Automobile intelligent terminal alarm test method and system

By converting standard videos into simulated videos, replacing real-person simulates illegal driving behaviors, identifying and evaluating the alarm timeliness of smart terminals, the problems of low testing efficiency and inability to evaluate alarm timeliness in the existing technology are solved, and more efficient testing is achieved and alarm accuracy.

CN119068635APending Publication Date: 2024-12-03CHONGQING VEHICLE TEST & RES INST CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411025520.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

In the prior art, smart terminals need real people to simulate illegal driving behavior during testing, resulting in low testing efficiency and inability to effectively evaluate the timeliness of alarms.

Method used

By converting standard videos into analog videos, using converters and detection modules to replace real-person analog driving behaviors, identify dangerous actions in the videos, and evaluate the interval between smart terminals from identification to alarms, calculate false alarm rates and missed alarm rates to evaluate the Timely alarm.

Benefits of technology

It improves the efficiency of smart terminal testing, can accurately evaluate the timeliness and accuracy of alarms, and reduces the dependence of manual simulation tests.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068635A_ABST
    Figure CN119068635A_ABST
Patent Text Reader

Abstract

The invention provides an automobile intelligent terminal alarm test method and system, and the method comprises the steps: S1, selecting a standard video in a standard library, converting the standard video into an analog video through a converter, and transmitting the analog video to an intelligent terminal; s2, the detection module identifies the standard video to obtain a first action type and a first moment, and the intelligent terminal identifies the analog video to obtain a second action type; s3, the intelligent terminal sends the second action type to the detection module, and the detection module records a second moment at the moment after receiving the second action type; and S4, the detection module determines a detection result according to the first action type, the second action type, the first moment and the second moment. The problems that in the prior art, when an intelligent terminal is tested, a real person needs to make illegal driving actions to trigger corresponding alarms, the testing efficiency is low, and the alarm timeliness of the intelligent terminal is not tested are solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of equipment detection, and in particular to an automobile intelligent terminal alarm testing method and system. Background Art

[0002] With the increasing demand for the supervision of "two passenger and one dangerous goods" vehicles and drivers by local transport management bureaus, and the increasingly complex driving environment, the traditional 2G network positioning vehicle-mounted terminal can no longer meet the needs of the market environment. In 2017, Jiangsu Province issued Su Biao T / JSATL 13-2017, announcing that the vehicle-mounted terminal has entered the era of 4G video surveillance. At the same time, the vehicle-mounted terminal was officially renamed the Active Safety Intelligent Prevention and Control System (hereinafter referred to as "smart terminal"). With the introduction of monitoring cameras such as DSM, ADAS, and DVR, the occurrence of accidents is effectively reduced by warning the driver's irregular driving behavior and the dangerous driving status of the vehicle, which brings convenience to the supervision of the "two passenger and one dangerous goods" vehicle industry.

[0003] After the monitoring camera captures the driver's actions as images or records them as videos, they are sent to the smart terminal. The smart terminal identifies whether the actions in the images or videos are dangerous actions, and then issues an alarm. In order to improve the smart terminal's recognition rate of dangerous actions, a large number of different action tests are usually carried out before sale. Since the smart terminal directly uses the monitoring camera as action images or videos as input, the test is still basically at the stage of real-car testing where the tester simulates illegal driving behaviors. Real people are required to perform illegal driving actions to trigger corresponding alarms, and the test efficiency is low. At the same time, during the test, the quality of the smart terminal is mostly judged based on the correctness of the test results. The interval time from the smart terminal's recognition of dangerous actions to the issuance of an alarm is not judged, and the timeliness of the smart terminal's alarm is not tested. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for testing the alarm of an automobile intelligent terminal, which solves the problems in the prior art that, during testing of the intelligent terminal, a real person is required to perform illegal driving actions to trigger a corresponding alarm, the testing efficiency is low, and the alarm timeliness of the intelligent terminal is not tested.

[0005] According to an embodiment of the present invention, a method for testing an alarm of an automobile intelligent terminal includes:

[0006] S1: select a standard video from the standard library, convert the standard video into an analog video using a converter, and send it to the smart terminal;

[0007] S2: The detection module identifies the standard video to obtain the first action type and the first moment, and the intelligent terminal identifies the analog video to obtain the second action type;

[0008] S3: The intelligent terminal sends the second action type to the detection module, and the detection module records the second moment after receiving the second action type;

[0009] S4: The detection module compares the first action type and the second action type to obtain a detection result, wherein the detection result includes a correct action judgment and an incorrect action judgment. If the first action type and the second action type are the same, the action judgment is judged to be correct; if the first action type and the second action type are different, the action judgment is judged to be incorrect; the ratio of the number of times the action judgment is judged to be incorrect to the total number of standard videos is calculated to obtain the false alarm rate.

[0010] Preferably, it also includes:

[0011] S5: Replace the standard video and repeat steps S1-S4 until all standard videos in the standard library are converted into analog videos and corresponding detection results are obtained.

[0012] Preferably, the file name of the standard video includes the first action type and the action completion time;

[0013] The detection module identifies the standard video, and the steps of obtaining the first action type and the first moment include:

[0014] Identify the file name of the standard video and obtain the first action type and the action completion time;

[0015] Calculate the sum of the current system time and the action completion time to get the first time.

[0016] Preferably, the detection result also includes timely alarm and delayed alarm;

[0017] Calculate the difference between the first moment and the second moment to obtain the alarm delay, and compare the alarm delay with a preset value;

[0018] If the alarm delay is greater than the preset value, it is considered as an alarm delay;

[0019] If the alarm delay is less than or equal to the preset value, the alarm is judged to be timely.

[0020] The ratio of the number of times judged as alarm delay to the total number of standard videos is calculated to obtain the missed alarm rate.

[0021] Preferably, if after comparing the first action type and the second action type, it is determined that the action is wrongly determined, the alarm delay is not calculated and it is directly determined as an alarm delay.

[0022] On the other hand, according to an embodiment of the present invention, there is also provided a vehicle intelligent terminal alarm test system, which uses the above-mentioned vehicle intelligent terminal alarm test method and comprises:

[0023] A standard library, wherein the standard library is used to store standard videos;

[0024] The test module includes a converter, a detection module and a communication module. The converter is electrically connected to the standard library and the smart terminal, and is used to convert the standard video into an analog video and transmit the analog video to the smart terminal. The communication module is communicatively connected to the smart terminal, and the detection module is electrically connected to the standard library and the communication module.

[0025] Preferably, the converter is provided with an AHD interface and a BNC interface, the AHD interface is connected to the standard library, and the BNC interface is connected to the DSM interface of the intelligent terminal.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] A converter is used to convert standard video into analog video and transmit it to the smart terminal to replace the image data generated by monitoring cameras such as DSM. There is no need for real people to simulate illegal driving behavior and trigger corresponding alarms by making illegal driving actions. At the same time, the detection module identifies the moment when the dangerous action occurs in the standard video and the moment when the second action type is received by the smart terminal to determine whether the interval time from the smart terminal identifying the dangerous action to the alarm is met, so as to test the timeliness of the alarm of the smart terminal. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 4 is a diagram of the test system architecture of an embodiment of the present invention.

[0029] Figure 2 The figure is a flow chart of a testing method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0031] like Figure 1 As shown, the embodiment of the present invention proposes an automobile intelligent terminal alarm test system, including:

[0032] Standard library, which is essentially a database for storing standard videos, is set up in an external computer terminal;

[0033] The test module includes a converter, a detection module and a communication module. The converter is provided with an AHD interface and a BNC interface. The AHD interface is connected to the standard library through an HDMI-AHD data cable to identify the standard video in the standard library. The BNC interface is connected to the DSM camera data input aviation head interface of the smart terminal through a data cable. The converter converts the standard video format into AHD simulated high definition in real time to replace the video acquisition input of the DSM camera, reproduces the picture taken by the DSM camera and transmits it to the smart terminal, thereby replacing the image data generated by the DSM and other monitoring cameras, so that there is no need for real people to simulate illegal driving behavior and make illegal driving actions to trigger corresponding alarms.

[0034] The communication module is connected to the intelligent terminal for communication (RS232 communication), and the detection module is electrically connected to the standard library and the communication module.

[0035] On the other hand, Figure 2 As shown, the embodiment of the present invention also provides a method for testing an alarm of an intelligent automobile terminal, which includes:

[0036] S1: select a standard video from the standard library, convert the standard video into an analog video using a converter, and send it to the smart terminal;

[0037] The common vehicle-mounted video intelligent terminal equipment on the market is set up in the cab of the "two passenger and one dangerous goods" vehicle, the DSM monitoring camera is adjusted to face the driver's position, the video recording resolution and frame rate are set to meet the standard requirements, and multiple drivers are recorded in all working conditions (including daytime, night, front light, side light, backlight, alternating flashes of sunlight under tree shade, vehicle vibration, wearing hats, glasses, sunglasses, masks, and completing illegal driving behaviors such as fatigue, smoking, making phone calls, playing with mobile phones, distraction, driver abnormality, equipment failure, both hands off the steering wheel, and not wearing seat belts). At the same time, some daily driving habits of drivers are recorded (such as normal driving, picking ears, picking noses, eating lollipops, chewing things, etc.), and the video library includes male and female drivers; after recording, the video is exported, and the video editing software is used to cut and combine the video into multiple 1-minute clips according to the alarm type, each clip is a standard video, and the illegal driving behavior (i.e., the first action type) and the action completion time at this time in each standard video are found through the video editing software, and the corresponding standard video is named in the form of the first action type + the action completion time.

[0038] Afterwards, they are grouped according to violations and normal driving behavior habits to form a test missed alarm and false alarm scenario library, and establish diversified simulation video scene materials; at the same time, some representative clips that do not participate in the test are selected as training videos to verify the entire test system.

[0039] After that, all standard videos are stored in the standard library. When testing the smart terminal, the BNC interface of the converter is connected to the DSM data input interface of the smart terminal, the communication module is connected to the signal acquisition board in the smart terminal, and then the detection module and the converter are started at the same time. The detection module and the converter read the standard videos in the standard library according to the file arrangement order.

[0040] A converter is used to convert standard video into analog video and transmit it to a smart terminal, replacing the image data generated by monitoring cameras such as DSM. This eliminates the need for real people to simulate illegal driving behaviors and trigger corresponding alarms by making illegal driving actions.

[0041] S2: The detection module identifies the standard video to obtain the first action type and the first moment, and the intelligent terminal identifies the analog video to obtain the second action type;

[0042] The detection module recognizes the standard video. It only needs to recognize the file name of the standard video, extract the first action type and the action completion time, and then calculate the sum of the current system time and the action completion time to obtain the first time.

[0043] The intelligent terminal itself has the image recognition capability to recognize the simulated video, identify the action in the simulated video and make a judgment to determine whether it is a dangerous action, and use the action as the second action type.

[0044] Both the first action type and the second action type include normal actions and dangerous actions.

[0045] S3: The intelligent terminal sends the second action type to the detection module, and the detection module records the second moment after receiving the second action type;

[0046] S4: The detection module compares the first action type with the second action type to obtain a detection result.

[0047] The detection results include correct action judgment, incorrect action judgment, timely alarm and delayed alarm;

[0048] In action judgment, if the first action type and the second action type are the same, the action judgment is judged to be correct;

[0049] If the first action type and the second action type are different, it is judged that the action judgment is wrong;

[0050] After comparing the first action type and the second action type, the difference between the first moment and the second moment is calculated to obtain the alarm delay, and the alarm delay is compared with the preset value. When calculating the alarm delay, two errors need to be added, namely the video frame interval (±0.03s) and the RS232 communication speed (±0.075s), that is, the alarm delay = the difference between the first moment and the second moment + the video frame interval + the RS232 communication speed;

[0051] If the alarm delay is greater than the preset value, it is considered as an alarm delay;

[0052] If the alarm delay is less than or equal to the preset value, the alarm is judged to be timely.

[0053] In addition, if after comparing the first action type and the second action type, it is determined that the action is wrong, the alarm delay is not calculated and it is directly determined as an alarm delay.

[0054] The detection module determines whether the interval time from the smart terminal identifying the dangerous action to the alarm issuing meets the requirements by identifying the moment when the dangerous action occurs in the standard video and the moment when the second action type is received by the smart terminal, so as to test the timeliness of the alarm of the smart terminal.

[0055] S5: Replace the standard video and repeat steps S1-S4 until all standard videos in the standard library are converted into analog videos and corresponding detection results are obtained;

[0056] According to the detection results, the false alarm rate and the missed alarm rate are obtained respectively.

[0057] Identify and judge all standard videos in the standard library to obtain corresponding detection results.

[0058] The false alarm rate is calculated by calculating the ratio of the number of incorrect action judgments to the total number of standard videos. The false alarm rate reflects the test of the intelligent terminal's action recognition ability. If the driver's action is misjudged, false alarms will appear everywhere, and false alarms will affect the driver's normal driving.

[0059] The ratio of the number of times the alarm is delayed to the total number of standard videos is calculated to obtain the missed alarm rate. The missed alarm rate mainly reflects the timeliness of the alarm of the smart terminal. Different smart terminals have different internal signal transmission speeds, so the interval time from identifying dangerous actions to issuing an alarm will also be different. However, in actual use, after the driver makes an illegal action (dangerous action), it still takes a short time to respond after receiving the alarm from the smart terminal. In order to avoid possible driving risks, the only way is to shorten the interval time from identifying dangerous actions to issuing an alarm as much as possible.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A method for testing an alarm of an automobile intelligent terminal, characterized in that: include: S1: select a standard video from the standard library, convert the standard video into an analog video using a converter, and send it to the smart terminal; S2: The detection module identifies the standard video to obtain the first action type and the first moment, and the intelligent terminal identifies the analog video to obtain the second action type; S3: The intelligent terminal sends the second action type to the detection module, and the detection module records the second moment after receiving the second action type; S4: The detection module compares the first action type and the second action type to obtain a detection result, wherein the detection result includes a correct action judgment and an incorrect action judgment. If the first action type and the second action type are the same, the action judgment is judged to be correct; if the first action type and the second action type are different, the action judgment is judged to be incorrect; the ratio of the number of times the action judgment is judged to be incorrect to the total number of standard videos is calculated to obtain the false alarm rate.

2. A vehicle intelligent terminal alarm test method as claimed in claim 1, characterized in that: Also includes: S5: Replace the standard video and repeat steps S1-S4 until all standard videos in the standard library are converted into analog videos and corresponding detection results are obtained.

3. A vehicle intelligent terminal alarm test method as claimed in claim 1, characterized in that: The file name of the standard video includes the first action type and the action completion time; The detection module identifies the standard video, and the steps of obtaining the first action type and the first moment include: Identify the file name of the standard video and obtain the first action type and the action completion time; Calculate the sum of the current system time and the action completion time to get the first time.

4. A method for testing an automobile intelligent terminal alarm as claimed in claim 1, characterized in that: The detection results also include timely alarm and delayed alarm; Calculate the difference between the first moment and the second moment to obtain the alarm delay, and compare the alarm delay with a preset value; If the alarm delay is greater than the preset value, it is considered as an alarm delay; If the alarm delay is less than or equal to the preset value, the alarm is judged to be timely. The ratio of the number of times judged as alarm delay to the total number of standard videos is calculated to obtain the missed alarm rate.

5. A method for testing an automobile intelligent terminal alarm as claimed in claim 4, characterized in that: If, after comparing the first action type and the second action type, it is determined that the action is wrong, the alarm delay is not calculated and it is directly determined as an alarm delay.

6. An automobile intelligent terminal alarm test system, characterized in that: The system uses the automobile intelligent terminal alarm test method according to any one of claims 1 to 5, comprising: A standard library, wherein the standard library is used to store standard videos; The test module includes a converter, a detection module and a communication module. The converter is electrically connected to the standard library and the smart terminal, and is used to convert the standard video into an analog video and transmit the analog video to the smart terminal. The communication module is communicatively connected to the smart terminal, and the detection module is electrically connected to the standard library and the communication module.

7. The automobile intelligent terminal alarm test system as claimed in claim 6, characterized in that: The converter is provided with an AHD interface and a BNC interface, the AHD interface is connected to the standard library, and the BNC interface is connected to the DSM interface of the intelligent terminal.