JPWO2018142596A1 - Encoding apparatus, encoding method, and encoding program - Google Patents

Abstract

The motion compensation prediction unit performs a motion compensation prediction process on the image frame and outputs a prediction frame, and also outputs prediction information (24) used for the motion compensation prediction process. The transform encoding unit encodes the difference between the image frame and the prediction frame and outputs it as encoded data (25). The variable length encoding unit (141) encodes the encoded data (25) using the variable length code, thereby compressing the encoded data (25) as the variable length code data (27). Then, the addition control unit (142) adds the prediction information (24) to the variable length code data (27).

Description

  The present invention relates to an encoding device, an encoding method, and an encoding program.

One of basic information for image processing is optical flow. Optical flow is defined as apparent movement on the image. As a method for detecting an optical flow, there is a method such as a gradient method. In general, video data to be subjected to image processing is handled by being compressed from the viewpoint of reducing the amount of data. In recent years, MPEG-2, MPEG-4, AVC / H. H.264, and HEVC / H. An international standard system of MPEG such as H.265 is widely used. The MPEG method performs motion compensation prediction in order to reduce the correlation in the time direction during the compression process. This motion compensation prediction process is similar to the optical flow detection method. The motion vector obtained as a result of the motion compensation prediction process may be used as a substitute for the optical flow. The motion vector is information that can be acquired in the compression process, and there is no need to newly generate information related to motion necessary for image processing. Therefore, the acquisition of the motion vector requires less arithmetic processing than the calculation of the optical flow, and enables image processing to be performed with low arithmetic processing.
Patent Document 1 discloses a method for specifying an object position as a more specific example of image processing in addition to substituting a motion vector for an optical flow.

JP 2007-214886 A

The motion compensated prediction process is aimed at maximizing the compression performance within a given condition, and may not be able to correctly follow the motion in the video. As a specific example, when the motion search range is smaller than the motion amount in the video, the motion compensation result may indicate no motion or may indicate the maximum value of the search range. In motion compensation prediction, there are a plurality of motion search methods such as full search, hierarchical search, and telescopic search. The characteristics of the obtained motion vector differ depending on which of the plurality of motion search methods is used. Therefore, in image processing, it is necessary to know the method of generating a motion vector in order to improve the accuracy of image processing.
However, in the related art, when video data is encoded by an encoding process using a motion compensation prediction process, prediction information related to a motion search process used in the motion compensation prediction process such as a motion vector generation method is unknown. There is a problem.

  An object of the present invention is to extract prediction information related to a motion search process used for a motion compensation prediction process from encoded data even when video data is encoded by an encoding process using a motion compensation prediction process. To do.

The encoding device according to the present invention is:
A motion compensation prediction unit that performs motion compensation prediction processing on an image frame and outputs a prediction frame, and outputs prediction information used in the motion compensation prediction processing;
A transform encoding unit that encodes a difference between the image frame and the prediction frame and outputs the encoded difference as encoded data;
A variable length encoding unit that compresses the encoded data as variable length code data by encoding the encoded data using a variable length code;
And an addition control unit that adds the prediction information to the variable-length code data.

  In the encoding device according to the present invention, the motion compensation prediction unit performs motion compensation prediction processing on an image frame and outputs a prediction frame, and also outputs prediction information used in the motion compensation prediction processing. The transform encoding unit encodes the difference between the image frame and the predicted frame and outputs the encoded data as encoded data. A variable length encoding unit compresses the encoded data as variable length code data by encoding the encoded data using a variable length code. Then, the addition control unit adds the prediction information to the variable length code data. Therefore, according to the encoding device of the present invention, even when video data is encoded by an encoding process using a motion compensation prediction process, prediction information used for the motion compensation prediction process is extracted from the encoded data. There is an effect that can be done.

1 is a configuration diagram of an encoding apparatus 100 according to Embodiment 1. FIG. FIG. 3 is a configuration diagram of an entropy encoding unit 140 according to Embodiment 1. FIG. 6 is a flowchart showing an encoding process S100 by the encoding method 610 and the encoding program 620 according to the first embodiment. 3 is a configuration diagram of compressed data 23 according to Embodiment 1. FIG. FIG. 6 is a flowchart showing a detailed operation of additional control processing S42 according to the first embodiment. FIG. 6 is a configuration diagram of an encoding apparatus 100 according to a modification of the first embodiment. The figure which shows the example of the search range information 242 contained in the prediction information 24 concerning Embodiment 2. FIG. Is a schematic diagram showing a method of full search. Is a schematic diagram showing a method of step search. Is a schematic diagram showing a diamond search method. Is a schematic diagram showing a method of searching for a hexagon. Is a flow diagram showing a method of full search. FIG. 3 is a flowchart showing a step search method. Is a flow diagram showing a diamond search method. Is a flowchart showing a method of searching for a hexagon.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, the same code | symbol is attached | subjected to the part which is the same or it corresponds in each figure. In the description of the embodiments, the description of the same or corresponding parts will be omitted or simplified as appropriate.

Embodiment 1 FIG.
*** Explanation of configuration ***
FIG. 1 is a configuration diagram of an encoding apparatus 100 according to the present embodiment. FIG. 2 is a configuration diagram of the entropy encoding unit 140 according to the present embodiment. The configuration of encoding apparatus 100 according to the present embodiment will be described using FIG. 1 and FIG.
As shown in FIG. 1, the encoding apparatus 100 is a computer. The encoding device 100 is also referred to as an encoder.
The encoding device 100 includes hardware such as a processor 910, a storage device 920, an image input interface 930, an output interface 940, and a communication device 950. The storage device 920 includes a memory 921 and an auxiliary storage device 922.

The encoding apparatus 100 includes a motion compensation prediction unit 110, a subtraction unit 120, a transform encoding unit 130, an entropy encoding unit 140, and a storage unit 150 as functional configurations. The entropy encoding unit 140 includes a variable length encoding unit 141 and an addition control unit 142. In addition, the addition control unit 142 includes a multiplexing unit 421, a switch unit 422, and a multiplexing control unit 423.
The functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 are realized by software.
The storage unit 150 is realized by the memory 921. In addition, the storage unit 150 may be realized only by the auxiliary storage device 922 or by the memory 921 and the auxiliary storage device 922. A method for realizing the storage unit 150 is arbitrary.

  The processor 910 is connected to other hardware via a signal line, and controls these other hardware. The processor 910 is an IC (Integrated Circuit) that performs arithmetic processing. Specific examples of the processor 910 are a CPU (Central Processing Unit), a DSP (Digital Signal Processor), and a GPU (Graphics Processing Unit).

  The memory 921 is a storage device that temporarily stores data. Specific examples of the memory 921 are SRAM (Static Random Access Memory) and DRAM (Dynamic Random Access Memory).

  The auxiliary storage device 922 is a storage device that stores data. A specific example of the auxiliary storage device 922 is an HDD (Hard Disk Drive). The auxiliary storage device 922 includes an SD (registered trademark) (Secure Digital) memory card, a CF (Compact Flash), a NAND flash, a flexible disk, an optical disk, a compact disk, a Blu-ray (registered trademark) disk, a DVD (Digital Versatile Disk), and the like. It may be a portable storage medium.

  The image input interface 930 is a port connected to an input device such as an imaging device that captures video. Specifically, the image input interface 930 is a USB (Universal Serial Bus) terminal. Note that the image input interface 930 may be a port connected to a LAN (Local Area Network). Each image frame 21 of the plurality of image frames constituting the video data is input to the encoding device 100 via the image input interface 930.

  The output interface 940 is a port to which a cable of a display device such as a display is connected. Specifically, the output interface 940 is a USB terminal or a HDMI (registered trademark) (High Definition Multimedia Interface) terminal. The display is specifically an LCD (Liquid Crystal Display). Note that the encoding apparatus 100 may not include the output interface 940.

The communication device 950 communicates with other devices via a network.
The communication device 950 includes a receiver and a transmitter. The communication device 950 is wired or wirelessly connected to a communication network such as a LAN, the Internet, or a telephone line. The communication device 950 is specifically a communication chip or a NIC (Network Interface Card). The communication device 950 is a communication unit that communicates data. The receiver is a receiving unit that receives data. The transmitter is a transmission unit that transmits data. The compressed data 23 generated by the encoding device 100 is sent to the network via the communication device 950.

  The auxiliary storage device 922 stores programs that realize the functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142. Yes. A program that realizes the functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 is also referred to as a coding program 620. This program is loaded into the memory 921, read into the processor 910, and executed by the processor 910. The auxiliary storage device 922 stores an OS. At least a part of the OS stored in the auxiliary storage device 922 is loaded into the memory 921. The processor 910 executes the encoding program 620 while executing the OS.

  The encoding apparatus 100 may include only one processor 910 or may include a plurality of processors 910. The plurality of processors 910 cooperate to execute programs that realize the functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142. May be.

  Information, data, signal values, and variable values indicating the results of the processes of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 are: The data is stored in the auxiliary storage device 922 of the encoding device 100, the memory 921, or a register or cache memory in the processor 910.

Programs that realize the functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 may be stored in a portable recording medium. Good. Specifically, the portable recording medium is a memory card such as a magnetic disk, a flexible disk, an optical disk, a compact disk, a Blu-ray (registered trademark) disk, a DVD (Digital Versatile Disc), or an SD card.
The encoding program product is a storage medium and a storage device in which the encoding program 620 is recorded. An information providing program product refers to a program loaded with a computer-readable program regardless of its appearance.

*** Explanation of operation ***
The encoding process S100 by the encoding method 610 and the encoding program 620 according to the present embodiment will be described using FIG.
The encoding process S100 includes a motion compensation prediction process S10, a subtraction process S20, a transform encoding process S30, and an entropy encoding process S40.

  First, in step S101, the image input interface 930 acquires one image frame 21 among a plurality of image frames constituting a video. The image input interface 930 outputs the image frame 21 to the motion compensation prediction unit 110 and the subtraction unit 120.

<Motion compensation prediction process S10>
Next, the motion compensation prediction unit 110 executes a motion compensation prediction process S10. The motion compensation prediction unit 110 performs a motion compensation prediction process on the image frame 21 to output a prediction frame 22 and outputs prediction information 24 used for the motion compensation prediction process. The motion compensation prediction unit 110 is also referred to as MC (Motion Compensation).
Specifically, in step S <b> 102, the motion compensation prediction unit 110 generates the prediction frame 22 using the image frame 21 and the image frame that has already been encoded by the transform encoding unit 130.
In step 103, the motion compensation prediction unit 110 outputs information used when the motion compensation prediction process is performed on the image frame 21 as the prediction information 24. The prediction information 24 is also referred to as a prediction parameter. The motion compensation prediction unit 110 outputs prediction information 24 including search method information 241 indicating a search method used in the motion compensation prediction process and search range information 242 indicating a search range used in the motion compensation prediction process. . The search method indicates what compensation method is used for motion compensation. The search range indicates how much the search range for motion is.
More specifically, the motion compensation prediction unit 110 captures the past image frames stored in the memory 921 and the image frames 21 input via the image input interface 930 into the processor 910. The motion compensation prediction unit 110 compares the past image frame and the image frame 21 in units of small blocks, searches for similar image positions, and outputs prediction information 24 including coordinates indicating the similar positions. In addition, the motion compensation prediction unit 110 writes the prediction frame 22 that is a similar video in the memory 921.

<Subtraction process S20>
In step S <b> 104, the subtraction unit 120 calculates a difference between the image frame 21 and the predicted frame 22. The subtraction unit 120 is also referred to as a differentiator. Specifically, the subtraction unit 120 takes the difference between the prediction frame 22 generated by the motion compensation prediction unit 110 and the image frame 21 and outputs the difference 26 to the transform coding unit 130. More specifically, the subtraction unit 120 takes the predicted frame 22 stored in the memory 921 into the processor 910 and subtracts the image frame 21 and the predicted frame 22 input via the image input interface 930 by the processor 910. The difference 26 that is the result of the subtraction process is stored in the memory 921.

<Transform coding process S30>
Next, the transform encoding unit 130 encodes the difference 26 between the image frame 21 and the predicted frame 22 and outputs the encoded difference 26 as encoded data 25.
Specifically, in step S <b> 105, the transform encoding unit 130 acquires the difference 26 output from the subtracting unit 120, and performs discrete cosine transform (DCT: Discrete Cosine Transform) on the difference 26. The transform coding unit 130 performs DCT such as orthogonal transform and quantization. The transform encoding unit 130 outputs the difference 26 subjected to DCT as encoded data 25. More specifically, the transform encoding unit 130 takes the difference 26 stored in the memory 921 into the processor 910 and writes the result of the DCT as encoded data 25 to the memory 921.

<Entropy encoding process S40>
The entropy encoding unit 140 includes a variable length encoding process S41 and an addition control process S42. The entropy encoding unit 140 is also referred to as VLC (Variable-Length Coding).

<Variable length encoding process S41>
In step S106, the variable length coding unit 141 of the entropy coding unit 140 executes a variable length coding process S41. The variable length encoding process S41 is a process in which the variable length encoding unit 141 compresses the encoded data 25 as the variable length code data 27 by encoding the encoded data 25 using the variable length code. Specifically, the variable length encoding unit 141 performs variable length code processing in which the encoded data 25 written to the memory 921 by the transform encoding unit 130 is taken into the processor 910 and encoded using the variable length code. The result is written in the memory 921 as the variable length code data 27.

<Additional control processing S42>
In step S107, the addition control unit 142 of the entropy encoding unit 140 executes an addition control process S42. The addition control process S42 is a process in which the addition control unit 142 adds the prediction information 24 to the variable length code data 27. The addition control unit 142 outputs the variable length code data 27 to which the prediction information 24 is added as the compressed data 23. Specifically, the addition control unit 142 detects the data break position for the variable-length code data 27 and adds the prediction information 24 to the detected break position. The addition control unit 142 adds the search method information 241 of the prediction information 24 to the sequence layer of the variable length code data 27, and the search range of the prediction information 24 to the frame layer of the variable length code data 27. Information 242 is added.

The compressed data 23 according to the present embodiment will be described with reference to FIG.
The compressed data 23 has a sequence layer RS, a frame layer RF, and a macroblock layer RMB. The addition control unit 142 detects a data delimiter position P for the variable-length code data 27 serving as the compressed data 23 and adds the prediction information 24 to the detected delimiter position P. The addition control unit 142 adds the search method information 241 of the prediction information 24 to the sequence layer RS of the variable length code data 27 and adds the search method information 241 of the prediction information 24 to the frame layer RF of the variable length code data 27. Search range information 242 is added.
Thus, the addition control unit 142 stores the search method in a layer that stores information related to the entire compressed data called the sequence layer RS. Further, the addition control unit 142 stores a search range in a layer that stores information related to each image called a frame layer RF.
MPEG compressed data has a data area that can be freely used by a user. This data area is called a user area or SEI (Supplemental Enhancement Information). The addition control unit 142 stores the prediction information 24 in this data area. The data area released to these users is defined as skipping data when it cannot be interpreted on the receiving side, and the compressed data 23 in which these pieces of information are multiplexed is stored in a device that does not perform image processing. Even if input, it is possible to correctly decode the image. Therefore, encoding apparatus 100 according to the present embodiment can completely ensure compatibility with existing apparatuses.

The detailed operation of the additional control process S42 according to the present embodiment will be described with reference to FIGS.
The addition control unit 142 of the entropy encoding unit 140 includes a multiplexing unit 421, a switch unit 422, and a multiplexing control unit 423. The switch unit 422 switches the input to the addition control unit 142 according to the control of the multiplexing control unit 423. The type of the prediction information 24 includes a parameter related to the entire compressed data and a parameter related to each image frame of a plurality of image frames constituting the video. The multiplexing control unit 423 controls the switch unit 422 according to the type of the prediction information 24, and sends the variable length code data 27 and the prediction information 24 to the multiplexing unit 421.

The detailed operation of the addition control process S42 will be further described with reference to FIG.
In step S421, the multiplexing control unit 423 switches the input of the switch unit 422 to the variable length coding unit 141.
In step S422, the multiplexing control unit 423 causes the variable length coding unit 141 to output the variable length code data 27 of the sequence layer RS to the multiplexing unit 421.
In step S423, the multiplexing control unit 423 determines whether the output of the variable length code data 27 of the sequence layer RS has been completed. If the output has not ended, the process returns to step S422. If the output has been completed, the process proceeds to step S424.
In step S424, the multiplexing control unit 423 switches the input of the switch unit 422 to the prediction information 24.
In step S425, the multiplexing control unit 423 causes the multiplexing unit 421 to output the prediction information 24 related to the sequence layer RS among the prediction information 24. The prediction information 24 related to the sequence layer RS is search method information 241.
In step S426, the multiplexing control unit 423 determines whether the output of the prediction information 24 related to the sequence layer RS in the prediction information 24 is completed. If the output has not ended, the process returns to step S425. If the output has been completed, the process proceeds to step S427.

In step S427, the multiplexing control unit 423 switches the input of the switch unit 422 to the variable length coding unit 141.
In step S428, the multiplexing control unit 423 causes the variable length coding unit 141 to output the variable length code data 27 of the frame layer RF to the multiplexing unit 421.
In step S429, the multiplexing control unit 423 determines whether the output of the variable length code data 27 of the frame layer RF has been completed. If the output has not ended, the process returns to step S427. If the output has been completed, the process proceeds to step S430.
In step S430, the multiplexing control unit 423 switches the input of the switch unit 422 to the prediction information 24.
In step S431, the multiplexing control unit 423 causes the multiplexing unit 421 to output the prediction information 24 related to the frame layer RF among the prediction information 24. The prediction information 24 related to the frame layer RF is search range information 242.
In step S432, the multiplexing control unit 423 determines whether or not the output of the prediction information 24 related to the frame layer RF in the prediction information 24 is completed. If the output has not ended, the process returns to step S431. If the output has been completed, the process proceeds to step S433.
In step S433, the multiplexing control unit 423 switches the input of the switch unit 422 to the variable length coding unit 141, and causes the multiplexing unit 431 to output all the variable length code data 27 of the macroblock layer RMB.
The multiplexing unit 431 multiplexes the variable length code data 27 and the prediction information 24 input via the switch unit 422 and outputs the result as compressed data 23. Specifically, the multiplexing unit 431 writes the compressed data 23 to the memory 921. The communication device 950 sends the compressed data 23 written in the memory 921 by the multiplexing unit 421 to the network. Alternatively, the output interface 940 may output the compressed data 23 to an output device such as a display. The multiplexing unit 421 is also referred to as a MUX (multiplexer).

*** Other configurations ***
In the present embodiment, the functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 are realized by software. However, as a modification, the functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 may be realized by hardware.

The configuration of encoding apparatus 100 according to a modification of the present embodiment will be described using FIG.
As illustrated in FIG. 6, the encoding device 100 includes hardware such as a processing circuit 909, an image input interface 930, an output interface 940, and a communication device 950.

  The processing circuit 909 is a dedicated electronic device that realizes the functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 and the storage unit 150 described above. Circuit. Specifically, the processing circuit 909 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA, an ASIC, or an FPGA. GA is an abbreviation for Gate Array. ASIC is an abbreviation for Application Specific Integrated Circuit. FPGA is an abbreviation for Field-Programmable Gate Array.

  The functions of the motion compensation prediction unit 110, the subtraction unit 120, the transform coding unit 130, the variable length coding unit 141, and the addition control unit 142 may be realized by one processing circuit 909, or a plurality of functions may be realized. The processing circuit 909 may be distributed and realized.

  As another modification, the function of each unit of each device may be realized by a combination of software and hardware. That is, some functions of the encoding device 100 may be realized by dedicated hardware, and the remaining functions may be realized by software.

  The processor 910, the storage device 920, and the processing circuit 909 of the encoding apparatus 100 are collectively referred to as “processing circuit”. That is, regardless of the configuration of the encoding device 100 shown in FIG. 1 or FIG. 6, the motion compensation prediction unit 110, the subtraction unit 120, the transform encoding unit 130, and the variable length encoding unit 141. The function of the additional control unit 142 and the storage unit 150 are realized by a processing circuit.

  “Part” may be read as “process” or “procedure” or “processing”. Further, the function of “unit” may be realized by firmware.

*** Explanation of effects of this embodiment ***
According to encoding apparatus 100 according to the present embodiment, compressed data includes parameters related to motion compensation, specifically, a motion compensation method and a motion search range. In a normal encoder, prediction information is not included in the compressed data, and on the device side that performs the image processing for receiving the compressed data, what compensation method is used for the motion compensation, There was no way to know what the range of motion search was. The fact that these pieces of prediction information cannot be known means that the maximum amount of motion cannot be assumed when performing image processing for detecting the motion of an object, and the accuracy of image processing cannot be defined. That is, the image processing is inaccurate. According to encoding apparatus 100 according to the present embodiment, it is possible to notify a parameter relating to motion compensation, specifically, a motion compensation search method and a motion search range, to the image processing apparatus side. Therefore, according to coding apparatus 100 according to the present embodiment, it is possible to improve accuracy by limiting to motion estimation within the search range. Also, according to encoding apparatus 100 according to the present embodiment, motion information can be corrected according to a motion search method, and the efficiency of image processing can be improved. Further, according to the encoding device 100 according to the present embodiment, even when there are encoders having different search methods or search ranges among manufacturers, by performing normalization based on this parameter, It is possible to minimize differences between encoder manufacturers. This is extremely important when building a multi-vendor system.

  Usually, in the video compression method using the MPEG method, control is performed so that the amount of information when an image is divided into fine macro blocks and compressed for each macro block is reduced. In this control, motion compensation processing for searching for a region similar to the target macroblock is performed. As a result of this motion compensation processing, a motion vector is generated and becomes information similar to the optical flow in the sense of motion tracking. However, in the video compression method, it is determined whether to use a motion vector by using the amount of information after compression as an evaluation scale. If it can be determined that the amount of information is smaller without using a motion vector, no motion vector is assigned to the target macroblock. However, according to coding apparatus 100 according to the present embodiment, it is possible to obtain motion vectors for all areas of the screen.

  According to encoding apparatus 100 according to the present embodiment, it is possible to improve and stabilize the processing accuracy of image processing that uses motion vectors obtained by motion compensation prediction. Furthermore, according to encoding apparatus 100 according to the present embodiment, it is possible to prevent the generated compressed data from affecting the operation of an existing receiving apparatus. That is, it means that the compressed data generated by the encoding apparatus 100 according to the present embodiment is completely compliant with the MPEG system.

Note that the encoding apparatus according to the present embodiment performs an operation according to an encoding international standard system such as MPEG / H261 except for prediction information. In this case, the entropy encoding unit is configured by only two elements of the variable length encoding unit 141 and the multiplexing unit 421 in the configuration shown in FIG. 2, and the variable length encoding unit 141 and the multiplexing unit 421 are directly connected. Is done.
An encoding device that conforms to the international standard described above performs an operation in accordance with a video international encoding system represented by ISO / MPEG or ITU / H261. H. For example, Yasuda et al. "International standard for MPEG / multimedia coding" (1994), Fujiwara et al. "Latest MPEG textbook" (1994), IEICE "Knowledge Base". It is described in literature such as Chapter 11 International Standard.

Embodiment 2. FIG.
In this embodiment, matters to be added to the description of Embodiment 1 will be described. In particular, a specific example of the prediction information 24 will be described.
FIG. 7 is a diagram illustrating an example of a search range included in the prediction information 24 according to the present embodiment.
FIG. 7 shows that the search range for motion compensation differs greatly for each camera manufacturer. For example, the camera of camera manufacturer A has a search range of ± 80 pixels in the horizontal direction and ± 40 pixels in the vertical direction. At this time, the search range information 242 is information of ± 80 pixels in the horizontal direction and ± 40 pixels in the vertical direction. The camera of camera manufacturer B has ± 40 pixels in the horizontal direction and ± 80 pixels in the vertical direction. At this time, the search range information 242 is information of ± 40 pixels in the horizontal direction and ± 80 pixels in the vertical direction. The addition control unit 142 embeds the search range information 242 representing these search ranges in the compressed data 23 as the search range information 242 of the frame layer RF described in the first embodiment.

8 to 11 are diagrams illustrating examples of search methods represented by search method information included in the prediction information 24 according to the present embodiment. As shown in FIGS. 8 to 11, there are a plurality of search methods as search methods. FIG. 8 is a schematic diagram showing a full search method. FIG. 9 is a schematic diagram showing a step search method. Step search is also referred to as multi-stage search. FIG. 10 is a schematic diagram showing a diamond search method. FIG. 11 is a schematic diagram showing a hexagon search method.
In addition, steps S201 to S211 in FIG. 12 are flowcharts showing a full search method. Steps S301 to S313 in FIG. 13 are flowcharts showing a step search method. Steps S401 to S406 in FIG. 14 are flowcharts showing a diamond search method. Steps S501 to S506 in FIG. 15 are flowcharts showing a hexagon search method.
In the full search, optimal motion information can be generated within the search range by searching all areas. The amount of processing can be reduced by performing the step search while thinning out. Diamond search is a method of gradually expanding the search range from the vicinity of the starting point where the search is started. Hexagon search is a method that achieves both search accuracy and search processing reduction.
Specific examples of the search range method 241 include identifiers of these search methods. Identifiers are assigned to these search methods, and the addition control unit 142 embeds the identifiers in the compressed data 23 as the sequence layer search method information 241 described in the first embodiment.

  Although a rectangular search range is shown as an example of the search range of the prediction information, the search range is not necessarily a rectangle, and there may be a search range having a circle or ellipse range or a combination of rectangles. It goes without saying that information defining these figures can be embedded as search range information. Also, as an example of the motion search method, an example of a specific method has been shown, but it is not necessarily a single method, and it is possible to embed such information even in the case of a combination of a plurality of search methods. Needless to say.

In Embodiments 1 and 2, each unit of encoding apparatus 100 constitutes encoding apparatus 100 as an independent functional block. However, the configuration may not be the same as that in the above-described embodiment, and the configuration of the encoding device 100 is arbitrary. The functional blocks of the encoding device 100 are arbitrary as long as the functions described in the above-described embodiments can be realized. These functional blocks may be configured in any other combination or arbitrary block configuration in the encoding apparatus 100.
Also, the encoding apparatus 100 may be a system configured from a plurality of apparatuses instead of a single apparatus.

Although Embodiments 1 and 2 have been described, a combination of a plurality of portions may be implemented among these embodiments. Alternatively, one part of these embodiments may be implemented. In addition, these embodiments may be implemented in any combination as a whole or in part.
The above-described embodiments are essentially preferable examples, and are not intended to limit the scope of the present invention, its application, and uses, and various modifications can be made as necessary. .

  21 picture frame, 22 prediction frame, 23 compressed data, 24 prediction information, 241 search method information, 242 search range information, 25 encoded data, 26 difference, 27 variable length code data, 100 encoding device, 110 motion compensation prediction unit , 120 subtracting unit, 130 transform coding unit, 140 entropy coding unit, 141 variable length coding unit, 142 additional control unit, 421 multiplexing unit, 422 switch unit, 423 multiplexing control unit, 150 storage unit, 909 processing Circuit, 910 processor, 920 storage device, 921 memory, 922 auxiliary storage device, 930 image input interface, 940 output interface, 950 communication device, 610 encoding method, 620 encoding program, S100 encoding process, S10 motion compensation prediction process , S20 Subtraction process S30 transform encoding processing, S40 entropy coding process, S41 variable length coding process, S42 addition control process, RS sequence layer, RF frame layer, RMB macroblock layer, P-separated position.

Claims (6)

A motion compensation prediction unit that performs motion compensation prediction processing on an image frame and outputs a prediction frame, and outputs prediction information used in the motion compensation prediction processing;
A transform encoding unit that encodes a difference between the image frame and the prediction frame and outputs the encoded difference as encoded data;
A variable length encoding unit that compresses the encoded data as variable length code data by encoding the encoded data using a variable length code;
An encoding apparatus comprising: an addition control unit that adds the prediction information to the variable-length code data. The additional controller is
The encoding apparatus according to claim 1, wherein a data delimiter position is detected for the variable-length code data, and the prediction information is added to the detected delimiter position. The motion compensation prediction unit
3. The prediction information including search method information representing a search method used for the motion compensation prediction process and search range information representing a search range used for the motion compensation prediction process is output. Encoding device. The additional controller is
The search method information of the prediction information is added to a sequence layer of the variable length code data, and the search range information of the prediction information is added to a frame layer of the variable length code data. Item 4. The encoding device according to Item 3. The motion compensation prediction unit performs a motion compensation prediction process on the image frame and outputs a prediction frame, and outputs prediction information used in the motion compensation prediction process,
The transform encoding unit encodes the difference between the image frame and the prediction frame, and outputs the encoded difference as encoded data.
A variable length encoding unit compresses the encoded data as variable length code data by encoding the encoded data using a variable length code,
An encoding method in which an addition control unit adds the prediction information to the variable-length code data. A motion compensated prediction process for outputting the prediction information used in the motion compensated prediction process while performing a motion compensated prediction process on the image frame and outputting a predicted frame;
A transform encoding process that encodes a difference between the image frame and the prediction frame and outputs the encoded difference as encoded data;
A variable length encoding unit that compresses the encoded data as variable length code data by encoding the encoded data using a variable length code;
An encoding program that causes a computer to execute an addition control unit that adds the prediction information to the variable-length code data.

Images