Data distribution system, communication quality prediction device, data transmission device, and data transmission method

Abstract

According to the present invention, a data distribution system includes a first prediction means that, on the basis of first sensor data, predicts the communication quality of a network used to transmit the first sensor data, a determination means that, in accordance with the communication quality predicted by the first prediction means, determines a parameter related to the transmission quality of the first sensor data, an encoding means that uses the parameter related to the transmission quality of the first sensor data to encode the first sensor data, and a transmission means that transmits the encoded first sensor data over the network.

Description

Technical Field

[0001] The present invention relates to a data distribution system, a communication quality prediction device, a data transmission device, and a data transmission method.

Background Art

[0002] Data captured by cameras mounted on moving bodies such as autonomous vehicles and drones, or wearable cameras worn by workers, is live-distributed to a remote control center or the like and used for monitoring operations and the like. Since the transmission of the captured data by these cameras passes through a wireless section, it is known that it is affected by the communication quality of the network.

[0003] Patent Document 1 discloses a communication quality adjustment system including an environment information acquisition unit that acquires environment information indicating the environment where a receiver is placed and that affects the communication state of the receiver, and that can receive video content with stable quality at the receiver.

[0004] Patent Document 2 discloses a configuration in an unmanned automatic driving system, drone control, and robot control for remotely managing devices, in which terminals mounted on these devices predict the communication quality between the terminals and an external communication device. Then, based on the predicted communication quality, these terminals perform control of improving the communication quality, avoiding a fatal decrease in communication quality, or satisfying the control conditions of the terminals with respect to the communication quality.

[0005] Patent Document 3 discloses a traffic control device that performs traffic control by deep reinforcement learning with a camera image as an input and can automatically adapt to various communication environments to effectively utilize a wireless band.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0007] The following analysis is provided by the inventor of the present invention. The transmission device for sensor data such as the above-described captured data moves together with a moving body or a worker. As a result of the movement, a sudden communication throughput variation may occur due to an obstacle intervening between the transmission device and the base station. In this regard, Patent Document 1 changes the communication state of the receiver and the reception form when the receiver receives content based on environmental information such as the position and speed of the receiver when the receiver is a mobile station, and cannot cope with sudden communication throughput variations caused by the movement of the transmission device.

[0008] Further, the terminal of Patent Document 2 is configured to predict future communication quality using the surrounding environment information of the terminal and perform control of its own device according to the future communication quality, thereby avoiding sudden communication throughput variations. Therefore, for example, when the control rules when the communication quality in FIG. 4 of the same document is the worst, such as limiting the speed to 5 km / h or less and stopping on the road shoulder, sudden communication throughput variations cannot be avoided.

[0009] Patent Document 3 only discloses a configuration in which a camera is used to photograph the communication environment between the first communication device and one or more second communication devices and predict the communication quality of each radio section to increase the total throughput.

[0010] An object of the present invention is to provide a data distribution system, a communication quality prediction device, a data transmission device, and a data transmission method that can cope with sudden communication throughput variations accompanying the movement of a transmission device for sensor data.

Means for Solving the Problems

[0011] According to a first aspect, based on first sensor data, there is provided a data distribution system including: a first prediction means for predicting the communication quality of a network used for transmitting the first sensor data; a determination means for determining a parameter related to the transmission quality of the first sensor data according to the communication quality predicted by the first prediction means; an encoding means for encoding the first sensor data using the parameter related to the transmission quality of the first sensor data; and a transmission means for transmitting the encoded first sensor data via the network.

[0012] According to a second aspect, there is provided a communication quality prediction apparatus including: a first prediction means for predicting the communication quality of a network used for transmitting first sensor data based on the first sensor data; and a transmission means for transmitting the predicted communication quality of the network used for transmitting the first sensor data to the device that is the source of the first sensor data. This communication quality prediction apparatus causes the device that is the source of the first sensor data to perform encoding of the first sensor data according to the predicted communication quality of the network used for transmitting the first sensor data and transmission of the encoded first sensor data.

[0013] According to a third aspect, there is provided a data transmission apparatus capable of receiving, from a communication quality prediction apparatus including: a first prediction means for predicting the communication quality of a network used for transmitting first sensor data based on the first sensor data; and a transmission means for transmitting the predicted communication quality of the network used for transmitting the first sensor data to the device that is the source of the first sensor data, the predicted communication quality of the network used for transmitting the first sensor data, and performing encoding of the first sensor data according to the communication quality and transmission of the encoded first sensor data.

[0014] According to a fourth aspect, based on the first sensor data, the communication quality of the network used for transmitting the first sensor data is predicted, and according to the predicted communication quality, a parameter related to the transmission quality of the first sensor data is determined. The first sensor data is encoded using the parameter related to the transmission quality of the first sensor data, and the encoded first sensor data is transmitted via the network. A data transmission method is provided. This method is associated with a specific machine, namely a computer, which predicts the communication quality of the network used for transmitting the sensor data based on the sensor data.

[0015] According to a fifth aspect, a program (computer program) for realizing the functions of the constituent devices of the above-described data distribution system is provided. This program is input into a computer device via an input device or from the outside through a communication interface, stored in a storage device, and causes a processor to be driven according to a predetermined step or process. Further, this program can display the processing result including intermediate states step by step via a display device as necessary, or communicate with the outside via a communication interface. A computer device for that purpose typically includes, as an example, a processor, a storage device, an input device, a communication interface, and a display device that can be connected to each other by a bus as necessary. Further, this program can be recorded on a computer-readable (non-transitory) storage medium.

Advantages of the Invention

[0016] According to the present invention, there are provided a data distribution system, a communication quality prediction device, a data transmission device, and a data transmission method that can cope with sudden communication throughput fluctuations accompanying the movement of a sensor data transmission device.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Modes for Carrying Out the Invention

[0018] First, an overview of an embodiment of the present invention will be described with reference to the drawings. Note that the reference numerals attached to the drawings in this overview are for convenience of each element as an example to assist understanding, and are not intended to limit the present invention to the illustrated embodiments. Also, the connection lines between blocks such as the drawings referred to in the following description include both bidirectional and unidirectional ones. The one-way arrow schematically shows the flow of the main signal (data) and does not exclude bidirectionality. The program is executed via a computer device, and the computer device includes, for example, a processor, a storage device, an input device, a communication interface, and a display device as necessary. Further, this computer device is configured to be communicable with devices inside or outside the device (including computers) via a communication interface, whether wired or wireless. Also, ports or interfaces are provided at the input / output connection points of each block in the figure, but the illustration thereof is omitted.

[0019] In one embodiment of the present invention, as shown in FIG. 1, it can be realized in a data distribution system 10 including a first prediction means 11, a determination means 12, an encoding means 13, and a transmission means 14.

[0020] More specifically, as shown in FIG. 2, the first prediction means 11 predicts the communication quality of the network used for transmitting the first sensor data based on the first sensor data (step S01). For example, when the first sensor data is image data captured by various sensors, the first prediction means 11 predicts the communication quality of the network based on the image data.

[0021] The determination means 12 determines a parameter related to the transmission quality of the first sensor data according to the communication quality predicted by the first prediction means (step S02). For example, when a prediction result that the communication quality of the network deteriorates is obtained, the determination means 12 changes to a value corresponding to the deterioration of the communication quality as a parameter related to the transmission quality of the first sensor data. For example, the determination means 12 changes the video bit rate lower than the reference value according to the communication quality.

[0022] The encoding means 13 encodes the first sensor data using parameters related to the transmission quality of the first sensor data (step S03).

[0023] The transmitting means 14 transmits the encoded first sensor data via the network (step S04). As described above, the parameters related to the transmission quality of the first sensor data are determined according to the communication quality of the network. Therefore, when a prediction result that the communication quality of the network deteriorates is obtained, the parameters related to the transmission quality of the first sensor data are determined to be values corresponding to the prediction result. Thus, it is possible to cope with sudden communication throughput fluctuations.

[0024] For example, when the sensor data is imaging data, there may be a case where an object that is expected to affect the communication quality of the network used for transmitting the imaging data appears in the imaging data. In this case, the first prediction means 11 predicts that the communication quality of the network used for transmitting the sensor data will decrease in the near future. The determination means 12 determines to change the parameters related to the transmission quality of the sensor data to contents suitable for the case where the communication quality of the network is low, based on the prediction that the communication quality of the network used for transmitting the sensor data will decrease. As the parameters related to the transmission quality of the sensor data, for example, the value of the video bit rate (hereinafter simply referred to as "bit rate") can be used. Then, the encoding means 13 encodes the sensor data using the parameters related to the transmission quality of the sensor data. By doing so, when the sensor data is imaging data, it is possible to avoid deterioration of the imaging data and abnormalities during reproduction due to the influence of an object that hinders the transmission of the imaging data.

[0025] As shown in FIG. 3, the above-described data distribution system 10 can be realized by using a data transmission device 10c that acquires and transmits first sensor data. According to this configuration, for example, when the sensor data is shooting data (video), there is an advantage that the communication quality can be predicted using high-quality video before encoding and transmission.

[0026] In addition to the form in which each of the above-described processing means is provided in a single device as shown in FIG. 3, a configuration in which each processing means of the above-described data distribution system 10 is distributed among a plurality of devices can be adopted. For example, as shown in FIG. 4, a configuration in which the first prediction means 11 of the data distribution system 10 is provided in another device (communication quality prediction device 10b) can also be adopted. In this case, a determination means 12, an encoding means 13, and a transmission means 14 are arranged in the data transmission device 10a. Further, a dedicated server or an MEC (Multi-access Edge Computing, or Mobile Edge Computing) server can be used as the communication quality prediction device 10b, and a prediction algorithm with a large processing amount can be used. Also, a plurality of data transmission devices 10a may be connected to the communication quality prediction device 10b, and the predicted communication results may be sent to these plurality of data transmission devices 10a. In the example of FIG. 4, the determination means 12 is arranged in the data transmission device 10a, but the determination means 12 may be arranged on the communication quality prediction device 10b side. In this case, the communication quality prediction device 10b will transmit parameters related to the transmission quality of the shooting data to the data transmission device 10a instead of the predicted communication quality.

[0027] Further, as shown in FIG. 5, the communication quality prediction device 10b may be arranged on the data receiving device 80 side. In this case, the communication quality prediction device 10f receives the encoded sensor data from the data receiving device 80 and predicts the communication quality of the network 90. Then, the communication quality prediction device 10f provides the predicted communication quality to the data transmission device 10e. The communication quality prediction device 10f may receive the captured data from the communication devices constituting the network 90. In the example of FIG. 5, the determination means 12 is arranged in the data transmission device 10e, but the determination means 12 may be arranged on the communication quality prediction device 10f side. In this case, the communication quality prediction device 10f transmits a parameter related to the transmission quality of the sensor data to the data transmission device 10e instead of the predicted communication quality. Also, when the data receiving device 80 is arranged on a cloud platform that provides cloud computing services, the communication quality prediction device 10f may be arranged on the cloud platform.

[0028] [First Embodiment] Subsequently, a first embodiment in which the present invention is applied to a system for live distribution of video captured by a camera will be described in detail with reference to the drawings. FIG. 6 is a diagram showing the configuration of a captured data distribution system according to the first embodiment of the present invention. Referring to FIG. 6, a configuration is shown in which a video transmission device 200 and a video reception device 300 are connected via a network 90. Further, a communication quality prediction device 100 is provided in the captured data distribution system. The communication quality prediction device 100 can acquire the captured data from the video transmission device 200 and the communication quality when receiving the past captured data from the video reception device 300.

[0029] The video transmission device 200 is a device that transmits the captured data captured by the camera 201 to the video reception device 300. The video transmission device 200 includes a camera 201, an encoding control unit 202, and an encoding unit 203. This video transmission device 200 corresponds to the aforementioned data transmission devices 10a and 10e. The camera 201 is a camera that captures video to be live-streamed and is mounted on a moving body (corresponding to the first device). The encoding control unit 202 corresponds to the aforementioned determination means 12, and determines the bit rate when encoding the video based on the predicted communication quality value received from the communication quality prediction device 100. The encoding unit 203 corresponds to the aforementioned encoding means 13 and transmission means 14, encodes the video at the bit rate determined by the encoding control unit 202, and transmits it to the communication quality prediction device 100 and the video reception device 300 respectively. Note that the bit rate can be changed using at least one or more of the video resolution, frame rate, target bit rate, QP (Quantization Parameter), CRF (Constant Rate Factor), encoding method (CODEC type), etc. Also, when changing the bit rate, if the priority of the information element to be changed by the user has been determined in advance, the bit rate may be changed by changing the information element according to the priority. By doing so, it becomes possible to change the bit rate according to the user's request.

[0030] The video reception device 300 is a device that receives the captured data captured by the camera 201 via the video transmission device 200. The video reception device 300 includes a communication quality measurement unit 301, a decoding unit 302, and a playback unit 303. This video reception device 300 corresponds to the aforementioned data reception device 80. The communication quality measurement unit 301 measures the communication quality when receiving the video from the video transmission device 200 and transmits it to the communication quality prediction device 100. The decoding unit 302 decodes the video received from the video transmission device 200. The playback unit 303 plays back the video decoded by the decoding unit 302.

[0031] The communication quality prediction device 100 is a device that predicts the communication quality of a network used when the video transmission device 200 transmits data to the video reception device 300. The communication quality prediction device 100 includes a prediction unit 101 and a data acquisition unit 102. The data acquisition unit 102 receives video from the video transmission device 200 and provides it to the prediction unit 101. Also, the data acquisition unit 102 receives past communication quality from the video reception device 300 and provides it to the prediction unit 101. The prediction unit 101 predicts the future communication quality of the network 90 based on the past communication quality acquired from the video reception device 300 and the video received from the video transmission device 200. The method for predicting the future communication quality of the network 90 in the prediction unit 101 will be described in detail later together with the description of the operation of this embodiment.

[0032] Subsequently, the operation of this embodiment will be described in detail with reference to the drawings. FIG. 7 is a sequence diagram showing the operation of the shooting data distribution system according to the first embodiment of the present invention. Referring to FIG. 7, the communication quality prediction device 100 acquires video (shooting data) from the video transmission device 200 at a predetermined time interval (step S001). Note that this predetermined time interval is determined according to the system configuration, the network quality between the video transmission device 200 and the communication quality prediction device 100, and the performance of the communication quality prediction device 100. For example, when the video transmission device 200 and the communication quality prediction device 100 are executed on the same server, or when they are connected by a high-speed network, when the performance of the communication quality prediction device 100 is high, or when it is desired to improve the prediction accuracy of the prediction unit 101, the communication quality prediction device 100 may acquire all frames from the video transmission device 200. In this case, for example, when the video transmission device 200 transmits video (shooting data) at 30 fps, the predetermined time interval is 30 fps. Of course, the communication quality prediction device 100 may acquire the video (shooting data) of the video transmission device 200 by thinning it out. In this case, the predetermined time interval is 30 fps or less.

[0033] On the one hand, the video receiving device 300 measures the communication quality of the network 90 when receiving video from the video transmitting device 200 in the past (step S002). In the present embodiment, the video receiving device 300 measures the communication throughput as the communication quality and provides the time series data thereof to the communication quality prediction device 100 for explanation. The communication throughput can be calculated, for example, by dividing the size of each video frame by the time required for receiving the frame (the time from receiving the first packet to receiving the last packet).

[0034] The communication quality prediction device 100 predicts the communication throughput of the network 90 at a predetermined time in the future as the future communication quality by using the past communication throughput acquired from the video receiving device 300 and the video (shooting data) acquired from the video transmitting device 200 (step S003). The communication quality prediction device 100 transmits the predicted communication throughput to the video transmitting device 200.

[0035] Specifically, the communication quality prediction device 100 predicts the future communication throughput based on the time series data of the communication throughput (communication quality history data) acquired from the video receiving device 300. As a method for predicting the communication throughput, for example, a method of predicting the probability distribution of the time series data based on a pre-created prediction model can be used. Further, the communication quality prediction device 100 checks whether an event that affects the communication throughput of the network 90 at a predetermined time in the future has occurred based on the video (shooting data) acquired from the video transmitting device 200. As a result of the check, if it is determined that an event that affects the communication throughput at a predetermined time in the future has occurred, the communication quality prediction device 100 increases or decreases the predicted value of the communication throughput according to the content thereof.

[0036] The video transmission device 200 that has acquired the predicted communication throughput determines an appropriate bitrate based on the predicted communication throughput (step S004). Note that the bitrate setting may be realized by setting any one or more of the target bitrate, QP (Quantization Parameter), and CRF (Constant Rate Factor). Also, the resolution and frame rate may be increased or decreased. For example, when the communication throughput is below the lower limit of a predetermined range, the video transmission device 200 determines to lower either or both of the resolution and frame rate in addition to the bitrate of the video to be sent to the video reception device 300. When only the resolution is lowered, a smooth video with a high frame rate can be delivered even if the bitrate is lowered. When only the frame rate is lowered, the deterioration of the image quality of each video frame when the bitrate is lowered can be reduced. For example, when the communication throughput exceeds the upper limit of a predetermined range, the video transmission device 200 determines to increase the resolution or frame rate in addition to the bitrate of the video to be sent to the video reception device 300. Note that when increasing or decreasing the bitrate, a configuration for increasing or decreasing the bitrate can be adopted by increasing or decreasing any one or more of the resolution, frame rate, QP, and CRF according to a predetermined setting.

[0037] Note that the predetermined range for comparison with the communication throughput can be increased or decreased according to the currently adopted bitrate. For example, when the bitrate exceeds the communication throughput, packet loss occurs and appears as image quality distortion. Also, when the bitrate is much lower than the communication throughput, network resources are not being effectively utilized. Therefore, by setting a value considering the bitrate as the predetermined range and comparing it with the predicted communication throughput, prior countermeasures become possible.

[0038] The video transmission device 200 encodes the video at the determined bitrate and transmits it to the video reception device (step S005). The video reception device 300 decodes and plays back the received captured data (step S006).

[0039] The operation of this embodiment will be described more specifically with reference to FIGS. 8 and 9. In the following description, the video transmission device 200 is mounted on a vehicle (target vehicle) as an in-vehicle terminal, and transmits the video in front of the vehicle captured by the camera 201 to the monitoring center. The video reception device 300 is installed in the monitoring center, and decodes and plays back the captured data received from the vehicle (target vehicle). Also, the communication quality prediction device 100 is assumed to be mounted on the vehicle (target vehicle). Further, the wireless network between the vehicle (target vehicle) and the base station is a wireless network such as 5G (including local 5G), LTE (Long Term Evolution), or wireless LAN (Local Area Network), and communication is performed at a frequency at which the communication quality fluctuates due to the influence of obstacles such as millimeter waves.

[0040] The communication quality prediction device 100 mounted on the target vehicle predicts the future (a predetermined time ahead) communication throughput based on the time-series data of the past communication throughput acquired from the video reception device 300 of the monitoring center. Also, the communication quality prediction device 100 checks whether an event that affects the future communication throughput has occurred based on the video (captured data) acquired from the camera 201. For example, as shown in the upper figure (a) of FIG. 8, when the inter-vehicle distance between the target vehicle and the vehicle ahead is equal to or greater than a predetermined distance, line-of-sight communication (LOS (Line of Sight)) is ensured between the base station and the vehicle (target vehicle) as shown by the broken line in FIG. 8. In this case, the communication quality prediction device 100 determines that no event that affects the future communication throughput has occurred, predicts the future communication throughput from the past communication throughput, and notifies the video transmission device 200. The video transmission device 200 encodes the video at the bitrate determined based on the communication throughput predicted from the past communication throughput, and transmits it to the monitoring center.

[0041] On the one hand, as shown in the lower diagram (b) of FIG. 8, the target vehicle and the preceding vehicle are approaching, and there is a preceding vehicle between the base station and the vehicle (target vehicle), and non-line-of-sight (NLOS) communication may occur as shown by the dashed line in FIG. 8. In this case, the preceding vehicle is prominently reflected in the captured data. The communication quality prediction device 100 determines from such an image that an event affecting the future communication throughput has occurred, predicts the future communication throughput taking into account the influence of the preceding vehicle, and notifies the video transmission device 200. In this case, the video transmission device 200 encodes the video at the bit rate determined based on the communication throughput underestimated due to the presence of the preceding vehicle and transmits it to the monitoring center. Thereby, even in a situation where the distance to the preceding vehicle decreases and the communication throughput drops sharply, it is possible to continue stable live video distribution.

[0042] In the above example, it has been described that it is determined that an event affecting the future communication throughput has occurred when the preceding vehicle is prominently reflected in the captured data. However, the method for determining the occurrence of an event affecting the communication throughput is not limited to this method. For example, when the preceding vehicle shown in the temporally continuous captured data is getting larger, it can be determined that the target vehicle is approaching the preceding vehicle. Based on this approach (approach speed), it is also possible to estimate the timing when the communication throughput is affected. Also, when the state where the preceding vehicle is prominently reflected continues in the captured data for a predetermined period, it can be determined that the target vehicle is maintaining a state of approaching the preceding vehicle.

[0043] In addition, the events that affect the future communication throughput described above are not limited to the approach of the vehicle (target vehicle) to the vehicle ahead. For example, as shown in FIGS. 9(a) and 9(b), when the vehicle (target vehicle) approaches a tunnel or the like, similarly, the bit rate can be determined based on the underestimated communication throughput. Of course, depending on the tunnel, there may be a tunnel where the communication throughput does not decrease due to reasons such as a base station being installed inside. In that case, the communication quality prediction device 100 may identify a tunnel where the communication throughput does not decrease from the appearance, position, etc. of the tunnel. Of course, after predicting that the communication throughput will decrease once, a method of restoring the bit rate based on the actual communication throughput can also be adopted.

[0044] Also, from the states of FIGS. 8(b) and 9(b), there may be a case where the future communication throughput is greatly improved due to events such as the vehicle-to-vehicle distance increasing or passing through a tunnel. In this case, the communication quality prediction device 100 predicts the future communication throughput considering these events and notifies the video transmission device 200. Then, the video transmission device 200 encodes the video at the bit rate determined based on the overestimated communication throughput and transmits it to the monitoring center. As a result, after the improvement of the communication throughput, it becomes possible to resume the delivery of high-quality live video promptly.

[0045] As described above, according to the present embodiment, while basing on the future communication throughput predicted based on the past communication throughput, when an event that affects the future communication throughput is foreseen from the video, it is possible to adjust the communication throughput.

[0046] [Second Embodiment] Next, a second embodiment in which the configuration of the prediction unit of the communication quality prediction device is changed will be described in detail with reference to the drawings. FIG. 10 is a diagram showing the configuration of a shooting data distribution system according to the second embodiment of the present invention. In the shooting data distribution system according to the second embodiment, the configuration of the prediction unit 101a in the communication quality prediction device 100a is different from that of the first embodiment. Since the other configurations are the same as those of the first embodiment shown in FIG. 6, the differences will be mainly described below.

[0047] FIG. 11 is a diagram showing the configuration of a communication quality prediction device according to the second embodiment of the present invention. Referring to FIG. 11, the configuration of a communication quality prediction device 100a including a prediction unit 101a and a data acquisition unit 102 is shown.

[0048] The data acquisition unit 102 includes a video acquisition unit 1021 and a communication quality acquisition unit 1022. The video acquisition unit 1021 acquires video (shooting data) from the video transmission device 200 and provides it to the prediction unit 101a. The communication quality acquisition unit 1022 acquires the past communication quality of the network 90 from the video reception device 300 and provides it to the prediction unit 101a.

[0049] The prediction unit 101a includes a first predictor 1011, a second predictor 1012, and an integration unit 1013. In the present embodiment, the first predictor 1011 functions as the first prediction means, and the second predictor 1012 functions as the second prediction means. Specifically, the first predictor 1011 predicts the future communication quality based on the video (shooting data) sent from the video transmission device 200 and outputs it to the integration unit 1013. Such a first predictor 1011 can be configured, for example, using a machine learning model that takes the video (shooting data) sent from the video transmission device 200 as an input and outputs a predicted value. Also, for example, the first predictor 1011 can be configured using a model that identifies the objects shown in the video (shooting data) sent from the video transmission device 200, calculates the degree to which the size and distance affect the communication throughput, and outputs a predicted value.

[0050] The second predictor 1012 predicts future communication quality based on the past communication quality (time-series data of communication throughput) of the network 90 sent from the video receiving device 300, and outputs it to the integration unit 1013. The integration unit 1013 integrates the predicted communication qualities of the first predictor 1011 and the second predictor 1012 according to a predetermined rule, and outputs a predicted value of the future communication quality. The predetermined rule may be, for example, a rule that outputs, as a predicted value, the result of weighting the predicted value of the first predictor 1011 and the predicted value of the second predictor 1012 by a predetermined formula. Also, the predetermined rule may be a rule that compares the predicted value of the first predictor 1011 and the predicted value of the second predictor 1012, adopts the lower one, and outputs it as the predicted value.

[0051] Also, the integration process in the integration unit 1013 may adopt either the predicted value of the first predictor 1011 or the predicted value of the second predictor 1012 based on the change amount of the first predictor 1011 as follows.

[0052] [Sudden change 1 in the predicted value of the first predictor 1011] FIG. 12 is a diagram for explaining another example of the integration process of the integration unit 1013. The black circles in FIG. 12 indicate the output of the first predictor (communication quality predicted based on the video). For example, as shown in FIG. 12, when a decrease of a specified amount or more occurs in the output of the first predictor 1011, the integration unit 1013 adopts the output of the first predictor 1011. On the other hand, when a decrease of a specified amount or more does not occur in the output of the first predictor 1011, the integration unit 1013 adopts the output of the second predictor 1012. By adopting such an integration process, it is possible to appropriately respond to a decrease in communication quality caused by an event that cannot be predicted from the transition of communication throughput (for example, approaching the vehicle ahead, passing through a tunnel).

[0053] Also, for example, as shown in FIG. 13, even when an increase equal to or greater than a specified amount occurs in the output of the first predictor 1011, the integration unit 1013 may adopt the output of the first predictor 1011. On the other hand, when an increase equal to or greater than the specified amount does not occur in the output of the first predictor 1011, the integration unit 1013 adopts the output of the second predictor 1012. By adopting such integration processing, it becomes possible to appropriately respond to a rapid recovery of the communication throughput by returning from the states shown in FIGS. 8(b) and 9(b) (for example, approaching the vehicle ahead, traveling in a tunnel) to the states shown in FIGS. 8(a) and 9(a) (for example, ensuring a distance from the vehicle ahead, traveling outside the tunnel).

[0054] [Sudden change in the predicted value of the first predictor 1011 2] FIG. 14 is a diagram for explaining another example of the integration processing of the integration unit 1013. In the example of FIG. 14, when a decrease equal to or greater than a specified amount occurs from the lowest value of the output of the first predictor 1011 over a certain past period, the integration unit 1013 adopts the output of the first predictor 1011. On the other hand, when the output of the first predictor 1011 does not satisfy the above conditions, the integration unit 1013 adopts the output of the second predictor 1012. By adopting such integration processing as well, it is possible to appropriately respond to a decrease in communication quality caused by an event that cannot be predicted from the transition of the communication throughput.

[0055] Also, for example, as shown in FIG. 15, even when an increase equal to or greater than a specified amount occurs from the highest value of the output of the first predictor 1011 over a certain past period, the integration unit 1013 may adopt the output of the first predictor 1011. On the other hand, when the output of the first predictor 1011 does not satisfy the above conditions, the integration unit 1013 adopts the output of the second predictor 1012. By adopting such integration processing as well, it becomes possible to appropriately respond to a rapid recovery of the communication throughput by returning from the states shown in FIGS. 8(b) and 9(b) (for example, approaching the vehicle ahead, traveling in a tunnel) to the states shown in FIGS. 8(a) and 9(a) (for example, ensuring a distance from the vehicle ahead, traveling outside the tunnel).

[0056] Note that the above-mentioned "specified amount" can be changed according to the performance of the camera 201, the content of the live distribution image, and the type of monitoring operation by the monitoring center. For example, when the performance of the camera 201 is low, changes in the output of the first predictor 1011 are likely to occur due to noise or deterioration of shooting conditions. In that case, by setting the "specified amount" that functions as a threshold to a value larger than the reference, false determination due to noise can be avoided. For example, when the performance of the camera 201 is low, by setting a value larger than the standard value as the "specified amount", it becomes difficult for the output of the first predictor 1011 to be adopted. Thereby, false determination due to the performance of the camera 201, weather, etc. can be avoided. Also, when continuous video distribution is required rather than image quality, the "specified amount" may be set to a value smaller than the reference. Thereby, it becomes possible to enhance the tolerance to sudden changes in throughput due to sudden events. For example, by setting a value smaller than the standard value as the "specified amount", it becomes easier for the output of the first predictor 1011 to be adopted. Thereby, it becomes possible to perform stable video distribution even in a situation where sudden events occur frequently.

[0057] Also, in the example shown in FIGS. 12 to 15, although it has been described as comparing the latest output of the first predictor 1011 with the past output of the first predictor 1011, the latest output of the first predictor 1011 may be compared with the actual communication throughput sent from the video receiving device 300.

[0058] Also, in the above-described embodiment, the integration unit 1013 has been described as determining whether to adopt the predicted value of the second predictor 1012 based on the change in the output of the first predictor 1011, but the integration process of the predicted value in the integration unit 1013 is not limited to this. For example, the integration unit 1013 may compare the predicted value of the first predictor 1011 with the predicted value of the second predictor 1012 and adopt the lower predicted value.

[0059] [Third Embodiment] Next, a third embodiment in which the configuration of the prediction unit of the communication quality prediction device is changed will be described in detail with reference to the drawings. FIG. 16 is a diagram showing the configuration of a shooting data distribution system according to the third embodiment of the present invention. In the shooting data distribution system according to the third embodiment, the configuration of the prediction unit 101b in the communication quality prediction device 100b is different from that of the second embodiment. Since the other configurations are the same as those of the second embodiment shown in FIGS. 10 and 11, the differences will be mainly described below.

[0060] FIG. 17 is a diagram showing the configuration of an integration unit 1013b in the prediction unit 101b of the communication quality prediction device 100b according to the third embodiment of the present invention. Referring to FIG. 17, the output of the first predictor 1011 and the output of the second predictor 1012 are input to the integration unit 1013b. Using these as inputs, the integration unit 1013b outputs a prediction value with a higher likelihood.

[0061] Such an integration unit 1013b can be configured by a vector autoregressive model (VAR model; Vector AutoRegression), a neural network, or the like. For example, in the case of a vector autoregressive model, a model can be created by identifying parameters from the time series data of the output of the first predictor 1011 and the time series data of the output of the second predictor 1012. In the case of a neural network, an RNN (Recurrent Neural Network) or the like is used, and a model is created by learning using the time series data of the output of the first predictor 1011, the time series data of the output of the second predictor 1012, and teacher data labeled with the actual communication throughput. The above-described method for configuring the integration unit 1013b is merely an example, and statistical models and machine learning models can be used.

[0062] Also, in the example of FIG. 17, one of the inputs to the integration unit 1013b is the output of the second predictor 1012. However, instead of the output of the second predictor 1012, the past communication quality of the network 90 sent from the video receiving device 300 itself may be used. In this case, the second predictor 1012 can also be omitted.

[0063] Also, as shown in FIG. 18, past communication quality of the network 90 sent from the video receiving device 300 and position information of the camera 201 and the video transmitting device 200 may be input to the integration unit 1013b, and a predicted value considering these may be output. For example, by inputting the past communication quality to the integration unit 1013b, it is also possible to further improve the predicted value of the second predictor 1012 in the integration unit 1013b. Also, by inputting the position information to the integration unit 1013b, it is possible to output a predicted value considering the position. By doing so, for example, even when a tunnel is shown in the video in the same way, depending on the past communication quality and position information, a tunnel where the communication throughput drops and a tunnel where the communication throughput does not drop can be distinguished, and communication control according to each location becomes possible.

[0064] According to the present embodiment, as in the second embodiment, it is possible to appropriately respond to the sharp decrease in the communication throughput shown in FIGS. 8(b) and 9(b) and the subsequent recovery of the communication throughput.

[0065] [Fourth Embodiment] Subsequently, a fourth embodiment in which the input of the communication quality prediction device is changed will be described in detail with reference to the drawings. FIG. 19 is a diagram showing the configuration of a shooting data distribution system according to the fourth embodiment of the present invention. In the shooting data distribution system according to the fourth embodiment, the communication quality prediction device 100c receives video (shooting data) directly from the camera 201, rather than the video (shooting data) encoded by the video transmitting device 200. Other configurations are the same as those of the first to third embodiments, and similarly, the configuration of the communication quality prediction device 100c can adopt the first to third embodiments.

[0066] According to this embodiment, it is possible to predict communication quality using high-quality video before encoding. Since this embodiment deals with video (shooting data) with a large size before encoding, it can be preferably adopted in a form where the communication quality prediction device 100c is directly connected to the camera 201 via a cable or the like. For example, this embodiment can also be preferably adopted when the communication quality prediction device 100c is mounted on a moving body together with the video transmission device 200.

[0067] [Fifth Embodiment] Subsequently, a fifth embodiment in which a plurality of cameras connected to the communication quality prediction device 100d are provided will be described in detail with reference to the drawings. FIG. 20 is a diagram showing the configuration of a shooting data distribution system according to the fifth embodiment of the present invention. In the shooting data distribution system according to the fifth embodiment, in addition to the camera 201 of the video transmission device 200, a second camera 401 is connected to the communication quality prediction device 100d. Since the other configurations are the same as those of the first to fourth embodiments, the differences will be mainly described below.

[0068] Therefore, in this embodiment, the data acquisition unit 102d receives videos from the video transmission device 200 and the second camera 401, respectively, and provides them to the prediction unit 101d. Here, the second camera 401 corresponds to the second device, and the video (shooting data) received from the second camera 401 corresponds to the second sensor data.

[0069] The prediction unit 101d predicts the future communication quality of the network 90 based on the past communication quality acquired from the video reception device 300, the video received from the video transmission device 200, and the video of the second camera.

[0070] The operation of this embodiment will be described in detail with reference to the drawings. FIG. 21 is a diagram for explaining the operation of the shooting data distribution system according to the fifth embodiment of the present invention. In the example of FIG. 21, the video transmission device 200 is mounted on a vehicle (target vehicle) as an in-vehicle terminal, and transmits the video of the front of the vehicle captured by the camera 201 to the monitoring center. It is assumed that the communication quality prediction device 100d is arranged on the base station side. Also, it is assumed that communication between the vehicle (target vehicle) and the base station is performed at a frequency that is easily affected by obstacles such as millimeter waves. In the example of FIG. 21, a camera installed near the traffic signal at the intersection transmits video (shooting data) to the communication quality prediction device 100d as the second camera 401.

[0071] The communication quality prediction device 100d arranged on the base station side predicts the future communication throughput based on the time-series data of the past communication throughput acquired from the video reception device 300 of the monitoring center. Also, the communication quality prediction device 100d checks whether an event that affects the future communication throughput has occurred based on the video (shooting data) acquired from the cameras 201 and 401. For example, as shown in FIG. 21, although there are no preceding vehicles or obstacles on the road where the target vehicle is traveling, if a large vehicle is approaching the intersection from the intersecting road, the second camera 401 can capture this large vehicle. In this case, the communication quality prediction device 100d determines from the video of the second camera 401 that an event that affects the future communication throughput has occurred, predicts the future communication throughput taking into account the influence of the vehicle ahead, and notifies the vehicle (target vehicle). In this case, the video transmission device 200 mounted on the vehicle (target vehicle) encodes the video at the bitrate determined based on the communication throughput estimated for the crossing of the large vehicle, and transmits it to the monitoring center. As a result, it becomes possible to continue the stable live video distribution even under the situation where the communication throughput has decreased due to the crossing of the large vehicle.

[0072] In this embodiment, the number of the second cameras 401 that provide videos to the communication quality prediction device 100d is not limited to one. For example, as shown in FIG. 22, a plurality of second cameras 401 may be arranged. In the example of FIG. 22, a plurality of second cameras 401 are installed, enabling vehicles entering the intersection from a high position to be grasped. As a result, not only large vehicles entering the intersection from the intersection direction but also oncoming vehicles can be grasped early, and it becomes possible to predict future communication throughput taking their influence into account.

[0073] As described above, according to this embodiment that uses the video of the second camera 401, it is possible to further improve the prediction accuracy of future communication throughput. In the examples of FIGS. 21 and 22, an example where the communication quality prediction device 100d is arranged on the base station side has been described, but the communication quality prediction device 100d may be arranged on the monitoring center side. The arrangement and configuration of each device in this case are the same as those shown in FIG. 5. Further, the communication quality prediction device 100d may be mounted on the target vehicle. The arrangement and configuration of each device in this case are the same as those shown in FIG. 4.

[0074] Also, in the above-described embodiment, the communication quality prediction device 100d has been described as predicting future communication quality using both the videos of the camera 201 and the second camera 401, but a configuration that predicts future communication quality using only the video of the second camera 401 may also be used. For example, as shown in FIGS. 21 and 22, when an aerial view video can be obtained from the second camera 401, the input of the shooting data of the camera 201 to the communication quality prediction device 100d can be omitted.

[0075] [Sixth Embodiment] The present invention is also applicable to applications such as monitoring using live images from cameras installed at construction sites, inside factories, etc. in addition to transmitting live videos of the video of the camera 201 mounted on a vehicle (target vehicle).

[0076] FIG. 23 is a diagram for explaining the operation of the shooting data distribution system according to the sixth embodiment of the present invention. In FIG. 23, the camera 201 is a camera installed at a construction site, inside a factory, or the like. As shown in the upper diagram (a) of FIG. 23, line-of-sight communication (LOS) is normally ensured between a video transmission device (not shown) to which the camera 201 is connected and the base station, as shown by the broken line in FIG. 23(a). In this case, the communication quality prediction device 100e determines that no event affecting the future communication throughput has occurred, predicts the future communication throughput from the past communication throughput, and notifies the video transmission device (not shown). In this case, the video transmission device (not shown) encodes the video at the bit rate determined based on the communication throughput predicted from the past communication throughput and transmits it to the monitoring center.

[0077] On the other hand, as shown in the lower diagram (b) of FIG. 23, there may be a case where a construction vehicle crosses and non-line-of-sight communication (NLOS) occurs, as shown by the broken line in FIG. 23(b). In this case, the construction vehicle is reflected in the shooting data of the camera 201. The communication quality prediction device 100e determines from such a video that an event affecting the future communication throughput has occurred, predicts the future communication throughput taking into account the influence of the construction vehicle, and notifies the video transmission device (not shown). In this case, the video transmission device (not shown) encodes the video at the bit rate determined based on the communication throughput estimated considering the influence of the crossing of the construction vehicle and transmits it to the monitoring center. Thereby, it becomes possible to continue stable live video distribution even in a situation where the communication throughput drops sharply due to the crossing of the construction vehicle.

[0078] Further, the camera 201 may be a wearable camera worn on a helmet or work clothes of a worker or the like, rather than a fixed camera. In Fig. 24, the camera 201 is a camera worn on the worker's helmet. As shown in the upper diagram (a) of Fig. 24, line-of-sight communication (LOS) is usually ensured between the video transmission device (not shown) to which the camera 201 is connected and the base station, as indicated by the dashed line in Fig. 24(a). In this case, the communication quality prediction device 100e determines that no event affecting the future communication throughput has occurred, predicts the future communication throughput from the past communication throughput, and notifies the video transmission device (not shown). In this case, the video transmission device (not shown) encodes the video at the bit rate determined based on the communication throughput predicted from the past communication throughput and transmits it to the monitoring center.

[0079] On the other hand, as shown in the lower diagram (b) of Fig. 24, when the worker moves inside the building, it becomes non-line-of-sight communication (NLOS), as indicated by the dashed line in Fig. 24(b). In this case, it is possible to grasp from the shooting data of the camera 201 that the worker has moved inside the building. The communication quality prediction device 100e determines from such video that an event affecting the future communication throughput has occurred, predicts the future communication throughput taking into account the influence of the worker's movement, and notifies the video transmission device (not shown). In this case, the video transmission device (not shown) encodes the video at the bit rate determined based on the communication throughput estimated for the influence of the worker's movement into the building and transmits it to the monitoring center. Thereby, it becomes possible to continue the stable live video distribution even in a situation where the worker actually moves inside the building and the communication throughput drops sharply.

[0080] Alternatively, the camera 201 may be a camera mounted on a construction vehicle or the like. As shown in the upper diagram (a) of FIG. 25, a line-of-sight communication (LOS) is usually ensured between the video transmission device (not shown) to which the camera 201 is connected and the base station, as indicated by the dashed line in FIG. 25(a). In this case, the communication quality prediction device 100e determines that no event affecting the future communication throughput has occurred, predicts the future communication throughput from the past communication throughput, and notifies the video transmission device (not shown). In this case, the video transmission device (not shown) encodes the video at the bitrate determined based on the communication throughput predicted from the past communication throughput and transmits it to the monitoring center.

[0081] On the other hand, as shown in the lower diagram (b) of FIG. 25, when the construction vehicle equipped with the video transmission device (not shown) moves away from the base station, the communication quality deteriorates. In this case, it is possible to grasp from the captured data of the camera 201 that the construction vehicle is moving in the direction away from the base station. The communication quality prediction device 100e determines from such video that an event affecting the future communication throughput has occurred, predicts the future communication throughput taking into account the influence of the movement of the construction vehicle, and notifies the video transmission device (not shown). In this case, the video transmission device (not shown) encodes the video at the bitrate determined based on the communication throughput estimated for the influence of the movement of the construction vehicle and transmits it to the monitoring center. Thereby, it becomes possible to continue stable live video distribution even in a situation where the construction vehicle actually moves away from the base station and the communication throughput decreases.

[0082] Also, in any of the cases of FIGS. 23 to 25, the communication throughput is expected to recover due to the end of the crossing of the construction vehicle, the movement of the worker outside the building, and the approach of the construction vehicle to the base station. Also in these cases, similar to the first to fifth embodiments, the communication quality prediction device 100e can detect these events based on the captured data of the camera 201, predict that the future communication throughput will increase, and notify the video transmission device (not shown). Thereby, it is also possible to quickly increase the bitrate that has once decreased and improve the image quality of the live video.

[0083] As described above, each embodiment of the present invention has been explained. However, the present invention is not limited to the above-described embodiments, and further modifications, substitutions, and adjustments can be made without departing from the basic technical idea of the present invention. For example, the system configuration shown in each drawing, the configuration of each element, the arrangement of devices, etc. are examples for helping the understanding of the present invention, and are not limited to the configurations shown in these drawings.

[0084] For example, in each of the above-described embodiments, the video transmission device 200 has been described as determining the bit rate when encoding video according to the communication quality. However, a configuration that adjusts parameters related to the quality of shooting data other than the bit rate can also be adopted. Examples of such parameters include resolution (size), gradation, frame rate, color gamut, luminance dynamic range, etc.

[0085] Also, the camera 201 in each of the above-described embodiments has been described as a visible light camera that shoots in front of the camera, but it is not limited to this. For example, a 360-degree camera with an unlimited shooting range, a camera that can acquire depth in addition to video (Depth camera), an infrared camera may also be used. Furthermore, LiDAR (Laser Detection and Ranging) may also be used.

[0086] In addition, in each of the above-described embodiments, the communication quality prediction apparatus has been described as predicting the communication throughput as the communication quality of the network used for transmitting the first sensor data. However, information other than the communication throughput can also be used as the communication quality. Examples of such communication quality include information related to signal quality such as RSRP (Reference Signal Received Power), RSRQ (Reference Signal Received Quality), RSSI (Received Signal Strength Indicator), and SINR (Signal-to-Interference-plus-Noise Ratio). In addition to the information related to signal quality, a configuration in which the communication quality prediction apparatus predicts a parameter controlled according to signal quality such as MCS (Modulation and Coding Scheme) can also be adopted.

[0087] In addition, the procedures shown in the first to sixth embodiments described above can be realized by a program that causes a computer (9000 in FIG. 26) that functions as the communication quality prediction apparatuses 100 to 100e to realize the functions of the communication quality prediction apparatuses 100 to 100e. Such a computer is exemplified by a configuration including a CPU (Central Processing Unit) 9010, a communication interface 9020, a memory 9030, and an auxiliary storage device 9040 in FIG. 26. That is, the CPU 9010 in FIG. 26 may execute a data acquisition program and a communication quality prediction program, and perform an update process of each calculation parameter held in the auxiliary storage interface.

[0088] That is, each part (processing means, function) of the communication quality prediction apparatuses 100 to 100e shown in each of the above-described embodiments can be realized by a computer program that causes a processor mounted on these apparatuses to execute each of the above-described processes using its hardware.

[0089] Finally, the preferred forms of the present invention will be summarized. [First Form] (Refer to the data distribution system from the first perspective above) [Second form] The first prediction means of the data distribution system described above can acquire second sensor data acquired by a second device different from the first device that acquires the first sensor data, instead of the first sensor data, The first prediction means can adopt a configuration that predicts the communication quality of the network used for transmitting the first sensor data based on the second sensor data. [Third form] The first prediction means of the data distribution system described above can further acquire second sensor data acquired by a second device different from the first device that acquires the first sensor data, The first prediction means can adopt a configuration that predicts the communication quality of the network used for transmitting the first sensor data based on the first sensor data and the second sensor data. [Fourth form] The prediction means of the data distribution system described above can further adopt a configuration that predicts the communication quality using the history of the communication quality of the network. [Fifth form] The data distribution system described above further further includes a second prediction means that predicts the communication quality using the history of the communication quality of the network, The determination means can adopt a configuration that determines a parameter related to the transmission quality of the first sensor data according to the communication quality predicted by the second prediction means according to the change in the communication quality predicted by the first prediction means. [Sixth form] The determination means of the data distribution system described above can adopt a configuration that determines a parameter related to the transmission quality of the first sensor data based on the lower communication quality among the prediction result of the first prediction means and the prediction result of the second prediction means. [Seventh form] The prediction means of the data distribution system described above can adopt a configuration that predicts communication quality using the position information of the first device. [Eighth embodiment] As a parameter related to the quality of the first sensor data in the data distribution system described above, a configuration using the video bit rate can be adopted. [Ninth embodiment] The data distribution system described above can adopt a configuration that increases or decreases the video bit rate by increasing or decreasing at least one of the video resolution, frame rate, target bit rate, QP (Quantization Parameter), CRF (Constant Rate Factor), and encoding method (CODEC type) according to a predetermined setting. [Tenth embodiment] The first device of the data distribution system described above is a camera mounted on a moving body, The first prediction means can adopt a configuration that predicts the communication quality of the network based on an event appearing in the first sensor data as the moving body moves. [Eleventh embodiment] The second device of the data distribution system described above can adopt a configuration that is a fixed camera capable of photographing the moving body. [Twelfth embodiment] The prediction means of the data distribution system described above can adopt a configuration that is arranged in a predetermined receiving device or a relay server on the network. [Thirteenth embodiment] (Refer to the communication quality prediction device from the second perspective above) [Fourteenth embodiment] (Refer to the data transmission device from the third perspective above) [Fifteenth embodiment] (Refer to the data transmission method from the fourth perspective above) [Sixteenth embodiment] (Refer to the program from the fifth perspective above) Note that, similar to the first embodiment, the thirteenth to sixteenth embodiments can be expanded to the second to twelfth embodiments.

[0090] In addition, the disclosures of the above patent documents are hereby incorporated by reference in this document and can be used as the basis or part of the present invention as necessary. Within the scope of the entire disclosure of the present invention (including the claims), modifications and adjustments of the embodiments or examples can be made based on the basic technical idea. Also, within the scope of the disclosure of the present invention, various combinations or selections (including partial deletion) of various disclosure elements (including each element of each claim, each element of each embodiment or example, each element of each drawing, etc.) are possible. That is, the present invention naturally includes all the disclosures including the claims and various modifications and corrections that could be made by those skilled in the art according to the technical idea. In particular, for the numerical ranges described in this document, any numerical value or small range included within the range should be construed as specifically described even without separate description. Furthermore, each disclosure item of the above-cited documents can be used, as necessary and in accordance with the spirit of the present invention, by combining part or all of it with the description items of this document as part of the disclosure of the present invention and is considered to be included in the disclosure items of this application.

Explanation of Reference Numerals

[0091] 10 Data distribution system 10a, 10c, 10e Data transmission device 10b, 10f Communication quality prediction device 11 First prediction means 12 Decision means 13 Encoding means 14 Transmission means 80 Data reception device 90 Network 100, 100a, 100b, 100c, 100d, 100e Communication quality prediction device 101, 101a, 101b, 101d Prediction unit 102, 102c, 102d Data acquisition unit 200 Video transmission device 201 Camera 202 Encoding control unit 203 Encoding unit 300 Video Receiver 301 Communication Quality Measurement Unit 302 Decoding Unit 303 Reproduction Unit 1011 First Predictor 1012 Second Predictor 1013, 1013b Integration Unit 1021 Video Acquisition Unit 1022 Communication Quality Acquisition Unit 9000 Computer 9010 CPU 9020 Communication Interface 9030 Memory 9040 Auxiliary Storage Device

Claims

1. A first prediction means for predicting the communication quality of a network used for transmitting the image based on an object depicted in the image; A second prediction means for predicting the communication quality of the network based on the communication history of the network; An integration means for using, as a prediction result, either the communication quality predicted by the first prediction means or the communication quality predicted by the second prediction means, based on the amount of change in the communication quality predicted by the first prediction means and a threshold value of the amount of change; A determination means for determining a parameter related to the transmission quality of the image according to the prediction result; An encoding means for encoding the image data using the parameter related to the transmission quality of the image; A transmission means for transmitting the encoded image data via the network; A data distribution system including the above.

2. The first prediction means can acquire a second image acquired by a second device different from the first device that acquires the image, instead of the image; The data distribution system according to Claim 1, wherein the first prediction means predicts the communication quality of a network used for transmitting the image based on an object depicted in the second image.

3. The first prediction means can further acquire a second image acquired by a second device different from the first device that acquires the image; The data distribution system according to Claim 1, wherein the first prediction means predicts the communication quality of a network used for transmitting the image data based on objects depicted in the image and the second image, respectively.

4. The integration means uses, as the prediction result, the communication quality predicted by the first prediction means when the amount of change in the communication quality predicted by the first prediction means is greater than the threshold value of the amount of change. The data distribution system according to any one of Claims 1 to 3.

5. The data distribution system according to any one of Claims 1 to 4, wherein the parameter related to the transmission quality of the image is the bit rate.

6. A first prediction means for predicting the communication quality of a network used for transmitting the image based on an object depicted in the image; A second prediction means for predicting the communication quality of the network based on the communication history of the network; An integration means for using, as a prediction result, either the communication quality predicted by the first prediction means or the communication quality predicted by the second prediction means, based on the amount of change in the communication quality predicted by the first prediction means and a threshold value of the amount of change; transmission means for transmitting the prediction result to the device that is the source of the image; to the device that is the source of the image A communication quality prediction device that causes the device that is the source of the image to perform encoding of the image data according to the prediction result and transmission of the encoded image data. **Claim 7** The first prediction means can acquire a second image acquired by a second device different from the first device that acquires the image, instead of the image. The communication quality prediction device according to claim 6, wherein the first prediction means predicts the communication quality of the network used for transmitting the image based on an object depicted in the second image. **Claim 8** The first prediction means can further acquire a second image acquired by a second device different from the first device that acquires the image. The communication quality prediction device according to claim 6, wherein the first prediction means predicts the communication quality of the network used for transmitting the image data based on objects depicted in the image and the second image, respectively. **Claim 9** The integration means uses, as the prediction result, the communication quality predicted by the first prediction means when the amount of change in the communication quality predicted by the first prediction means is greater than the threshold value of the amount of change. The communication quality prediction device according to any one of claims 6 to 8. **Claim 10** A first prediction means for predicting the communication quality of the network used for transmitting the image based on an object depicted in the image; a second prediction means for predicting the communication quality of the network based on the communication history of the network; integration means for using, as the prediction result, either the communication quality predicted by the first prediction means or the communication quality predicted by the second prediction means based on the amount of change in the communication quality predicted by the first prediction means and the threshold value of the amount of change; a communication quality prediction device including transmission means for transmitting the prediction result to the device that is the source of the image; receiving the prediction result; A data transmission device capable of performing encoding of the image data according to the prediction result and transmission of the encoded image data. **Claim 11** The first prediction means predicts the communication quality of the network used for transmitting the image based on an object depicted in the image. The second prediction means predicts the communication quality of the network based on the communication history of the network. Based on the amount of change in communication quality predicted by the first prediction means and the threshold value of the amount of change, either the communication quality predicted by the first prediction means or the communication quality predicted by the second prediction means is used as the prediction result. According to the prediction result, a parameter related to the transmission quality of the image is determined. The data of the image is encoded using the parameter related to the transmission quality of the image. The encoded image data is transmitted via the network. Data transmission method. **Claim 12**: A communication quality prediction device comprising: a first prediction means for predicting the communication quality of a network used for transmitting an image based on an object depicted in the image; a second prediction means for predicting the communication quality of the network based on the communication history of the network; an integration means for using, as a prediction result, either the communication quality predicted by the first prediction means or the communication quality predicted by the second prediction means based on the amount of change in communication quality predicted by the first prediction means and the threshold value of the amount of change; and a transmission means for transmitting the prediction result to the device at the transmission source of the image. The prediction result is transmitted to the device at the transmission source of the image. A data transmission method for causing the device at the transmission source of the image to execute encoding of the image data according to the prediction result and transmission of the encoded image data. **Claim 13**: A data transmission device capable of receiving the prediction result from a communication quality prediction device comprising: a first prediction means for predicting the communication quality of a network used for transmitting an image based on an object depicted in the image; a second prediction means for predicting the communication quality of the network based on the communication history of the network; an integration means for using, as a prediction result, either the communication quality predicted by the first prediction means or the communication quality predicted by the second prediction means based on the amount of change in communication quality predicted by the first prediction means and the threshold value of the amount of change; and a transmission means for transmitting the prediction result to the device at the transmission source of the image. The prediction result is received. A data transmission method for executing encoding of the image data according to the prediction result and transmission of the encoded image data.

Images