 is more than threshold value (DELTA) GV (y). As a result of the determination, if the difference T (y) is equal to or greater than the threshold value ΔGV (y), the process proceeds to step S413, and if not, the process proceeds to step 414 to obtain the search range correction application region for the boundary in the y direction.

  In step S413, the application range of the search range correction is set to “± (m + α) macroblock”. In step S414, the application range of the search range correction is set to “± m macroblock”.

  The relationship between the search range correction application area obtained in step S413 and the search range correction application area obtained in step S414 is proportional to the magnitude of the motion vector difference for each area. That is, when the magnitude T (y) of the difference between the global vectors in the y direction is equal to or greater than the threshold value ΔGV (y), a wider area becomes the application area.

  Next, an example of a procedure for obtaining a search range correction application region with respect to the boundary in the x direction will be described with reference to the flowchart of FIG.

  In step S421, a difference T (x) between global vectors in the x direction between adjacent first regions is obtained. Next, in step S422, it is determined whether or not the difference T (x) obtained in step S421 is greater than or equal to a threshold value ΔGV (x).

  As a result of the determination in step S422, if the difference T (x) is greater than or equal to the threshold value ΔGV (x), the process proceeds to step S423; otherwise, the process proceeds to step S424 to apply the search range correction application area to the boundary in the x direction. Ask for.

  In step S423, the application range of the search range correction is “± (n + α) macroblock”, and in step S424, the application region of the search range correction is “± n macroblock”. The relationship between the application range of the search range correction obtained in step S423 and the application region of the search range correction obtained in step S424 is that the magnitude T (x) of the global vector difference in the x direction is greater than or equal to the threshold ΔGV (x). In this case, a wider area becomes the application area.

FIG. 5 is a conceptual diagram of an interpolation area in the present embodiment.
When the search range correction process as in the present embodiment is performed, as shown in FIG. 5, the difference between the global vectors for each adjacent first region is as follows. That is, when the difference of the global vector is greater than or equal to a certain threshold value, a region for correcting the motion vector search range in the second region located near the boundary is compared with a predetermined region as compared with the case where the difference is less than the threshold value. Choose as much as the amount.

FIG. 6A and FIG. 6B are conceptual diagrams showing a search range with respect to an application range of search range correction in the present embodiment.
Hereinafter, the correction processing result of the search range correction unit 106 will be described with reference to FIG. Since the correction processing of the search range correction unit 106 performed in the present embodiment has the same configuration as that of the first embodiment, detailed description thereof is omitted. As a result of the correction process, the search range of each macroblock after the search range correction is as shown in FIG.

  As shown in FIG. 6B, when the difference between the global vectors of adjacent divided areas is larger than a certain threshold, the application range of the search range correction is widened, and otherwise, the search region correction application region is narrowed. By correcting the search range in this way, even when the difference between the global vectors of the adjacent first regions is large, a sudden change in the vector search range can be suppressed.

(Another embodiment according to the present invention)
Each unit constituting the encoding apparatus according to 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.

  In addition, the present invention can be implemented as a system, apparatus, method, program, storage medium, or the like, and can be applied to a system composed of a plurality of devices. Moreover, you may apply to the apparatus which consists of one apparatus.

  In the present invention, a software program (in the embodiment, a program corresponding to the flowchart shown in FIG. 4) for executing each step in the encoding method described above is directly or remotely supplied to a system or apparatus. In addition, this includes a 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.

  Various recording media can be used as a recording medium for supplying the program. For example, floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card, ROM, DVD (DVD-ROM, DVD- R).

  As another program supply method, a browser on a client computer is used to connect to an Internet home page. 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.

  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.

  In addition to the functions of the above-described embodiments being realized by the computer executing the read program, the OS running on the computer performs part or all of the actual processing. Also, the functions of the above-described embodiments can be realized.

  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 structural example of the encoding apparatus in the 1st Embodiment of this invention. It is a conceptual diagram of the application area | region of the search range correction | amendment in the 1st Embodiment of this invention. It is a conceptual diagram of the vector search range in the 1st Embodiment of this invention. It is a flowchart which shows an example of the process sequence of the search range correction | amendment part in the 2nd Embodiment of this invention. It is a conceptual diagram of the application area | region of the search range correction | amendment in the 2nd Embodiment of this invention. It is a conceptual diagram of the vector search range in the 2nd Embodiment of this invention. It is a block diagram of the conventional encoding apparatus. It is a conceptual diagram which shows the process of the global vector search by a block matching system. It is a conceptual diagram of the vector search range in a conventional example.

Explanation of symbols

101 Reference frame holding unit 102 Global vector search unit 103 Search range determination unit 104 Motion vector search unit 105 Encoding unit 106 Search range correction unit

Claims (10)

Reference frame holding means for dividing the input image data into a plurality of first regions and holding a reference frame used for searching for a motion vector for each of the divided first regions;
A global vector search means for obtaining a global vector between the input image data and a reference frame held in the reference frame holding means for each of the first areas;
Search range determination means for determining a search range in motion vector search for each macroblock using the global vector for each divided area obtained by the global vector search means;
A block having the highest matching degree with the video signal of the current macroblock within the search range determined by the search range determination unit is searched from the reference frame held in the reference frame holding unit, and the current macroblock Motion vector search means for searching for the final motion vector of
Encoding means for performing encoding using the motion vector of each macroblock obtained by the motion vector search means, and generating and outputting encoded image data;
Search range correction means for resetting and correcting a macroblock search range located near the boundary of the divided first region as a second region in the motion vector search for each macroblock. An encoding device.   The search range correction unit is configured to determine a horizontal motion vector search range of a second region located near a boundary of the divided first region, the first region including the second region, and the first region The encoding apparatus according to claim 1, wherein correction is performed using a motion vector of a first region adjacent to one region across a vertical boundary.   The search range correction means is configured to determine a vertical motion vector search range of a second region located near a boundary of the divided first region, the first region including the second region, and the first region The encoding apparatus according to claim 1, wherein correction is performed using a motion vector of a first region adjacent to one region across a horizontal boundary.   The search range correction unit is configured to determine a motion vector search range of a second region located near a boundary between the divided first regions, the first region including the second region, and the first region. The encoding apparatus according to claim 1, wherein correction is performed using an average value of motion vectors of first regions adjacent to each other with a boundary interposed therebetween.   The search range correction unit is configured to search for a motion vector in a second region located in the vicinity of the boundary based on a difference between motion vectors of the regions adjacent to each other across the boundary of the divided first region. The encoding apparatus according to claim 1, wherein an area whose range is to be corrected is selected.   The search range correction means includes a second area located in the vicinity of the boundary in proportion to the magnitude of the difference between the motion vectors of the adjacent areas across the boundary of the divided first area. 6. The encoding apparatus according to claim 5, wherein a wide area for correcting the motion vector search range is selected.   The search range correction means, when the difference of the motion vector for each of the adjacent regions across the boundary of the divided first region is greater than or equal to a threshold, compared with the case where the difference is less than the threshold, 6. The encoding apparatus according to claim 5, wherein a region for correcting a motion vector search range is widely selected by a predetermined amount from the second region located in the vicinity of the boundary. A reference frame holding step of dividing the input image data into a plurality of first areas and holding a reference frame used for searching for a motion vector for each of the divided first areas;
A global vector search step for obtaining a global vector between the input image data and the reference frame held in the reference frame holding step for each of the first regions;
A search range determination step for determining a search range in a motion vector search for each macroblock using a global vector for each divided region obtained in the global vector search step;
A block having the highest matching degree with the video signal of the current macro block within the search range determined in the search range determination step is searched from the reference frame held in the reference frame holding step, and the current macro block is determined. A motion vector search step for searching for a final motion vector of
An encoding step of performing encoding using the motion vector of each macroblock obtained in the motion vector search step, and generating and outputting encoded image data;
A search range correction step of resetting and correcting a macroblock search range located near the boundary of the divided first region as a second region in the motion vector search for each macroblock. The encoding method. A reference frame holding step of dividing the input image data into a plurality of first areas and holding a reference frame used for searching for a motion vector for each of the divided first areas;
A global vector search step for obtaining a global vector between the input image data and the reference frame held in the reference frame holding step for each of the first regions;
A search range determination step for determining a search range in a motion vector search for each macroblock using a global vector for each divided region obtained in the global vector search step;
A block having the highest matching degree with the video signal of the current macro block within the search range determined in the search range determination step is searched from the reference frame held in the reference frame holding step, and the current macro block is determined. A motion vector search step for searching for a final motion vector of
An encoding step of performing encoding using the motion vector of each macroblock obtained in the motion vector search step, and generating and outputting encoded image data;
Causing the computer to execute a search range correction step of resetting and correcting a macroblock search range located near the boundary of the divided first region as a second region in the motion vector search for each macroblock. A program characterized by that.   A computer-readable storage medium storing the program according to claim 9.

Images