JP5212319B2 - Encoding apparatus, encoding method, and encoding program - Google Patents

Description

  The present invention relates to an encoding device, an encoding method, and an encoding program for dividing encoded data and outputting the divided data to a network.

  2. Description of the Related Art Conventionally, an apparatus that divides data when encoding moving image data and outputting the data to a network is known (for example, see Patent Document 1). In this apparatus, an MTU (Maximum Transmission Unit) that is the maximum value of a datagram that can be transmitted with one output is taken into consideration, and the encoded data is divided into sizes that fit within the MTU size. In order to perform network communication, data such as a header (hereinafter referred to as “additional data”) is newly added to the divided data, and the data is output. Transmission efficiency can be improved by dividing and outputting the encoded data so as to fit in the MTU size.

Japanese Patent No. 3888505

  However, in the conventional apparatus, the encoded data is simply divided in accordance with the MTU size. Although the size of the simply divided data may be different, additional data must be added uniformly even for small data. Therefore, the ratio of the additional data to the entire data to be output may increase, which causes the processing efficiency to deteriorate. On the other hand, if the data is significantly compressed considering only the processing efficiency, the quality of the moving image is degraded. Therefore, it has been difficult in the past to appropriately reduce the ratio of additional data in the entire data.

  An object of the present invention is to provide an encoding device, an encoding method, and an encoding program capable of appropriately reducing the ratio of additional data in the entire output data.

  An encoding device according to a first aspect of the present invention is an encoding device that encodes image data input by an image input means and outputs the encoded data to another device via a network. Measuring means for measuring the MTU size of the network between the encoder and the encoding device, and dividing the encoded data into a size that fits within the MTU size measured by the measuring means, and adding additional data necessary for network communication A setting means for setting a data size of encoded data in which an overhead ratio, which is a ratio of the additional data to the entire data to be output, as a target value when newly added to each divided data; Encoding means for encoding the image data to a data size set as: and an encoding data encoded by the encoding means The data is divided by the size that fits MTU size measured by said measuring means and an output means for outputting to the device.

  Since the encoding apparatus according to the first aspect can encode the image data so that the overhead ratio becomes the target value, the overhead ratio can be appropriately reduced. Therefore, the processing efficiency can be improved.

  Preferably, the setting means sets the data size at the point where the overhead ratio switches from decrease to increase when the data size of the encoded data is increased, to the set value. In this case, the set value can be set accurately and easily, and the overhead ratio can be reduced to an appropriate value.

  The encoding apparatus may include a calculation unit that calculates a reference data size that is a data size of encoded data when the image data is encoded without setting the setting value. When the data size of the point is different from the reference data size, the setting means is a first point that is a point having the smallest data size among a plurality of points, the data size of which is larger than the reference data size. And the data size of the target point is set to the set value with any one of the second points having the largest data size among the points having a data size smaller than the reference data size as a target point. That's fine. In this case, the encoding apparatus can reduce the overhead ratio while generating encoded data having a data size close to the reference data size. Therefore, the processing efficiency can be improved without sharply degrading the image quality.

  The setting means sets the target point as either the first point or the second point depending on whether the reference data size is larger than a threshold value provided between the first point and the second point. You may decide whether to do it. In this case, the encoding apparatus can easily set the setting value using the threshold value.

  The setting means, when a plurality of MTU sizes having different values are measured as a result of measuring the MTU size of the network between the plurality of apparatuses by the measuring means, It is desirable to set one set value common to the plurality of devices based on the MTU size. The output means may divide the same encoded data by a size that fits in each MTU size measured for the output destination device, and output the divided data to each of the plurality of devices. In this case, the encoding apparatus can easily and appropriately reduce the overhead ratio by a series of processes even when the encoded data is output to a plurality of devices having different MTU sizes.

  The encoding device may include priority determination means for determining the priority of each of the plurality of devices. The setting unit may set the setting value common to the plurality of devices based on the priority determined by the priority determination unit. In this case, the encoding device can preferentially reduce the processing load of a high-priority device through a series of encoding processing without performing encoding processing separately for each output destination device.

  The priority determination means may determine the priority using at least one of network bandwidth information of each of the devices, MTU size, and setting information indicating a priority level. In this case, the encoding apparatus can appropriately determine the priority using at least one of information on a band that affects the processing speed of the device, the MTU size, and setting information indicating the degree of priority. .

  An encoding method according to a second aspect of the present invention is an encoding method performed by an encoding device that encodes image data input by an image input means and outputs the encoded data to another device via a network. A measurement step of measuring an MTU size of the network between the device and the encoding device; and encoding data is divided by a size that fits within the MTU size measured by the measurement step, and network communication is performed. Setting for setting the data size of the encoded data whose target value is the overhead ratio, which is the ratio of the additional data to the entire data to be output, when the additional data necessary for the data is newly added to each divided data An encoding step for encoding the image data to a data size set as the set value, and the code And an output step of outputting the coded data, the device is divided by the size that fits in the measured by the measuring step MTU size by step.

  According to the encoding method according to the second aspect, since the image data can be encoded so that the overhead ratio becomes the target value, the overhead ratio can be appropriately reduced. Therefore, the processing efficiency can be improved.

  An encoding program according to a third aspect of the present invention is an encoding program for encoding image data input by an image input means, and outputting the encoded data to another device via a network. A measurement step of measuring the MTU size of the network between the device and the computer is divided into encoded data by a size that fits within the MTU size measured by the measurement step, and additional data necessary for network communication is obtained. A setting step for setting a data size of encoded data in which an overhead ratio, which is a ratio of the additional data with respect to the entire data to be output, is a target value when a new value is added to each divided data; and the setting value An encoding step for encoding the image data to a data size set as: The coded data coded by step, characterized in that to execute an output step of outputting to the device is divided in a size fit to the measured MTU size by said measuring step.

  According to the encoding program according to the third aspect, since the image data can be encoded so that the overhead ratio becomes the target value, the overhead ratio can be appropriately reduced. Therefore, the processing efficiency can be improved.

2 is a block diagram showing an electrical configuration of the video conference apparatus 1. FIG. 4 is a schematic diagram of data stored in an MTU size storage area 121 of a RAM 12. FIG. 3 is a schematic diagram of data stored in an available bandwidth storage area 122 of a RAM 12. FIG. 4 is a schematic diagram of data stored in a setting priority value storage area 131 of the HDD 13; FIG. 2 is a functional block diagram of the video conference apparatus 1. FIG. It is a figure which shows the structure of a NAL unit, and the structure of a transmission packet. It is a figure which shows an example of the data size of the transmission packet before adjusting the data size of encoding data. It is a figure which shows an example of the data size of the transmission packet after adjusting the data size of encoding data. It is a flowchart of the encoding process which the video conference apparatus 1 performs. It is a figure which shows an example of the mapping table produced when the measured MTU size is one. It is a figure which shows an example of the mapping table produced when there are two or more measured MTU sizes. It is a flowchart of the data size adjustment process performed by an encoding process.

  Hereinafter, a video conference apparatus 1 which is an embodiment of an encoding apparatus according to the present invention will be described with reference to the drawings. The drawings to be referred to are used for explaining technical features that can be adopted by the present invention. The configuration of the apparatus, the flowcharts of various processes, and the like described in the drawings are not intended to be limited to these, but are merely illustrative examples.

  The video conference apparatus 1 is connected to one or more other video conference apparatuses 1 via the network 8 (see FIG. 1). Each video conference device 1 inputs and outputs image data and audio data. As a result, users at multiple bases can share video and audio. Therefore, even when all the users are not at the same base, the users can smoothly execute the conference.

  The electrical configuration of the video conference apparatus 1 will be described with reference to FIGS. The video conference apparatus 1 includes a CPU 10 that controls the video conference apparatus 1. A ROM 11, a RAM 12, a hard disk drive (hereinafter referred to as “HDD”) 13, and an input / output interface 19 are connected to the CPU 10 via a bus 18.

  The ROM 11 stores a program and an initial value for operating the video conference device 1. The RAM 12 temporarily stores various information used in the control program. The HDD 13 is a non-volatile storage device that stores various types of information. Instead of the HDD 13, a storage device such as an EEPROM or a memory card may be used.

  The input / output interface 19 is connected to an audio input processing unit 21, an audio output processing unit 22, a video input processing unit 23, a video output processing unit 24, an operation unit 25, and an external communication I / F 26. The voice input processing unit 21 processes input of voice data from the microphone 31 that inputs voice. The audio output processing unit 22 processes the operation of the speaker 32 that outputs audio. The video input processing unit 23 processes input of video data (moving image data) from the camera 33 that captures video. The video output processing unit 24 processes the operation of the display device 34 that displays video. The operation unit 25 is used for a user to input various instructions to the video conference apparatus 1. The external communication I / F 26 connects the video conference device 1 to the network 8.

  The RAM 12 will be described in detail. The RAM 12 is provided with various storage areas such as an MTU size storage area 121 and an available bandwidth storage area 122.

  As shown in FIG. 2, in the MTU size storage area 121, information related to the MTU size of the network 8 with a device (another video conference apparatus 1 in the present embodiment) connected via the network 8 is stored. Remembered. The MTU (Maximum Transmission Unit) size is the maximum value of data that can be transmitted in one data transfer. For example, in an Ethernet (registered trademark) type LAN environment, the maximum frame size of Ethernet is 1518 bytes. Therefore, 1500 bytes excluding the Ethernet header and FCS (Frame Check Sequence) is the MTU size. Therefore, when the data to be output is large, it is desirable to divide and output the data so that the sum of the IP header and the data portion of the IP packet is within 1500 bytes.

  The video conference apparatus 1 according to the present embodiment measures the MTU size for each connected device at the start of communication. The measured MTU size is stored in the MTU size storage area 121, and processing such as data division is performed. Further, the video conference apparatus 1 can generate one piece of encoded data that is optimal for a plurality of devices, using the priority of each device among the plurality of devices. The video conference apparatus 1 can calculate a count (G) indicating this priority from the MTU size. More specifically, processing is performed using a device with a small MTU size as a device with a high priority and a reciprocal of the MTU size as a coefficient (G). Details of this processing will be described later.

  As shown in FIG. 3, the available bandwidth storage area 122 stores the available bandwidth of the network 8 between the video conference device 1 and each device. The video conference apparatus 1 can also perform processing using a device with a low (narrow) usable bandwidth as a device with a high priority and a reciprocal of the usable bandwidth as a coefficient (G).

  The HDD 13 will be described in detail. The HDD 13 is provided with various storage areas such as a setting priority value storage area 131. As shown in FIG. 4, the setting priority value storage area 131 stores setting priority values input and set in advance by the user. The video conference apparatus 1 can generate one piece of encoded data that is optimal for a plurality of devices, using the setting priority value.

  Next, an outline of image data processing in the video conference apparatus 1 will be described. The video conference apparatus 1 converts the image data input from the camera 33 into H.264. Image compression encoding is performed based on the H.264 standard to generate encoded data. The generated encoded data is output to another device via the network 8. Note that the video conference device 1 decodes encoded data input from another device via the network 8 and causes the display device 34 to display an image. In addition, audio data is input / output, and audio is generated based on audio data input from another device. However, since these processes are not the main part of the present invention, the image compression encoding and encoded data output processes will be described below.

  Image compression coding includes intra-frame coding and inter-frame coding. Intraframe coding is coding performed by intra prediction within one frame of image data of a plurality of consecutive frames of image data input by a camera. An I picture (Intra-coded Picture) that is encoded data generated by intra-frame encoding can be decoded independently without referring to other pictures. On the other hand, in interframe coding, prediction data is calculated by referring to data of a frame different from data of a frame to be encoded among continuous frame data, and the calculated prediction error is encoded. Encoded data generated by interframe coding includes a P picture (Predictive-coded Picture) and a B picture (Bidirectional-coded Picture). Decoding of P and B pictures requires pictures that are referenced during encoding. However, the data size of the P picture and the B picture is smaller than the data size of the I picture.

  As shown in FIG. 5, in the video conference apparatus 1, first, DCT / quantization 42 is performed on the input image 41 from the camera 33. Here, the coefficient transformed by DCT (Discrete Cosine Transform) is quantized according to the quantization parameter. Next, inverse quantization / inverse DCT 43 is performed on a part of the quantized data. The data subjected to the inverse quantization / inverse DCT 43 is subjected to the deblocking filter 44 and stored in the frame memory 45.

  When intra-frame coding is performed, intra-screen prediction 46 is performed on the data stored in the frame memory 45, and further DCT / quantization 42 is performed. Entropy encoding 47 is performed on the quantized data, and encoded data 48 (I picture) is generated.

  When interframe coding is performed, motion prediction 51 is performed using the input image 41, and motion compensation 52 based on another predicted image in the frame memory 45 is performed. The prediction error calculated by the motion compensation 52 is subjected to weighted prediction 53 using a weighting factor related to brightness, and further, DCT / quantization 42 is performed. Entropy encoding 47 is performed on the quantized data, and encoded data 48 (P picture or B picture) is generated.

  Based on the result of the MTU size measurement 55, the generated encoded data 48 is divided into transmission packets 49 that fit within the measured MTU size. The divided encoded data becomes the target of the network output 50.

  Next, division of encoded data will be described. Packetization modes defined for packet transmission include a single NAL (Network Abstraction Layer) unit mode, a non-interleave mode, and an interleave mode. In the video conference apparatus 1, the single NAL unit mode is adopted. In the single NAL unit mode, each picture that is encoded data is encapsulated in a single NAL unit packet. One encapsulated NAL unit is a unit of data to be output.

  As shown in FIG. 6, the NAL unit has a NAL header, an RBSP, and a trailing bit in order from the top. An identifier “nal_unit_type” indicating the type of the NAL unit is added to the NAL header. RBSP (Raw Byte Sequence Payload) stores raw data that has been compression-encoded. The trailing bit is an adjustment bit for specifying the last bit position of the RBSP.

  The encapsulated NAL unit is divided so as to fit within the MTU size of the network 8 with the output destination device. The divided packets (hereinafter referred to as “divided packets”) are output to the device via the network 8. Here, it is necessary to newly add data such as a header necessary for network communication (hereinafter referred to as “additional data”) to the divided packet. According to the communication mode employed in the present embodiment, an IP header and a TCP header are included at the head of the first fragmented packet ((1) in FIG. 6), and the subsequent fragmented packets ((2) and It is necessary to add an IP header as additional data at the beginning of (3)). One packet to be output to the network 8 is constituted by the divided packet and the additional data.

  However, if the NAL unit is simply divided based on the MTU size, the data size of the divided encoded data may not be uniform. In the example shown in FIG. 7, the data size of the transmission packet (1) and (2) is equal to the MTU size, but the data size of the transmission packet (3) is smaller than the MTU size. In order to perform network communication, additional data must be added to small data such as the encoded data of (3). Therefore, the ratio of additional data to the entire data to be output (hereinafter referred to as “overhead ratio”) may increase. As the overhead ratio increases, the processing efficiency deteriorates and the video quality decreases.

  As shown in FIG. 5, the video conference apparatus 1 according to the present embodiment can adjust the data size of the encoded data by the NAL unit code amount control 56 based on the result of the MTU size measurement 55. Specifically, based on the measured MTU size, the data size of the encoded data whose overhead ratio is the target value is set. The quantization parameter used in the DCT / quantization 42 is determined so that encoded data having the set data size is generated. For example, the data is compressed and encoded so as to be reduced by the encoded data (3) shown in FIG. As a result, as shown in FIG. 8, it is not necessary to add additional data to encoded data having a small data size, and the overhead ratio is reduced. As described above, the video conference apparatus 1 can reduce the overhead ratio and prevent the video quality from deteriorating.

  In addition, when the quantization parameter is determined only for the purpose of reducing the overhead ratio, a problem occurs. For example, if the compression ratio is rapidly increased so as to reduce both the encoded data of (2) and (3) shown in FIG. 7, the overhead ratio decreases, but the video quality deteriorates. If the compression ratio is rapidly decreased, the processing load such as decoding increases. The video conference apparatus 1 according to the present embodiment can reduce the overhead ratio without abruptly changing the compression ratio by appropriately setting the data size of the encoded data to be generated. The above processing will be described in detail below.

  The encoding process performed by the video conference apparatus 1 will be described with reference to FIGS. The encoding process is executed by the CPU 10 in accordance with a program stored in the ROM 11. The encoding process is started when an instruction to execute input / output of image data is input.

  As shown in FIG. 9, when the encoding process is started, various data are initialized (S1). For each device connected via the network 8, the MTU size of the network 8 with the device is measured (S2). The measurement of the MTU size may be performed using a known method such as Path MTU Discovery.

  Next, it is determined whether or not a plurality of MTU sizes having different values have been measured (S3). When the number of connected devices is one, or when the MTU sizes of the plurality of connected devices are all the same (S3: NO), based on one measured MTU size. A mapping table is created (S4). The mapping table is a table that associates the data size of the encoded data with the overhead ratio.

A method for creating a mapping table from one MTU size will be described. The numbers used are shown below.
MTU size (unit: number of bytes): mtu
IP header data size (unit: number of bytes): ip
TCP header data size (unit: number of bytes): tcp
Data size of encoded data (unit: number of bytes): D
Number of divisions of encoded data: N
Overhead ratio (unit:%): X
N is an integer greater than or equal to 0, and N = 0 when there is one packet, and N = 1 when there are two divided packets.

First, when the encoded data is the data size D, the division number N that satisfies the following equation (1) is calculated.
{D− (mtu−ip−tcp)} / (mtu−ip) ≦ N <{D + (mtu−ip) − (mtu−ip−tcp)} / (mtu−ip) (1)
The overhead ratio X in this case is calculated by the following formula (2) using the calculated value of N.
X = {ip × (N + 1) + tcp} / D (2)
By performing the above processing while changing the data size D, a mapping table for a single MTU size is created.

  FIG. 10 shows an example of a mapping table created when the measured MTU size is one. As shown in FIG. 10, if the number of divisions does not change, the overhead ratio decreases as the data size of the encoded data increases. When the data size increases and the number of divisions increases by 1, the added IP header increases by 1, so the overhead ratio increases. Therefore, in consideration of the overhead ratio, it is desirable to encode to the data size where the number of divisions increases when the data size is increased, that is, the overhead ratio is switched from decrease to increase.

  Returning to the description of FIG. When a plurality of devices are connected to the video conference device 1 and a plurality of MTU sizes having different values are measured (S3: YES), a coefficient (G) indicating the priority of each device is acquired. (S6). As described above, the video conference apparatus 1 can use various coefficients (G) (see FIG. 2 to FIG. 4). In this embodiment, the user determines in advance which coefficient (G) is used. Let me choose. In the process of S6, the coefficient (G) selected by the user is acquired from the RAM 12 or the HDD 13. Next, a mapping table is created using the acquired coefficient (G) and the plurality of acquired MTU sizes (S7).

A method for creating a mapping table from a plurality of MTU sizes will be described. First, when the MTU sizes mtux = mtu1, mtu2,..., Mtum of m devices are measured, the data size D is a certain value from the above equation (1) for the MTU size of each device. Nm = N1, N2,..., Nm are calculated. Next, the overhead ratio X is calculated by dividing the total amount of additional data output to m devices by the total amount of encoded data output to m devices. The calculation formula of the overhead ratio X in this case is shown below.
When considering the priority of each device X = Σ [{Gi × ip × (Nj + 1) + tcp} / (Σ (Gi) × D)] (i = 1,..., M) (j = 1,. .., m)
When priority of each device is not considered X = Σ {ip × (Ni + 1) + tcp} / (D × m) (i = 1,..., M)
The mapping table is created by performing the above processing while changing the data size D.

  FIG. 11 shows an example of a mapping table created when there are a plurality of measured MTU sizes. In FIG. 11, for reference, a table created using only the MTU size of device A, a table created using only the MTU size of device B, and created using two MTU sizes of devices A and B Four tables are shown: a table (with priority taken into account) and a table created using the two MTU sizes of devices A and B (without taking into account priority). By creating a table using the two MTU sizes of the devices A and B, a mapping table common to the devices A and B can be created. Furthermore, by considering the priority, the MTU size of the device having a high priority can be reflected in the mapping table larger.

  Returning to the description of FIG. When the mapping table is created (S4 or S7), the image data is encoded for one frame (S8). This encoding is performed without considering the MTU size. For example, it may be performed based on a quantization parameter set as a default value, a quantization parameter set considering only the performance of the output destination device without considering the MTU size, and the like. Next, the data size of the encoded data generated by the encoding is calculated (S9). Whether the generated encoded data is encoded data whose data size can be changed by changing the quantization parameter is determined based on the type of data (S10). If it is encoded data whose data size cannot be changed (S10: NO), the process directly proceeds to S12. If the encoded data is related to a picture and the encoded data can change the data size (S10: YES), a data size adjustment process is performed (S11).

  As shown in FIG. 12, in the data size adjustment processing, the data size (S9, see FIG. 9) of the encoded data generated without considering the MTU size is set as the reference data size (S21). Next, processing for selecting an inflection point as a target point is performed. The inflection point indicates that when the data size of the encoded data is increased, the number of divisions of the encoded data is increased, that is, the overhead ratio is switched from decrease to increase. By encoding the image data so as to have the data size of the inflection point, the overhead ratio can be appropriately reduced. Selection of the inflection point is performed by the created mapping table.

  First, it is determined whether or not the reference data size matches the data size of any inflection point (S22). If they match (S22: YES), the inflection point is selected (S23), and the process proceeds to S27.

  If the reference data size does not match the data size of any inflection point (S22: NO), one of the two inflection points whose data size is close to the reference data size is selected. Specifically, of the first inflection point that is the smallest inflection point among the inflection points whose data size is larger than the reference data size, and among the inflection points whose data size is smaller than the reference data size One of the second inflection points, which is the inflection point having the largest data size, is selected. Specifically, an intermediate value between the data size of the first inflection point and the data size of the second inflection point is set as a threshold, and it is determined whether or not the reference data size is larger than the threshold (S24). If it is equal to or smaller than the threshold (S24: NO), the second inflection point whose data size is smaller than the reference data size is selected (S25). If the reference data size is larger than the threshold (S24: YES), the first inflection point whose data size is larger than the reference data size is selected (S26).

  Next, the data size of the selected inflection point is set as a setting value for the data size of the encoded data (S27). The set value indicates the data size of the encoded data whose overhead ratio is the target value. Next, from the difference between the reference data size and the set value, a quantization parameter for which the data size of the encoded data becomes the set value is calculated (S28). Image data for one frame is encoded again with the calculated quantization parameter (S29). The original encoded data generated in S8 (see FIG. 9) is deleted (S30), and the process returns to the encoding process.

  Returning to the description of FIG. When the encoded data is generated (S8 or S11), the generated encoded data is divided so as to fit within the MTU size of the output destination device. Additional data is added to the divided data (divided packet) and output to the network 8 (S12). The data size of the divided packet is a size obtained by subtracting the data size of the additional data from the MTU size of the output destination device. For example, if the MTU size is 1500 bytes and the data size of the additional data is 20 bytes, the encoded data is divided by 1480 bytes. As a result, the data size of the transmission packet composed of the divided packet and the additional data becomes equal to the MTU size.

  Next, processing related to the coefficient (G) is performed (S13). For example, when the user has set the priority of the device using the available bandwidth of the network 8, the available bandwidth is measured. A coefficient (G) is calculated from the measured available bandwidth and stored in the available bandwidth storage area 122 (see FIG. 3). If the user gives an instruction to change the setting priority value, the changed setting priority value and the coefficient (G) are stored in the setting priority value storage area 131 (see FIG. 4). Next, it is determined whether or not there is an end instruction (S14). If there is no termination instruction (S14: NO), the process returns to the determination of S3. If there is an instruction to end (S14: YES), the encoding process ends.

  As described above, the video conference apparatus 1 according to the present embodiment measures the MTU size of the network 8 between the encoded data output destination devices. When the encoded data is divided so as to be within the measured MTU size, the data size of the encoded data whose overhead ratio is the target value is set as a set value. The overhead ratio is a ratio of the data size of the additional data to the data size of the entire data to be output. The additional data is data newly added to the divided data in order to perform network communication. The video conference device 1 encodes the image data to the data size set as the set value. The encoded data is divided in accordance with the MTU size of the output destination device and output to the network 8. Therefore, since the video conference apparatus 1 can encode the image data so that the overhead ratio becomes the target value, the overhead ratio can be appropriately reduced. Therefore, the efficiency of processing such as decoding of encoded data can be improved.

  When the data size of the encoded data is increased, the video conference device 1 can set the data size of the inflection point at which the overhead ratio switches from decrease to increase as the data size setting value. Therefore, the set value can be set appropriately and easily. Specifically, the video conference apparatus 1 sets the data size of the encoded data generated without setting the setting value as the reference data size. One of two inflection points having a data size close to the reference data size is selected as a target point. The image data is encoded using the data size of the selected target point as a set value. Therefore, the overhead ratio can be reduced while generating encoded data having a data size close to the reference data size. Therefore, it is possible to improve the efficiency of processing such as decoding while preventing deterioration in video quality due to a significant change in data size. The video conference apparatus 1 can set a threshold value between two inflection points close to the reference data size, and can easily set a setting value using the threshold value.

  As a result of measuring the MTU size of the network 8 between a plurality of devices, the video conference apparatus 1 creates one mapping table using the plurality of MTU sizes when a plurality of MTU sizes having different values are measured. . One set value common to a plurality of devices is set using one created mapping table. The same encoded data encoded to the data size of the set value is divided in accordance with the MTU size of each output destination device and output to each of a plurality of devices. Therefore, even when encoded data is output to a plurality of devices having different MTU sizes, the overhead ratio can be appropriately reduced by a series of processes.

  The video conference apparatus 1 can determine a coefficient (G) indicating the priority among a plurality of devices, and can create one mapping table using the coefficient (G). Therefore, it is possible to reduce the processing load of a high-priority device with priority by a series of encoding processes. Note that the video conference apparatus 1 can appropriately determine the coefficient (G) using at least one of the band information of the network 8, the MTU size, and the setting priority value set by the user.

  In the above embodiment, the video conference device 1 corresponds to the “encoding device” of the present invention. The camera 33 corresponds to “image input means”. The CPU 10 that measures the MTU size in S2 of FIG. 9 functions as a “measurement unit”. In S4 to S11 in FIG. 9 and S21 to S27 in FIG. 12, the CPU 10 that sets the data size at which the overhead ratio becomes the target value as the set value functions as a “setting unit”. The CPU 10 that calculates and encodes the quantization parameter in S28 and S29 in FIG. 12 functions as an “encoding unit”. The CPU 10 that divides and outputs the encoded data in S12 of FIG. 9 functions as an “output unit”.

  The CPU 10 that calculates the reference data size in S9 of FIG. 9 functions as a “calculation unit”. CPU10 which determines the priority of each apparatus by calculating a coefficient (G) by S30 of FIG. 12, etc. functions as a "priority determination means".

  The process of measuring the MTU size in S2 of FIG. 9 corresponds to a “measurement step”. The process of setting the data size at which the overhead ratio is the target value in S4 to S11 of FIG. 9 and S21 to S27 of FIG. 12 corresponds to a “setting step”. The process of calculating and encoding the quantization parameter in S28 and S29 in FIG. 12 corresponds to the “encoding step”. The process of dividing and outputting the encoded data in S12 of FIG. 9 corresponds to an “output step”.

  Needless to say, the present invention is not limited to the above-described embodiment, and various modifications are possible. For example, it is needless to say that the present invention is not limited to a video conference apparatus. For example, the present invention can be applied to any device that outputs encoded image data via a network, such as a server that distributes video. In the above embodiment, the H.264 standard. Coding is performed based on the H.264 standard, but other standards may be adopted.

  In the above embodiment, an Ethernet (registered trademark) LAN environment is adopted, and the IP header is the additional data. However, the type of additional data varies depending on the network environment employed. For example, a UDP header, an RTP header, or the like may be additional data. Even in such a case, the present invention can be applied.

  In the above embodiment, the single NAL unit mode is adopted as the packetization mode defined for packet transmission. Therefore, processing such as adjustment of the data size of encoded data is performed in units of NAL units. However, the present invention can be applied even when the non-interleave mode or the interleave mode is employed. In this case, the data size may be adjusted in units of a plurality of NAL units that are transmission units.

  In the above embodiment, the mapping table is created based on the measured MTU size after the processing is started. However, a plurality of mapping tables corresponding to the MTU size may be stored in advance in the HDD 13 or the like, and the setting value may be set using a mapping table having the closest MTU size that matches or is close to the measured MTU size.

  The video conference apparatus 1 according to the above embodiment sets a threshold value between two inflection points, and selects a target point depending on whether or not the reference data size is larger than the threshold value. However, the method for selecting the target point can also be changed. For example, an inflection point (second inflection point) having a data size smaller than the reference data size may be uniformly selected as the target point. Similarly, the second inflection point may be selected without fail. It is also possible to change the position of the threshold according to the size of the reference data size.

  In the above embodiment, when a plurality of MTU sizes are measured for a plurality of devices, one set value common to the plurality of devices is set. However, different set values may be set for each device, and encoding may be performed with different quantization parameters for each device to generate a plurality of encoded data from the same image data. The processing in this case is only performed for each device when the measured MTU size is one.

  The specific method of creating one mapping table using priority can also be changed. It goes without saying that the formula in the above embodiment for creating the mapping table is also an example, and is only one method for setting the setting value.

1 Video conference device 8 Network 10 CPU
12 RAM
13 HDD
33 Camera

Claims (9)

An encoding device that encodes image data input by an image input means and outputs the encoded data to another device via a network,
Measuring means for measuring the MTU size of the network between the device and the encoding device;
When the encoded data is divided by a size that fits within the MTU size measured by the measurement means, and additional data necessary for network communication is newly added to each divided data, the additional data for the entire data to be output is added. A setting means for setting the data size of the encoded data in which the overhead ratio, which is the ratio, becomes a target value,
Encoding means for encoding the image data to the data size set as the set value;
An encoding apparatus comprising: output means for dividing the encoded data encoded by the encoding means into a size that fits in the MTU size measured by the measuring means and outputting the divided data to the device.   2. The encoding according to claim 1, wherein the setting unit sets a data size at a point where the overhead ratio is switched from a decrease to an increase when the data size of the encoded data is increased. apparatus. A calculation unit that calculates a reference data size that is a data size of encoded data when the image data is encoded without setting the setting value;
When the data size of the point is different from the reference data size, the setting means is a first point that is a point having the smallest data size among a plurality of points, the data size of which is larger than the reference data size. And the data size of the target point is set to the set value, with any one of the second points being the largest data size among the points having a data size smaller than the reference data size as a target point The encoding apparatus according to claim 2.   The setting means sets the target point as either the first point or the second point depending on whether the reference data size is larger than a threshold value provided between the first point and the second point. 4. The encoding apparatus according to claim 3, wherein whether to perform the determination is determined. The setting means, when a plurality of MTU sizes having different values are measured as a result of measuring the MTU size of the network between the plurality of apparatuses by the measuring means, Based on the MTU size, one set value common to the plurality of devices is set,
The output means divides the same encoded data by a size that fits in each MTU size measured for the output destination device, and outputs the divided data to each of the plurality of devices. 5. The encoding device according to any one of 4 to 4. Priority determining means for determining the priority of each of the devices among the plurality of devices,
6. The encoding apparatus according to claim 5, wherein the setting unit sets the setting value common to the plurality of devices based on the priority determined by the priority determination unit.   The said priority determination means determines the said priority using at least any one of the network bandwidth information of each said apparatus, MTU size, and the setting information which shows the priority degree. The encoding device described. An encoding method performed by an encoding device that encodes image data input by an image input unit and outputs the encoded data to another device via a network,
A measuring step of measuring the MTU size of the network between the device and the encoding device;
When the encoded data is divided in a size that fits in the MTU size measured in the measurement step, and additional data necessary for network communication is newly added to each divided data, the additional data of the entire data to be output is added. A setting step for setting the data size of the encoded data in which the overhead ratio, which is the ratio, becomes the target value,
An encoding step for encoding the image data to the data size set as the set value;
An encoding method comprising: an output step of dividing the encoded data encoded by the encoding step into a size that fits in the MTU size measured by the measurement step and outputting the divided data to the device. An encoding program for encoding image data input by an image input means and outputting the encoded data to another device via a network,
On the computer,
A measuring step of measuring the MTU size of the network with the device;
When the encoded data is divided in a size that fits in the MTU size measured in the measurement step, and additional data necessary for network communication is newly added to each divided data, the additional data of the entire data to be output is added. A setting step for setting the data size of the encoded data in which the overhead ratio, which is the ratio, becomes the target value,
An encoding step for encoding the image data to the data size set as the set value;
An encoding program that executes the output step of dividing the encoded data encoded by the encoding step into a size that fits within the MTU size measured by the measurement step and outputting the divided data to the device.

Images