JP2024079364A - Driving assistance system, driving assistance method, and driving assistance program - Google Patents

Abstract

To suppress a load with respect to a calculation processing unit of an on-vehicle device while effectively utilizing a plurality of functions related to image recognition.SOLUTION: In a driving assistance system 100 including an on-vehicle device 3 mounted on a vehicle and a management device 1 outside the vehicle, the on-vehicle device 3 has an image recognition control unit 32 and a memory unit 33 for storing a plurality of libraries for operating functions related to the image recognition, and the management device 1 has a function selection unit 11 for selecting a library to be an operation target from the plurality of libraries. The image recognition control unit 32 operates a library selected by the function selection unit 11 from the plurality of libraries in accordance with the selection made by the function selection unit.SELECTED DRAWING: Figure 1

Description

The present invention relates to a driving assistance system, a driving assistance method, and a driving assistance program.

In-vehicle devices that record vehicle operation information, such as digital tachographs and drive recorders, have become increasingly sophisticated in recent years. As a result, not only do they have at least an image recognition function using a camera mounted on the vehicle, but the images that the in-vehicle devices can recognize are becoming more diverse, and the objects that can be detected by image recognition processing are becoming more diverse.

For example, Patent Document 1 discloses a control device that enables a vehicle surroundings display control device, which switches between images captured by a camera capturing the surroundings of the vehicle, to switch the display screen based on the road configuration. Patent Document 2 discloses a control device that selects which of images captured by multiple cameras capturing the surroundings of the vehicle to display based on the road shape that will be reached in a predetermined time, and displays the selected captured image on a display device. Patent Document 3 discloses an in-vehicle device that, when processing images from an in-vehicle camera to detect moving objects, switches the number of detection frames depending on the state of the vehicle, such as moving forward, stopped, or backward, making it possible to recognize important areas in any state and reducing the load of image processing.

JP 2005-309812 A JP 2017-194784 A JP 2019-28481 A

However, there is a limit to the processing power of the onboard device's arithmetic processing unit, and as the targets of image recognition processing become more diverse, it becomes difficult to use all pre-installed image recognition-related functions (hereinafter simply referred to as "functions") while the vehicle is traveling. Using a arithmetic processing unit with high processing power would make it possible to use all functions, but this would result in increased costs.

It is also possible to set the functions that can be used in advance, so long as it does not exceed the processing capacity of the arithmetic processing unit, and set up a setting that allows only the set functions to be used effectively. In this case, functions that were not selected will never be used thereafter, making it difficult to effectively use the multiple functions that have been developed.

The present invention provides a driving assistance system, a driving assistance method, and a driving assistance program that can effectively utilize multiple functions related to image recognition while reducing the load on the on-board device's processing unit.

In order to achieve the above-mentioned object, the driving assistance system according to the present invention has the following features.
A driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device includes:
An image recognition control unit;
A storage unit that stores a plurality of programs for operating a plurality of functions related to image recognition,
each of the plurality of programs corresponds to each of the plurality of functions;
The management device includes:
a function selection unit for selecting a program to be operated from among the plurality of programs,
the image recognition control unit operates a program selected by the function selection unit in response to a selection made by the function selection unit;
Driving assistance system.

In order to achieve the above-mentioned object, a driving assistance method according to the present invention has the following features.
A driving assistance method using a driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device has a plurality of programs for operating a plurality of functions,
each of the plurality of programs corresponds to each of the plurality of functions;
receiving, in the management device, a selection by a user of a program to be operated from among the plurality of programs;
the vehicle-mounted device operates a program selected by the management device among the plurality of programs in response to a selection by the management device;
Driving assistance methods.

In order to achieve the above-mentioned object, a driving assistance program according to the present invention has the following features.
A driving assistance program for executing driving assistance by a driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device has a plurality of programs for operating a plurality of functions,
each of the plurality of programs corresponds to each of the plurality of functions;
On the computer,
receiving, by the management device, a selection by a user of a program to be operated from among the plurality of programs;
a step of causing the vehicle-mounted device to operate a program selected by the management device among the plurality of programs in response to a selection made by the management device;
A driving assistance program that executes the following:

According to the present invention, the management device sets the programs to be operated in the vehicle-mounted device, so that it is possible to effectively use multiple functions related to image recognition while reducing the load on the vehicle-mounted device's processing unit.

The present invention has been briefly described above. The details of the present invention will become clearer by reading the following description of the embodiment of the invention (hereinafter referred to as "embodiment") with reference to the attached drawings.

FIG. 1 is a block diagram showing an example of the configuration of a driving assistance system according to an embodiment of the present invention. FIG. 2 is a table showing the relationship between each group and the selected function. FIG. 3 is a table showing groups selected for each date and time or each day of the week. FIG. 4A is a conceptual diagram showing an example of a display by the function selection unit. FIG. 4B is a conceptual diagram showing another example of the display by the function selection unit. FIG. 5 is a flowchart illustrating the procedure of the presetting. FIG. 6 is a flow chart illustrating the operation when the vehicle is leaving the garage. FIG. 7 is a flowchart illustrating the procedure of the post-mortem analysis. FIG. 8 is a flowchart showing a procedure for reflecting the analysis.

Specific embodiments of the present invention are described below with reference to the drawings.

[Configuration of driving assistance system]
1 is a block diagram showing an example of the configuration of a driving assistance system 100 according to an embodiment of the present invention. The driving assistance system 100 is a system operated by a vehicle operating company or the like to assist a driver in driving, and includes a management device 1 and an in-vehicle device 3.

The management device 1 is installed within the vehicle operation company, and manages the vehicle-mounted device 3 and analyzes its operation information. The management device 1 is used to assist the driver. The vehicle-mounted device 3 is a device that is mounted on the vehicle and records the vehicle's operation information; specific examples include a digital tachograph and a drive recorder. The user can exchange information between the management device 1 and the vehicle-mounted device 3 using a recording medium such as a specified operation card, or a communication medium such as short-range wireless communication.

The management device 1 includes a pre-setting unit 10 that performs various settings related to the on-board device 3 before the vehicle is operated, and a post-analysis unit 20 that performs various analyses based on information obtained from the on-board device 3 after the vehicle is operated. The pre-setting unit 10 includes a function selection unit 11, a management unit 12, and a configuration file creation unit 13. Meanwhile, the post-analysis unit 20 includes an alarm history input unit 21, an alarm history analysis unit 22, and an analysis result output unit 23.

The parts shown in FIG. 1 mainly represent the functional parts of the management device 1, and the functions are executed based on programs read from a memory or the like by an arithmetic processing unit (computer) that is responsible for the main control processing of the management device 1. Some functions also include specific hardware, and for example, the function selection unit 11 has an interface that allows the user to make specific selections by input. This interface is, for example, a touch panel.

The function selection unit 11 is an input unit that allows the user to select a program to be operated from among the multiple programs, and is configured as hardware such as a touch panel, keyboard, mouse, etc. The multiple programs are programs that operate functions related to image recognition in the vehicle-mounted device 3. In other words, each of the multiple programs corresponds to each of the multiple functions. In this embodiment, multiple libraries are used as a specific example of the multiple programs. The multiple libraries will be described in detail later.

The management unit 12 monitors whether the library selected as the operation target by the function selection unit 11 exceeds the processing capacity of the arithmetic processing unit that is responsible for the main control processing of the vehicle-mounted device 3. However, if the aforementioned libraries are grouped so as not to exceed the processing capacity of the arithmetic processing unit and the user selects one of the groups, the management unit 12 is not necessarily required. This point will be described later.

The setting file creation unit 13 creates a predetermined setting file that includes information indicating the library selected as the target for operation, such as a code indicating the library name, and a flag indicating whether each library is to be enabled or disabled. In other words, the setting file creation unit 13 creates a setting file for enabling the program selected by the function selection unit 11. The setting file creation unit 13 saves the created setting file in a memory, an operation card, etc. The setting file created by the setting file creation unit 13 is output to the vehicle-mounted device 3 by a reader/writer, a wireless communication unit, etc. of the operation card.

Details of the alarm history input unit 21, alarm history analysis unit 22, and analysis result output unit 23 will be described later.

The vehicle-mounted device 3 comprises a setting file input unit 31, an image recognition control unit 32, a memory unit 33, an alarm history measurement unit 34, and an alarm history output unit 35. Each of these units mainly represents a functional unit of the vehicle-mounted device 3, and the functions are executed based on a program read from a memory or the like by an arithmetic processing device (computer) that is responsible for the main processing of the vehicle-mounted device 3. Some functions also include specified hardware, and for example, the memory unit 33 comprises a memory that stores specified data.

The configuration file input unit 31 is an input unit that inputs the configuration file output from the management device 1, and its hardware consists of an operation card reader/writer, a wireless communication unit, etc.

The image recognition control unit 32 operates a library selected as enabled in the configuration file output from the management device 1, among multiple programs that operate multiple functions stored in the memory unit 33. That is, the image recognition control unit 32 issues an instruction to enable or disable each of the multiple libraries based on the information in the configuration file in accordance with the selection in the function selection unit 11. The image recognition control unit 32 operates the library selected by the function selection unit 11 and does not operate the library that was not selected.

The operation of the image recognition control unit 32 will be described in detail. When the image recognition control unit 32 receives video footage from a camera mounted on the vehicle, it saves the video footage in a specified location. Each library enabled by the image recognition control unit 32 performs object detection processing, etc. on the video footage saved in the specified location, and if an object is detected, it passes the results to the image recognition control unit 32. The image recognition control unit 32 takes into account vehicle information and determines whether to output the data processing results of detected objects, etc. by each enabled function as analysis results.

In this way, the image recognition control unit 32 analyzes the data processing results from each function while taking into account vehicle information, and outputs the results as the analysis results from the image recognition function. The image recognition control unit 32 may further output information such as an alarm, event, or trigger in addition to the analysis results. If the analysis results output from the image recognition control unit 32 are accompanied by information such as an alarm, another application program that receives this information can issue an alarm to alert the driver.

As an example, assume that the image recognition control unit 32 receives a result from a certain library indicating that a vehicle has been detected ahead in the lane in which the vehicle is traveling. The image recognition control unit 32 takes into account the vehicle information and determines that a vehicle has been detected when the vehicle is traveling at or above a predetermined speed, and passes the determination result to another application program. On the other hand, even if the image recognition control unit 32 receives a result from this library indicating that a vehicle has been detected ahead, if the vehicle is stopped or traveling at less than a predetermined speed, such as in a traffic jam, the image recognition control unit 32 does not determine that a vehicle has been detected and does not output the determination result. In such a situation, there is no danger even if there is another vehicle ahead, so the image recognition control unit 32 can prevent unnecessary alarms from being issued by not outputting such detection results.

The memory unit 33 is an operating program for all functions to be set in the vehicle-mounted device 3, and stores a plurality of libraries 1 to n for implementing a plurality of functions 1 to n (n is a natural number). Each of the plurality of libraries 1 to n corresponds to a respective one of the plurality of functions 1 to n. The memory unit 33 is configured as hardware, such as a memory. Of the plurality of libraries stored in the memory unit 33, each library that is enabled performs object detection processing, etc. on the video stored in a specified location in response to instructions from the image recognition control unit 32, and if an object is detected, passes the result to the image recognition control unit 32.

The alarm history measurement unit 34 measures the alarm occurrence history that occurs for each driver of the vehicle based on the operation results of each library. The alarm history output unit 35 is an output unit that outputs the alarm occurrence history measured by the alarm history measurement unit 34 to the management device 1, and is composed of hardware such as a reader/writer for the operation card and a wireless communication unit.

Returning to the post-analysis unit 20 of the management device 1, the alarm history input unit 21 is an input unit that inputs the alarm occurrence history output from the alarm history output unit 35 of the vehicle-mounted device 3, and is composed of hardware such as a reader/writer for the operation card and a wireless communication unit. The alarm history analysis unit 22 analyzes the input alarm occurrence history. The analysis result output unit 23 outputs the analysis result by the alarm history analysis unit 22, and is composed of hardware such as an LCD monitor and a touch panel.

[Image recognition function]
Next, a description will be given of a number of functions related to image recognition and a number of libraries that operate these functions, which are possessed by the in-vehicle device 3. Due to the increasing demand for safety performance, the number of cameras mounted on vehicles is increasing, functions related to image recognition are becoming more sophisticated, and detection targets by image processing are becoming more diverse.

Previous vehicles were limited in the number of cameras they could install. As a result, the image recognition function of the onboard device and the types of warnings that accompanied it were also limited. For example, the types of warnings that previous onboard devices had were lane departure warning, forward vehicle approach warning, and road marking warning.

・Lane departure warning: A warning that is issued when the vehicle deviates from the lane ・Front vehicle approach warning: A warning that is issued when a vehicle in front approaches ・Road marking warning: A warning that is issued when markings such as diamonds or speed limit signs appear on the road surface on which the vehicle is traveling

Meanwhile, modern vehicles are equipped with multiple types of cameras, including front cameras, rear cameras, and in-car cameras. Front cameras are mounted at the front of the vehicle and capture images of the vehicle ahead to detect the distance between the vehicle and the vehicle ahead. Rear cameras are mounted at the rear of the vehicle and capture images of people, objects, and vehicles behind when backing up to detect the distance between them, etc. In-car cameras are mounted inside the vehicle and capture images of the driver to detect their behavior.

As the number of cameras increases, image recognition capabilities improve, and new functions such as the following are now required for in-vehicle devices.

- Pedestrian detection function: Detects pedestrians using the front and rear cameras.
- Traffic light detection function: Detects traffic lights using the front camera.
- Detection of aggressive driving and/or tailgating: Using the front and rear cameras to detect the distance between vehicles, aggressive driving and/or tailgating is detected.
- Crosswalk detection function: Uses the front camera to detect crosswalks from road markings.
Distracted driving detection function: Detects when the driver is distracted while driving using an in-car camera.

In addition, there are cases where multiple objects are detected, such as vehicles and pedestrians ahead, and where a single detection result is output by multiple functions, such as detecting aggressive driving using the detection result from the forward vehicle detection function. Furthermore, libraries corresponding to each function include those that perform image recognition by pattern matching, those that perform image recognition by artificial intelligence (AI), and calibration for setting various parameters related to image recognition. In order to enable processing related to multiple functions related to image recognition as described above, it is necessary to improve the information processing capacity of the vehicle-mounted device 3, that is, the processing capacity of the arithmetic processing device (computer) that is responsible for the main control processing of the vehicle-mounted device 3, such as a CPU (Central Processing Unit). However, arithmetic processing devices with such high information processing capacity are expensive, which leads to an increase in the manufacturing cost of the vehicle-mounted device.

In the driving assistance system 100 of this embodiment, the user can use the function selection unit 11 to select a library to be operated from among multiple libraries that operate functions related to image recognition and are stored in the storage unit 33 of the vehicle-mounted device 3. The image recognition control unit 32 of the vehicle-mounted device 3 then issues an enable/disable instruction for each library based on the information in the setting file for enabling the library selected by the function selection unit 11, and can operate the enabled library. In addition, libraries that are disabled in the setting file will not operate. This makes it possible to reduce the processing power (load) required for the arithmetic processing unit of the vehicle-mounted device 3 in advance, thereby avoiding the use of a arithmetic processing unit with high processing power and reducing the manufacturing costs of the vehicle-mounted device 3.

In other words, the management device 1 sets the library to be operated on in the vehicle-mounted device 3, so that it is possible to effectively use multiple image recognition-related functions while reducing the load on the processing unit of the vehicle-mounted device 3.

In addition, previously, dealers of the in-vehicle device 3 would restrict the image recognition function, i.e., suspend the library, in advance based on information received from users, but this will now make it possible to omit such work.

In particular, in this embodiment, the management unit 12 monitors whether the library selected by the function selection unit 11 exceeds the processing capacity of the arithmetic processing unit that is responsible for the main control processing of the vehicle-mounted device 3. This makes it possible to prevent a library that exceeds the processing capacity of the arithmetic processing unit from being selected. However, as described below, if each library is grouped so as not to exceed the processing capacity of the arithmetic processing unit and the user selects one of the groups, the monitoring function of the management unit 12 is not necessarily required. Note that certain functions, such as the calibration function, may be set to be always enabled.

[How to select a function]
Next, a specific method for selecting a function (library) to be operated will be described. In this embodiment, the user pre-sets a plurality of "groups" that collectively include libraries corresponding to one or more image recognition functions among various functions (libraries) related to image recognition. The combination of libraries in one group is set so that the processing capacity (load occupancy) of the arithmetic processing device that is responsible for the main control processing of the vehicle-mounted device 3 is not exceeded even if all the libraries of the functions included in one group are operated simultaneously.

Figure 2 is a table showing the relationship between each group and the selected function. The vehicle-mounted device 3 has the following functions: function 1, function 2, function 3, function 4, function 5, .... According to the table in Figure 2, group A includes function 1, function 2, and function 3, but does not include function 4 or function 5. Group B includes function 1, function 3, and function 4, but does not include function 2 or function 5. Each of groups A to E does not exceed the processing capacity of the arithmetic processing unit of the vehicle-mounted device 3, even if the libraries of each included function all operate simultaneously.

Figure 3 is a table showing the groups selected for each date and time or day of the week. After determining each group as in Figure 2, the user can set which group to adopt and which function to enable for each operation day. The tables in Figures 2 and 3 can be created by the user on the management device 1, and then installed and set in the vehicle-mounted device 3.

Figure 4A is a conceptual diagram showing an example of a display by the function selection unit 11. The function selection unit 11 displays the image as shown, for example, on an LCD display. Check boxes are arranged beside each function displayed vertically, and the user can select the required function by checking at least one function using an input device such as a mouse. However, sometimes the user selects too many functions, exceeding the processing capacity of the arithmetic processing unit. In such a case, the management unit 12 may display an error on the LCD display or sound an error through the speaker.

Figure 4B is a conceptual diagram showing another example of display by the function selection unit 11. Check boxes are arranged next to each group (see Figure 2) displayed in a vertical row, and the user can check and select one group using an input device such as a mouse. As described above, even if all functions of each group are operated simultaneously, the processing capacity of the arithmetic processing unit of the vehicle-mounted device 3 is not exceeded. However, the user may mistakenly select multiple groups. In such a case, the management unit 12 may display an error on the LCD display or sound an error through the speaker.

That is, the user can select at least one group through the screen of FIG. 4B using the function selection unit 11. This allows the user to select a group that includes appropriate functions depending on the driving situation, the driver, etc.

The user can also select one group according to the operation day on which the vehicle is to operate, as shown in FIG. 3, via the screen of FIG. 4B using the function selection unit 11. This allows the user to select a group that includes appropriate functions according to the characteristics of the operation day.

[How to use the warning history]
Next, a method of utilizing the alarm occurrence history acquired by the vehicle-mounted device 3 during leaving the garage will be described. The alarm history measurement unit 34 of the vehicle-mounted device 3 measures the alarm occurrence history that occurred during the operation of each library for each driver of the vehicle. That is, the alarm history measurement unit 34 measures the occurrence history of alarms that occurred during the operation of each function. The alarm occurrence history is typically the number of alarms, and the alarm history measurement unit 34 counts the number of alarms during the operation of the function. However, the alarm history measurement unit 34 can measure a wide range of statistical information as the alarm occurrence history, such as not only the number of alarms but also the duration of the alarm and the time period in which the alarm occurred. The alarm history output unit 35 outputs the alarm occurrence history measured by the alarm history measurement unit 34 to the management device 1.

The warning history input unit 21 of the management device 1 inputs the warning occurrence history output from the warning history output unit 35 of the vehicle-mounted device 3. As shown in FIG. 3, the "group" applied changes for each driving day, so each driver will ultimately be placed in a situation where all image recognition functions are activated while driving. After all drivers have used all functions, the warning history analysis unit 22 of the management device 1 analyzes the warning occurrence history (e.g., the number of warnings) that occurs for each function per operation, and the analysis result output unit 23 outputs the analysis result by the warning history analysis unit 22. This analysis allows the user to understand the individual tendencies of each driver, for example, the frequency with which dangerous driving occurs when each function is activated.

For example, a high number of warnings in the warning history means that the driver often drives dangerously while the function is in operation. The user can notify, for example, the operation manager, of the functions that are determined to have a high number of warnings as a result of the analysis as the driver's "weakness." After extracting the "weakness," the vehicle-mounted device 3 automatically sets the image recognition function to be the driver's "weakness" when it recognizes the driver's information. Prior to this, for example, the setting file creation unit 13 of the management device 1 sets specific information that identifies the selected library in the setting file based on the analysis results by the warning history analysis unit 22. In this way, it becomes possible to set an effective image recognition function that is in line with the tendencies of each individual driver, even with a limited number of functions that can be operated simultaneously.

In addition, the operations manager can grasp the trends of image recognition warnings for each driver and for the entire business. By effectively utilizing the image recognition function installed in the vehicle-mounted device 3, it becomes possible to provide driving support that is close to the driver.

[Procedure for executing driving assistance method]
The following describes the execution procedure of the driving assistance method of the driving assistance system 100 described above. Fig. 5 is a flow chart explaining the procedure of advance setting. First, the user sets the functions to be used for each group as shown in Fig. 2 (step S1). Depending on the request of the operator, which is the user, the staff of the dealer of the in-vehicle device 3 may perform the setting work.

The user can make a selection from the screen of the function selection unit 11, for example, as shown in FIG. 4A. The user can check and select at least one function using an input device such as a mouse. If the user selects too many functions and exceeds the processing capacity of the arithmetic processing unit, the management unit 12 may display an error message on the LCD display or sound an error message through the speaker.

The user can also make a selection from the screen of the function selection unit 11, for example, as shown in FIG. 4B. The user can check and select one group using an input device such as a mouse. After selecting a group, the user can also replace some of the functions within the group; such settings are called "user custom" settings.

Next, the user sets the usage group to be used for each operation day, as shown in FIG. 3 (step S2). However, the settings that were previously set in the in-vehicle device 3 may be used. Alternatively, the operation manager or the vehicle driver may set the groups. If the driver sets the groups, the display of the in-vehicle device 3 displays a query message asking, "Which function would you like to set today?", and the driver selects using the button on the in-vehicle device 3.

When the driver sets the functions, previously set functions may be deleted from the options as time passes, encouraging the driver to use unused functions. The arithmetic processing unit of the management device 1 may also set the functions randomly, but depending on how the computer generates random numbers, there may be a bias in the number of times each function is selected.

After the user has set up the library as described above, the setting file creation unit 13 creates a setting file for enabling the selected library. That is, steps S1 and S2 correspond to the steps in which the management device 1 accepts the user's selection of the library to be operated from among multiple libraries. The user then installs the setting file created in steps S1 and S2 into the vehicle-mounted device 3 (step S3). This step corresponds to the step of outputting the setting file created by the setting file creation unit 13 to the vehicle-mounted device. For example, the setting file is prepared in an operation card, and the file is installed by card loading.

The management device 1 may also be installed via an Internet line (wireless LAN, LTE, etc.) using a PC, smartphone, tablet, etc., which is a device that operates the management device 1. The management device 1 may also be installed using short-range wireless communication such as Bluetooth (registered trademark) using these devices. The management device 1 may also be installed by displaying a two-dimensional code on these devices and reading it with the camera of the vehicle-mounted device 3.

Figure 6 is a flow chart explaining the operation during the vehicle's departure from the garage. After the driver leaves the garage (step S11), the image recognition control unit 32 of the vehicle-mounted device 3 operates one of the multiple libraries that is enabled by the information in the configuration file output from the management device 1. That is, the image recognition control unit 32 operates one of the multiple libraries that is selected in the management device 1 according to the selection in the management device 1. Furthermore, the alarm history measurement unit 34 of the vehicle-mounted device 3 measures and accumulates the alarm occurrence history for each function that has been operated (step S12). After completing the work, the driver returns the vehicle to the garage (step S13).

In step S12, the alarm occurrence history may be recorded in the internal memory (storage unit 33) of the vehicle-mounted device 3, and stored in a format that links the driver's crew code and the alarm occurrence history. If a driving card is used, it may be recorded on the driving card, and the result is also reflected in the daily report. It may also be recorded on a web server on the network. By recording the alarm occurrence history on the web server, it becomes possible for an operation manager at a remote location to understand the alarm occurrence history.

Figure 7 is a flowchart explaining the procedure of post-analysis. When the management device 1 receives the alarm occurrence history from the analysis result output unit 23 of the vehicle-mounted device 3, the alarm history analysis unit 22 checks whether the alarm history measurement has been performed for all functions (step S21). If not all functions have been performed, the process is temporarily terminated until all functions have been performed (No in step S21). If all functions have been performed (Yes in step S21), the alarm history analysis unit 22 analyzes the alarm history for each function (step S22). The analysis is performed, for example, based on a combination of the population parameter x judgment method shown below.

The population may include, for example, all users within a business establishment, company, or company that owns the vehicle-mounted device 3 of that model.

The following are examples of judgment methods. The following mainly focuses on the number of warnings in the warning occurrence history.
The number of warnings generated by the image recognition function per operation is compared with the average expected in the market to determine whether it is higher or lower.
- The operation manager determines the threshold for the number of warnings generated by the image recognition function per operation, and judges whether the number of warnings is high or low by comparing it with this threshold. Depending on the threshold, there may be a large number of people who receive a high number of warnings, or the opposite phenomenon may also occur.
The number of warnings generated by the image recognition function per operation is compiled for each parameter. From the compilation results, those who are in the worst n% of the parameter are treated as having a "high number of warnings" (n is a variable setting).
The number of warnings generated by the image recognition function per operation is compiled for each parameter. The average value for each parameter is calculated from the compiled results. It is judged whether the number is more or less than the average value.
The number of warnings generated by the image recognition function per operation is compiled for each parameter. From the compiled results, a standard deviation is calculated for each parameter. The standard deviation is used to judge whether the result is good or bad.

Once the analysis results are obtained, the analysis result output unit 23 outputs the results to a display or the like, and can notify, for example, the operations manager of the driver's weaknesses (step S23). The user of the management device 1 may also enter the results in a daily report, and the operations manager can know the results when the daily report is submitted. The analysis result output unit 23 may also display the results via separately prepared analysis software. The operations manager can know the results while the analysis software app is running.

Figure 8 is a flow chart explaining the procedure for reflecting the analysis. Functions related to the driver's weaknesses are set (step S31). For example, the warning history analysis unit 22 may automatically set this in the setting file creation unit 13. Furthermore, if the operations manager who has seen the output of the analysis result output unit 23 wishes, the analysis results may be reflected in the settings regardless of the driver's wishes. Furthermore, the weaknesses may be presented to the driver, and if the driver wishes, they may be reflected in the settings. Furthermore, the weaknesses may be presented to both the operations manager and the driver, and if both wish, they may be reflected in the settings.

[Image recognition function output settings]
As described above, the driving assistance system 100 can appropriately set the image recognition function of the in-vehicle device 3. Furthermore, the driving assistance system 100 can turn off output settings related to image recognition functions that are not selected by the function selection unit 11, either automatically or through a setting input by the user. For example, if the forward vehicle approach warning function is not selected, no warning will be issued by the speaker even if the default setting for audio output by the speaker is set to on.

For example, the user may input the output off setting at the same time as the input in FIG. 4A, or the management unit 12 of the management device 1 may set this setting. This setting is sent directly to the vehicle-mounted device 3, and the image recognition control unit 32 can turn off the output settings related to the image recognition functions that were not selected by the function selection unit 11. This makes it possible to prevent confusion caused by the output of image recognition functions that were not selected.

Note that the output method related to the image recognition function is not limited to audio. For example, when activating the lane departure detection function that detects lane departure (lane protrusion) of the vehicle, output can be provided by turning on and off the turn signal. Also, when activating the person detection function that detects people, output can be provided by a reverse signal or a brake signal. When activating the speed detection function that detects excessive speed, output can be provided by the speed signal from the speedometer.

In the case of vehicle distance detection, the output setting can be switched on or off depending on whether the vehicle is on an ordinary road or an expressway. Furthermore, depending on the degree of approach, the output setting can be switched on or off depending on whether the vehicle ahead is gradually approaching or suddenly approaching. In this way, even with a single selected image recognition function, the output setting can be switched on or off in detail.

Here, the features of the embodiments of the driving assistance system, driving assistance method, and driving assistance program according to the present invention described above are briefly summarized and listed below in [1] to [9].

[1] A driving assistance system (100) comprising an on-board device (3) mounted on a vehicle and a management device (1) outside the vehicle,
The vehicle-mounted device,
An image recognition control unit (32);
A storage unit (33) for storing a plurality of programs for operating a plurality of functions related to image recognition,
each of the plurality of programs corresponds to each of the plurality of functions;
The management device,
a function selection unit (11) for selecting a program to be operated from among the plurality of programs;
The image recognition control unit (32) operates a program selected by the function selection unit among the plurality of programs in response to a selection made by the function selection unit.
Driving assistance system.

According to the driving assistance system of the configuration [1] above, the management device sets the programs to be operated in the vehicle-mounted device, so that it is possible to effectively use multiple image recognition functions while reducing the load on the processing unit of the vehicle-mounted device.

[2] The management device has a management unit (12) that monitors whether the program selected by the function selection unit exceeds the processing capacity of a processing unit that processes the vehicle-mounted device.
The driving assistance system according to [1] above.

According to the driving assistance system of the configuration [2] above, the management unit monitors whether the program selected by the function selection unit exceeds the processing capacity of the arithmetic processing unit that is responsible for the main control processing of the vehicle-mounted device. This makes it possible to prevent a program that exceeds the processing capacity of the arithmetic processing unit from being selected.

[3] In the storage unit, a plurality of groups each including a program corresponding to one or more functions are set in advance,
the function selection unit is configured to be able to select at least one of the groups,
The driving assistance system according to any one of claims 1 to 2.

According to the driving assistance system of the above configuration [3], the user can select a group including appropriate functions depending on the driving situation, the driver, etc.

[4] The image recognition control unit does not operate any of the programs that has not been selected by the function selection unit.
A driving assistance system according to any one of the above [1] to [3].

According to the driving assistance system configured as described above in [4], the program corresponding to the image recognition function that was not selected does not operate, thereby reducing the load on the processing unit.

[5] The management device includes:
a setting file creation unit (13) that creates a setting file for enabling the program selected by the function selection unit;
The image recognition control unit (32) operates the program selected by the function selection unit in accordance with the setting file.
A driving assistance system according to any one of [1] to [4] above.

According to the driving assistance system having the configuration of [5] above, the settings of the program selected as the operation target in the function selection unit of the management device can be reflected in the in-vehicle device via the setting file.

[6] A driving assistance method using a driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device has a plurality of programs for operating a plurality of functions,
each of the plurality of programs corresponds to each of the plurality of functions;
In the management device, a user's selection of a program to be operated from among the plurality of programs is accepted (steps S1 and S2);
The vehicle-mounted device operates the program selected by the management device among the plurality of programs in response to the selection by the management device (step S12).
Driving assistance methods.

According to the driving assistance method configured as above in [6], the management device sets the program to be operated in the vehicle-mounted device, so that it is possible to effectively use multiple image recognition functions while reducing the load on the processing unit of the vehicle-mounted device.

[7] A driving assistance program for executing driving assistance by a driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device has a plurality of programs for operating a plurality of functions,
each of the plurality of programs corresponds to each of the plurality of functions;
On the computer,
a step of receiving a user's selection of a program to be operated from among the plurality of programs in the management device (steps S1 and S2);
a step (step S12) of the in-vehicle device operating a program selected by the management device among the plurality of programs in response to a selection by the management device;
A driving assistance program that executes the following:

According to the driving assistance program configured as above in [7], the management device sets the program to be operated in the vehicle-mounted device, so that it is possible to effectively use multiple image recognition functions while reducing the load on the processing unit of the vehicle-mounted device.

[8] In the storage unit, a plurality of groups each including a program corresponding to one or more functions are set in advance,
The function selection unit is configured to be able to select one of the groups depending on the operation day on which the vehicle operates.
The driving assistance system according to any one of claims 1 to 2.

According to the driving assistance system of the configuration [8] above, the user can select a group that includes appropriate functions depending on the characteristics of the operation day.

[9] The vehicle-mounted device is
a warning history measurement unit (34) for measuring a warning occurrence history that occurs during the operation of each program for each driver, and a warning history output unit (35) for outputting the warning occurrence history to the management device,
The management device has an alarm history analysis unit that analyzes the alarm occurrence history,
The setting file creation unit creates the setting file based on an analysis result by the alarm history analysis unit.
The driving assistance system according to [5] above.

According to the driving assistance system configured as described above in [9], it is possible to set effective image recognition functions that are in line with the tendencies of each individual driver, even though there is a limit to the number of functions that can be operated simultaneously.

REFERENCE SIGNS LIST 1 Management device 3 Vehicle-mounted device 10 Pre-setting unit 11 Function selection unit 12 Management unit 13 Setting file creation unit 20 Post-analysis unit 21 Alarm history input unit 22 Alarm history analysis unit 23 Analysis result output unit 31 Setting file input unit 32 Image recognition control unit 33 Memory unit 34 Alarm history measurement unit 35 Alarm history output unit 100 Driving assistance system

Claims (7)

A driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device includes:
An image recognition control unit;
A storage unit that stores a plurality of programs for operating a plurality of functions related to image recognition,
each of the plurality of programs corresponds to each of the plurality of functions;
The management device includes:
a function selection unit for selecting a program to be operated from among the plurality of programs,
the image recognition control unit operates a program selected by the function selection unit among the plurality of programs in response to a selection made by the function selection unit;
Driving assistance system. the management device has a management unit that monitors whether or not the program selected by the function selection unit exceeds the processing capacity of a processing unit that processes the vehicle-mounted device;
The driving assistance system according to claim 1 . A plurality of groups each including a program corresponding to one or more functions are set in advance in the storage unit,
the function selection unit is configured to be able to select at least one of the groups,
The driving assistance system according to claim 1 . the image recognition control unit does not operate any of the programs that has not been selected by the function selection unit,
The driving assistance system according to claim 1 . The management device includes:
a setting file creation unit that creates a setting file for enabling the program selected by the function selection unit;
the image recognition control unit operates the program selected by the function selection unit in accordance with the setting file;
A driving assistance system according to any one of claims 1 to 4. A driving assistance method using a driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device has a plurality of programs for operating a plurality of functions,
each of the plurality of programs corresponds to each of the plurality of functions;
receiving, in the management device, a selection by a user of a program to be operated from among the plurality of programs;
the vehicle-mounted device operates a program selected by the management device among the plurality of programs in response to a selection by the management device;
Driving assistance methods. A driving assistance program for executing driving assistance by a driving assistance system including an on-board device mounted in a vehicle and a management device outside the vehicle,
The vehicle-mounted device has a plurality of programs for operating a plurality of functions,
each of the plurality of programs corresponds to each of the plurality of functions;
On the computer,
receiving, by the management device, a selection by a user of a program to be operated from among the plurality of programs;
a step of causing the vehicle-mounted device to operate a program selected by the management device among the plurality of programs in response to a selection made by the management device;
A driving assistance program that executes the following:

Images