Analysis device and analysis method

Abstract

To provide an analysis device and an analysis method for optimizing an amount of data to be analyzed.SOLUTION: An analysis device according to the present invention that analyzes data that is repeatedly transmitted includes an analysis unit, a determination unit, and a change unit. The analysis unit outputs a result of analyzing first data as an analysis result, and analyzes second data having a smaller amount of data than the first data. The determination unit determines whether a difference between the analysis result for the first data and the analysis result for the second data is within a predetermined range. The change unit changes, when the determination unit determines that the difference is within the predetermined range, an amount of the first data analyzed by the analysis unit to an amount of data of the second data, and further reduces the amount of data of the second data.SELECTED DRAWING: Figure 1B

Description

[Technical Field]

[0001] The present invention relates to an analysis device and an analysis method. [Background technology]

[0002] Conventionally, there is an analysis system in which image data acquired from an external source is analyzed by a server device. In the field of such analysis systems, a technique has been proposed for reducing the communication load by suppressing the amount of image data, for example, by lowering the resolution of the image data (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-128948 Summary of the Invention [Problem to be solved by the invention]

[0004] However, with conventional techniques, it is not easy to determine to what extent the amount of data should be reduced, and there is room for improvement in optimizing the amount of data to be analyzed.

[0005] The present invention has been made in view of the above, and has as its object to provide an analysis device and an analysis method that can optimize the amount of data to be analyzed. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, an analysis device according to the present invention analyzes repeatedly transmitted data, and includes an analysis unit, a determination unit, and a change unit. The analysis unit outputs the result of analyzing first data as an analysis result, and analyzes second data having a smaller data volume than the first data. The determination unit determines whether a difference between the analysis result for the first data and the analysis result for the second data is within a predetermined range. When the determination unit determines that the difference is within the predetermined range, the change unit changes the first data analyzed by the analysis unit to the data volume of the second data, and also changes the data volume of the second data to a data volume that is further reduced. [Effects of the Invention]

[0007] According to the present invention, the amount of data to be analyzed can be optimized. [Brief explanation of the drawings]

[0008] [Figure 1A] FIG. 1A is a schematic diagram showing an example of the configuration of an analysis system. [Figure 1B] FIG. 1B is a diagram showing an outline of the analysis method. [Figure 2] FIG. 2 is a block diagram of the analysis device. [Figure 3] FIG. 3 is a diagram illustrating an example of the analysis result storage unit. [Figure 4] FIG. 4 is a diagram illustrating an example of the switching condition storage unit. [Figure 5] FIG. 5 is a flowchart showing a processing procedure executed by the analysis device. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, an analysis device and an analysis method according to the present application will be described in detail with reference to the drawings. Note that the analysis device and the analysis method according to the present application are not limited to the embodiments.

[0010] First, an overview of an analysis device and an analysis method according to an embodiment will be described with reference to Figures 1A and 1B. Figure 1A is a schematic diagram showing an example of the configuration of an analysis system. Figure 1B is a diagram showing an overview of the analysis method. Note that the analysis method is executed, for example, by an analysis device 10 shown in Figure 1A.

[0011] 1A, an analysis system 1 according to an embodiment includes an analysis device 10, a service providing device 100, and a plurality of cameras 50. For example, the plurality of cameras 50 are installed on roads at various locations and capture images of road conditions.

[0012] Analysis device 10 acquires image data transmitted from each camera 50 via a predetermined network and analyzes the image data to analyze the road congestion situation at each point. Note that, for example, the functions of analysis device 10 may be provided in each camera 50.

[0013] Furthermore, analysis device 10 may be installed in a center (base station) and perform centralized analysis of data from multiple cameras 50. When analysis device 10 is installed in camera 50, analysis device 10 and camera 50 may have a one-to-one relationship, and when multiple cameras 50 are set up in one location, multiple cameras 50 are connected to one analysis device 10 as shown in FIG.

[0014] The service providing device 100 provides various services according to the congestion status at each location acquired from the analysis device 10. For example, the service providing device 100 provides a service of notifying each vehicle (not shown) of the congestion status. Furthermore, for example, the service providing device 100 may provide each vehicle with services such as calculating the arrival time to the destination according to the congestion status, or providing route guidance that will alleviate the congestion status.

[0015] Incidentally, for example, when the analysis device 10 acquires and analyzes image data from each camera 50, it is possible to analyze the congestion situation more accurately if the image data is large-volume data (e.g., high resolution, high frame rate).

[0016] On the other hand, for example, if the amount of image data is excessive, the resource effectiveness will be reduced accordingly, such as the load of analyzing the image data and the communication load associated with transferring the image data.

[0017] Therefore, while it is preferable to optimally tune the amount of image data acquired from each camera 50, it is not realistic to manually adjust the settings of each camera 50. Furthermore, the optimal conditions for each camera 50 vary depending on, for example, the weather, the time of day, etc., making it difficult to adjust each time.

[0018] Therefore, in the analysis method of the embodiment, the second data, the data volume of which has been intentionally reduced, is analyzed in parallel to optimize the data volume of image data acquired from each camera 50.

[0019] 1B, in the analysis method according to the embodiment, in a configuration in which analysis processing based on data transmitted from camera 50 is repeatedly performed, for example, analysis processing is performed on first data in one analysis processing cycle, and the analysis results of the first data are output. Furthermore, in the analysis method according to the embodiment, analysis processing is performed on second data, which has a smaller data volume than the first data, during spare time in the analysis processing of the first data.

[0020] Here, the first data is image data having a predetermined amount of data, for example, data for which the accuracy of the analysis results has already been guaranteed, and the second data is image data having a smaller amount of data than the first data and captured at the same time as the first data.

[0021] That is, since the amount of the first data is larger than the amount of the second data, as shown in the figure, the processing load related to the analysis process of the first data is higher than the processing load related to the analysis process of the second data.

[0022] For example, in an analysis method according to an embodiment, after completing an analysis process for the first data and an analysis process for the second data in one processing cycle, the analysis results of both are compared. If there is no significant difference between the analysis results of both, the analysis method according to an embodiment changes the amount of the first data in the next processing cycle to the amount of the second data, and changes the amount of the second data to data with an even smaller amount of data.

[0023] That is, in the analysis method according to the embodiment, if the analytical accuracy of the second data is ensured in the current processing cycle, the data amount of the first data in the next and subsequent processing cycles is switched to the data amount of the second data.

[0024] Furthermore, for example, if there is no significant difference between the analysis results of the first data and the second data in the next processing cycle, the amount of data in the first data will be further reduced in the processing cycle after that.

[0025] Accordingly, for example, in the analysis method according to the embodiment, the amount of first data acquired from each camera 50 is switched to the amount of second data, and the shooting setting of the image data of each camera 50 is switched to the shooting mode of the second data. This makes it possible to reduce the amount of image data acquired from each camera 50 as needed.

[0026] Furthermore, in the analysis method according to the embodiment, in the next and subsequent processing cycles, the amount of the second data is further reduced before the above processing is performed. That is, the analysis method according to the embodiment reduces the amount of the first data as needed until a significant difference appears between the analysis results of the first data and the second data.

[0027] As a result, the analysis method according to the embodiment can optimize the amount of data to be analyzed. Also, the analysis method according to the embodiment can appropriately adjust each camera 50 to the optimal shooting conditions by switching the shooting settings of each camera 50 according to the amount of data to be analyzed.

[0028] Next, a configuration example of the analysis device 10 according to the embodiment will be described with reference to Fig. 2. Fig. 2 is a block diagram of the analysis device 10. As shown in Fig. 2, the analysis device 10 includes a communication unit 110, a control unit 120, and a storage unit 130.

[0029] The communication unit 110 is a communication module for performing data communication with each camera 50 and the service providing device 100 via a predetermined network.

[0030] The memory unit 130 is realized, for example, by a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, and in the example of Figure 2, the memory unit 130 has an analysis result memory unit 131 and a switching condition memory unit 132.

[0031] The analysis result storage unit 131 is a storage area that stores the analysis results obtained by the analysis unit 121, which will be described later. Fig. 3 is a diagram showing an example of information stored in the analysis result storage unit 131. As shown in Fig. 3, for example, the analysis result storage unit 131 stores information on items such as "camera ID," "first data amount," "analysis result," "second data amount," and "analysis result" in association with one another.

[0032] "Camera ID" is an identifier for identifying each camera 50. "First data amount" indicates the amount of first data, "analysis result" indicates the analysis result for the first data, "second data amount" indicates the amount of second data, and "analysis result" indicates the analysis result for the second data.

[0033] In the example shown in Figure 3, the first data amount of camera 50 identified by camera ID "C001" is "P1", the analysis result is "D1", the second data amount is "P2", and the analysis result is "D2".

[0034] Returning to the explanation of Fig. 2, the switching condition storage unit 132 will be described. The switching condition storage unit 132 is a storage area that stores switching conditions for the data amounts of the first data and the second data. Note that the switching conditions here refer to conditions for increasing the data amounts of the first data and the second data, for example.

[0035] Fig. 4 is a diagram showing an example of information stored in the switching condition storage unit 132. As shown in Fig. 4, for example, the switching condition storage unit 132 stores information items such as "condition ID" and "condition" in association with each other.

[0036] "Condition ID" is an identifier for identifying each switching condition. "Condition" indicates details of the switching condition. For example, in the example shown in FIG. 4, conditions such as "significant difference occurs three or more times per hour" and "predetermined time has elapsed" are stored as switching conditions.

[0037] Here, "a significant difference was found three or more times per hour" means that, for example, a significant difference was found between the analysis result for the first data and the analysis result for the second data three times within one hour.

[0038] As described above, if there is no significant difference between the analysis results for the first data and the analysis results for the second data, the analysis device 10 gradually reduces the amount of the first data. However, if the amount of the first data is reduced, there is a risk that the accuracy of the analysis of the image data will fall below the expected value.

[0039] Therefore, for example, if "significant differences are found three or more times per hour" is established, the data amounts of the first data and the second data can be increased to ensure the accuracy of the analysis of the image data.

[0040] Furthermore, the "predetermined time has elapsed" in FIG. 4 indicates, for example, that a predetermined time has elapsed since the amount of data of the first data and the second data was reduced. In other words, by increasing the amount of data of the first data and the second data at a predetermined cycle, the accuracy of the analysis of the image data can be ensured. Note that the "predetermined time" here is preferably longer than the processing cycle of the repeated analysis process.

[0041] Even if the data volume of the first data and the second data is increased, if there is no significant difference between the analysis results for the first data and the analysis results for the second data, the data volume of the first data will be gradually reduced, thereby optimizing the data volume to be analyzed.

[0042] Returning to the explanation of Fig. 2, the control unit 120 will be described. The control unit 120 is a controller, and is realized by, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs (not shown) stored in the storage unit 130 using RAM as a work area. The control unit 120 can also be realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0043] 2, the control unit 120 includes an analysis unit 121, a determination unit 122, a change unit 123, and an output unit 124. The analysis unit 121 outputs the result of analyzing the first data as the analysis result, and also analyzes second data having a smaller data amount than the first data.

[0044] For example, the analysis unit 121 analyzes traffic conditions from image data transmitted from each camera 50. The analysis unit 121 analyzes first data from the image data transmitted from the camera 50, and writes the analysis result to the analysis result storage unit 131.

[0045] For example, the analysis unit 121 performs an analysis of the image data using artificial intelligence (AI) to detect and track vehicles and analyze the number of passing vehicles for each lane. Furthermore, the analysis unit 121 analyzes second data, which has a smaller data volume than the first data, during spare time from the analysis process of the first data. The second data may be transmitted from the camera 50, or may be created by processing the first data on the analysis device 10 side.

[0046] After performing the same analysis process on the second data as on the first data, the analysis unit 121 writes the analysis results to the analysis result storage unit 131. Note that the analysis process on the second data by the analysis unit 121 may be performed less frequently than the analysis process on the first data.

[0047] The determination unit 122 determines whether the difference between the analysis result for the first data and the analysis result for the second data is within a predetermined range. For example, the determination unit 122 refers to the analysis result storage unit 131 and compares the analysis result for the first data with the analysis result for the second data.

[0048] Here, for example, the determination unit 122 compares the analysis result for the first data with the analysis result for the second data, and if the difference in traffic volume indicated by each analysis result is less than ±1%, it determines that the difference between the analysis result for the quasi-data and the analysis result for the second data is within a predetermined range, i.e., that there is no significant difference. Note that it may also be determined that there is no significant difference when there is no clear difference, for example, when there is no change in whether or not a vehicle is detected, or no change in the number of detected vehicles.

[0049] In other words, if the difference in the traffic volume exceeds ±1%, the determining unit 122 determines that there is a significant difference between the two analysis results. Then, the determining unit 122 passes the determination result to the changing unit 123 for each determination.

[0050] When the determination unit 122 determines that the difference is within a predetermined range, the change unit 123 changes the first data analyzed by the analysis unit 121 to the data amount of the second data, and also changes the data amount of the second data to a data amount that is further reduced.

[0051] For example, when it is determined that there is no significant difference between the analysis result for the first data and the analysis result for the second data, the change unit 123 issues a reduction instruction to the corresponding camera 50 to reduce the data volume of the image data to be transmitted to the current data volume of the second data. That is, the change unit 123 sets the shooting settings for each camera 50 based on the analysis results of the first data and the second data.

[0052] As a result, from now on, image data with reduced data volume will be sent from camera 50, and analysis unit 121 will analyze the image data with reduced data volume, thereby reducing the processing load on analysis unit 121 and the communication load between analysis device 10 and camera 50.

[0053] Here, reducing the amount of data refers to reducing at least one of the resolution and frame rate of the image data. That is, the amount of data may be reduced by reducing the resolution, or by reducing the frame rate.

[0054] This allows the resolution and frame rate of the image data captured by each camera 50 to be optimized to appropriate values.

[0055] Also, for example, if the amount of data is reduced, one of the resolution or the frame rate may be reduced and the other may be increased. Note that the extent to which the amount of image data is reduced may be determined using, for example, a predetermined optimization algorithm.

[0056] Furthermore, for example, when a predetermined switching condition is met, the change unit 123 issues an increase instruction to increase the amount of first data to the corresponding camera 50. As shown in Fig. 4, for example, the switching condition may be "there is a significant difference a predetermined number of times (e.g., three times) or more per predetermined time (e.g., one hour)" or "a predetermined time has passed."

[0057] When these switching conditions are met, the change unit 123 instructs the camera 50 to increase the amount of the first data, thereby ensuring the accuracy of analysis of the first data.

[0058] The output unit 124 outputs the analysis result for the first data to the service providing device 100 via the communication unit 110. This allows the service providing device 100 to provide various services using the analysis result by the analysis device 10.

[0059] Next, a processing procedure executed by the analysis device 10 according to the embodiment will be described with reference to Fig. 5. Fig. 5 is a flowchart showing the processing procedure executed by the analysis device 10. Note that the processing procedure shown below is repeatedly executed by the control unit 120 every time image data is acquired.

[0060] As shown in FIG. 5, when analysis device 10 acquires image data (step S101), it analyzes the first data (step S102) and, during the free time, analyzes the second data (step S103).

[0061] Next, analysis device 10 determines whether there is a difference between the analysis results of the first data and the second data (step S104). If analysis device 10 determines that there is no difference between the analysis results of the first data and the second data (step S104; Yes), analysis device 10 issues a reduction instruction to camera 50 to reduce the data amount of the first data to the data amount of the second data (step S105), and ends the process.

[0062] Furthermore, if the analysis device 10 determines in step S104 that there is a difference between the two analysis results (step S104; No), it proceeds to processing in step S106 and determines whether or not the switching condition, for example, "significant difference three or more times per hour," is met (step S106).

[0063] If the analysis device 10 determines that the switching condition is not met (step S106; No), the analysis device 10 proceeds to the processing of step S101. On the other hand, if the analysis device 10 determines that the switching condition is met (step S106; Yes), the analysis device 10 issues an increase instruction to the camera 50 to increase the amount of first data (step S107), and ends the processing.

[0064] As described above, the analysis device 10 according to the embodiment is an analysis device that analyzes repeatedly transmitted data, and includes an analysis unit 121, a determination unit 122, and a change unit 123. The analysis unit 121 outputs the result of analyzing the first data as the analysis result, and analyzes the second data, which has a smaller data volume than the first data. The determination unit 122 determines whether the difference between the analysis result for the first data and the analysis result for the second data is within a predetermined range.

[0065] When the determination unit 122 determines that the difference is within a predetermined range, the change unit 123 changes the amount of first data analyzed by the analysis unit 121 to the amount of second data, and also changes the amount of second data to a further reduced amount. Therefore, the analysis device 10 according to the embodiment can optimize the amount of data to be analyzed.

[0066] In the above-described embodiment, the case of analyzing image data has been described, but the present invention is not limited to this. That is, the present invention may be applied to the case of analyzing data that is not limited to image data but is a predetermined analysis target.

[0067] In the above-described embodiment, the second data is analyzed during a free period of the first data, but this is not limiting. For example, the first data may be analyzed in the current processing cycle, and second data having a smaller data volume than the first data may be acquired in the next processing cycle, and then the second data may be analyzed. In this case, if there is no significant difference between the analysis result of the first data analyzed in the current processing cycle and the analysis result of the second data analyzed in the next processing cycle, the data volume of the first data may be changed to the data volume of the second data in the next processing cycle or later.

[0068] Further advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents. [Explanation of symbols]

[0069] 1. Analysis system 10 Analysis device 50 cameras 100 Service providing device 101 Steps 120 control section 121 Analysis Department 122 Judgment section 123 Changes 124 Output section 130 Storage section 131 Analysis result storage unit 132 Switching condition memory unit

Claims

1. An analysis device that receives and analyzes data repeatedly transmitted from a camera, outputting the result of analyzing the first data received from the camera as an analysis result, and analyzing second data, which is image data of the same time period and has a smaller amount of data than the first data, received from the camera or created by processing the first data, during a free time from the analysis of the first data; When it is determined that the difference between the analysis result for the first data and the analysis result for the second data is within a predetermined range, the first data is changed to the data amount of the second data, and the data amount of the second data is changed to a data amount that is further reduced; increasing the amount of the first data when it is determined that the difference between the analysis result for the first data and the analysis result for the second data is outside a predetermined range; Analysis device.

2. The second data has at least one of a resolution and a frame rate lower than that of the first data. The analysis device according to claim 1 .

3. The analysis of the second data is performed less frequently than the analysis of the first data. The analysis device according to claim 1 or 2.

4. Vehicles are detected from image data of the road captured by the camera installed on the road, and the number of vehicles is used as the analysis result. The analysis device according to any one of claims 1 to 3.

5. An analysis method performed by an analysis device that analyzes repeatedly transmitted data, comprising: outputting a result of analyzing the first data as an analysis result, and analyzing second data having a smaller amount of data than the first data; When it is determined that the difference between the analysis result for the first data and the analysis result for the second data is within a predetermined range, changing the first data to the data amount of the second data and also changing the data amount of the second data to a data amount that is further reduced; increasing the amount of the first data when it is determined that the difference between the analysis result for the first data and the analysis result for the second data is outside a predetermined range; Analysis method.

Images