Traffic warning system, computer readable medium, and operation method therefor

Abstract

To provide a traffic warning system operated according to a reliability level of a radar sensor and a visual sensor.SOLUTION: A traffic warning system comprises a visual sensor, a radar sensor, and a processor. The visual sensor is constituted so as to detect visual sensing data about each of at least one object. The radar sensor is constituted so as to detect radar sensing data about each of the one or more objects. The processor is constituted to: acquire a radar reliability level of radar sensing data by analyzing the radar sensing data; acquire a visual reliability level of the visual sensing data by analyzing the visual sensing data; and acquire a position and a velocity of at least one object by analyzing the visual sensing data and the radar sensing data, when both of the radar reliability level and the visual reliability level each exceed a predetermined value.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a traffic warning system, a computer readable medium, and a method of operation thereof. [Background technology]

[0002] With the increase in the number and types of vehicles, the probability of traffic accidents is also increasing year by year. Obviously, in addition to the continuous advancement of transportation technology in power management systems, improving driving safety has also become an important issue. For example, on a bicycle's route, there are many objects that riders should pay attention to, such as traffic signs, pedestrians, and obstacles. Riders need to pay attention to the road environment as the basis for their next driving actions.

[0003] In modern society, traffic accidents are common when occupants do not pay attention to road conditions. Due to fatigue or distraction, occupants may not notice traffic signs or obstacles on the road and therefore may not be able to perform appropriate driving maneuvers in response to traffic signs or obstacles in front of them.

[0004] In U.S. Patent No. 10,668,971, the bicycle safety device preferably measures both the lateral distance and speed of an overtaking vehicle to the bicycle frame and calculates a driving safety rating based on predetermined safety thresholds. The rating is uploaded to a remote server along with video evidence. The device may have a highly distinctive light display / effect (which also functions as the bicycle's rear light) so that drivers will recognize the device and only react by overtaking if it is safe to do so.

[0005] The invention in Chinese Patent Application Publication No. 114559960 relates to the technical field of autonomous driving of intelligent vehicles and discloses a collision early warning system based on the fusion of a front-view camera and a rear millimeter-wave radar, the system comprising: a visual detection module for capturing and processing images of road conditions ahead of the intelligent vehicle to obtain information on road conditions ahead; a rear millimeter-wave radar for enabling a user to obtain information on obstacles behind the intelligent vehicle, the rear obstacle information including the obstacle's speed; a data processing module for processing the road condition information and the rear obstacle information; and a rear early warning processing unit for obtaining a collision time according to the relative speed in the y direction between the rear obstacle and the intelligent vehicle, and issuing a warning when the collision time reaches a rear collision threshold. The early warning function ensures the safe driving of intelligent vehicles in accordance with this scheme, greatly reduces the possibility of vehicle collisions, and improves the driving safety of intelligent vehicles. Summary of the Invention

[0006] In one embodiment of the present invention, a traffic warning system is provided, the traffic warning system comprising: a visual sensor, a radar sensor, and a processor. The visual sensor is configured to detect visual sensing data of each of at least one object. The radar sensor is configured to detect radar sensing data of each of the at least one object. The processor is configured to analyze the radar sensing data to obtain a radar confidence level of the radar sensing data, analyze the visual sensing data to obtain a visual confidence level of the visual sensing data, and, if both the radar confidence level and the visual confidence level are above a predetermined value, analyze the visual sensing data and the radar sensing data to obtain a position and velocity of the at least one object.

[0007] In another embodiment of the present invention, a method for operating a traffic warning system includes the following steps: detecting visual sensing data of each of at least one object by a visual sensor; detecting radar sensing data of each of the at least one object by a radar sensor; analyzing the radar sensing data by a processor to obtain a radar confidence level of the radar sensing data; analyzing the visual sensing data by a processor to obtain a visual confidence level of the visual sensing data; and if both the radar confidence level and the visual confidence level are above a predetermined value, analyzing the visual sensing data and the radar sensing data by the processor to obtain a position and velocity of the at least one object.

[0008] In one embodiment of the present invention, a computer-readable medium is provided having stored thereon instructions executable by at least one processor of a traffic warning system to cause the traffic warning system to implement a method for detecting visual sensing data of each of at least one object, detecting radar sensing data of each of the at least one object, analyzing the radar sensing data to obtain a radar confidence level of the radar sensing, analyzing the visual sensing data to obtain a visual confidence level of the visual sensing data, and, if both the radar confidence level and the visual confidence level are above a predetermined value, analyzing the visual sensing data and the radar sensing data to obtain a position and velocity of the at least one object.

[0009] Numerous objects, features, and advantages of the present invention will become readily apparent from the following detailed description of the embodiments of the invention, taken in conjunction with the accompanying drawings, although the various embodiments and drawings used herein are for illustrative purposes only and should not be construed as limiting the invention.

[0010] The above objects and advantages of the present invention will be more readily understood by those skilled in the art after having reference to the following detailed description and accompanying drawings. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a schematic functional block diagram of a traffic warning system according to an embodiment of the present invention; [Figure 2A] 2 shows a flow chart of a method of operation of the traffic warning system of FIG. 1; [Figure 2B] 2 shows a flow chart of a method of operation of the traffic warning system of FIG. 1; [Figure 3] FIG. 1 is a schematic diagram of object recognition in different lanes by a vision-based algorithm in the present disclosure. [Figure 4] 1 is a schematic diagram illustrating the relationship between the relative speed of an object (horizontal axis) and the corresponding stopping distance (vertical axis) to a traffic warning system. [Figure 5] FIG. 1 is a schematic diagram of an impact zone and a threat zone in one embodiment. [Figure 6] 4 is a schematic diagram illustrating the relationship between relative velocity of an object (horizontal axis) and corresponding stopping distance (vertical axis) to a traffic warning system according to another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] The present invention provides a traffic warning system that integrates two types of sensors, for example, a radar or microwave sensor and a visual sensor, and utilizes the advantages of both, while significantly reducing the power consumption required for system operation through an implementable operating method and control logic. This system is particularly applicable to various mobile light transportation vehicles, such as bicycles, electric bicycles (e-bikes), and other multi-wheeled personal transportation means, to improve the driving safety of passengers and timely output warnings to help passengers avoid potential dangers from vehicles approaching from behind or threats in their blind spots. The detection results of such emergencies can be detected throughout the system, and information that should be reflected first can be provided to passengers in a timely manner, thereby achieving beneficial effects such as power saving, low interference, and accurate warnings.

[0013] Referring to Figure 1, Figure 1 shows a functional block diagram of a traffic warning system 100 according to one embodiment of the present invention. The traffic warning system 100 may be mounted on a vehicle (e.g., a bicycle, an electric bicycle, or other multi-wheeled personal transportation means).

[0014] 1, the traffic warning system 100 includes a visual sensor 110, a radar sensor 120, and a processor 130. The visual sensor 110 receives visual sensing data D of each of at least one object 10. V The radar sensor 120 is configured to detect radar sensing data D of each of the at least one object 10. R The processor 130 is configured to detect the radar sensing data D R By analyzing the radar sensing data, the radar confidence level L R and visual sensing data D V By analyzing the visual sensing data D V The visual confidence level L V and obtain the radar confidence level L R and the visual confidence level L V If both of the visual sensing data DV and radar sensing data D R , to obtain the position and velocity of the object 10. In one embodiment, the visual sensor 110 and the radar sensor 120 can be used together to improve the accuracy of object detection.

[0015] The "visual sensor" herein may be implemented by an imaging device, such as a camera (e.g., a helmet-mounted GoPro camera, a wide-angle camera, an infrared camera, etc.). The traffic warning system 100 (e.g., the visual sensor 110 or the processor 130) may also recognize the object 10 using an AI (artificial intelligence) model. The AI ​​model may be acquired in advance using various appropriate machine learning techniques, such as deep learning. The object 10 may be, for example, an obstacle, a vehicle, or a pedestrian. The visual sensor 110 has the following characteristics: (1) a close and near sensing range, reducing blind spots within a 30-meter-long, 144-degree-wide field of view (FOV), and (2) visual detection capable of simultaneously detecting multiple targets (e.g., more than 10 target objects).

[0016] The radar sensor 120 is, for example, a 24G radar with the lowest power consumption, and various specifications are available, with or without lane recognition functionality. Generally, the power consumption of a visual sensor (e.g., 10 W) is much higher than that of a radar sensor, limiting its application range. In this embodiment, the traffic warning system integrates a radar sensor with a visual sensor to minimize power consumption and further enhance so-called "smart" functionality (i.e., lane recognition functionality with an adaptive power-saving scheme or mechanism). The radar (or microwave) sensor has a far and long sensing range and, for example, by using 24 GHz millimeter waves, has the characteristic of further extending coverage beyond a field of view of at least 100 meters (m) long and 8 m wide.

[0017] The processor 130 is electrically connected to the visual sensor 110 and the radar sensor 120 and may be configured to control the on / off of the sensors according to different traffic scenarios or environmental conditions and to process data sensed or generated by the sensors. The processor 130 may control the visual sensor 110 and the radar sensor 120 to implement the method of operation of the traffic warning system 100. The processor 130 may process the visual sensing data D V and obtains the position and size (of the object frame) of the object 10 based on the radar sensing data D R It is possible to obtain the position and velocity of the object 10 based on

[0018] 2A and 2B, which show sequential flow charts corresponding to different aspects of the method of operation of the traffic warning system 100 of FIG.

[0019] In step S110, the radar sensor 120 remains on. Furthermore, if no object is detected within the detection period, the radar sensor 120 may remain on and the visual sensor 110 may remain off to save power.

[0020] In step S120, the radar sensor 120 detects whether an object 10 has appeared around the traffic warning system 100. If no object 10 has appeared around the traffic warning system 100, the process proceeds to step S130A. If an object 10 has appeared around the traffic warning system 100, the process proceeds to step S140A.

[0021] In step S130A, the radar sensor 120 remains on and the visual sensor 110 remains off to save power.

[0022] In step S130B, the visual sensor 110 is turned on every sampling interval (e.g., 2-5 seconds) to detect whether an object 10 appears around the traffic warning system 100. In one embodiment, the visual sensor 110 may be turned on for a valid period (e.g., 2-5 seconds) every sampling interval. In one embodiment, the valid period may be shorter than, longer than, or equal to the sampling interval.

[0023] In step S140A, if an object is detected, the visual sensor 110 is turned on to enhance or improve the reliability of the detection result. The visual sensor 110 generates visual sensing data D V can be detected.

[0024] In steps S140B and S140C, the processor 130 performs a fusion procedure. In the fusion procedure, the processor 130 fusion-processes the radar sensing data D R By analyzing the radar sensing data D (e.g., based on reflected signal strength, time of flight, etc.), R Radar confidence level L R and acquiring visual sensing data D V By analyzing (for example, based on the captured image and the image processing results) visual sensing data D V The visual confidence level L V In one embodiment, the radar confidence level L R can be obtained using millimeter wave radar, for example, and the visual confidence level L V can be obtained using, for example, a CMOS (Complementary Metal Oxide Semiconductor) sensor camera, a QVGA (Quarter Video Graphics Array) sensor camera, and / or an infrared sensor camera.

[0025] In one embodiment, only one of the radar sensor and the visual sensor can be activated to provide more reliable detection information in certain situations. In the former case, because visual detection accuracy is significantly reduced in bad weather and / or nighttime conditions, the radar sensor alone can detect vehicles to further improve these adverse conditions. In the latter case, when an approaching object is outside the radar sensor's detection range, the visual sensor can detect vehicles in the radar's blind spot to further improve these deficiencies. Therefore, the above two types of sensors can be adaptively activated according to different conditions.

[0026] In step S140D, processor 130 calculates the radar confidence level L R and the visual confidence level L V The radar reliability level L is determined based on whether both of the above values ​​exceed a preset value. R and the visual confidence level L V If both of the radar confidence level L are greater than the preset value, the process proceeds to step S150A. R and visual confidence level L V If the difference is not greater than the preset value, the process proceeds to step S160. In one embodiment, the preset value may be, for example, in the range of 70% to 90% (e.g., 80%), but may be less or greater. The preset value may be an index value or percentage that reflects the quality or reliability of the data.

[0027] In step S150A, the processor 130 receives the visual sensing data D V and radar sensing data D R By analyzing the vectors 10, the position (eg, relative position) and velocity (eg, relative velocity) of the object 10 can be obtained.

[0028] In step S150B, the processor 130 may determine the number of at least one object 10 approaching the traffic warning system 100 (eg, the relative position and relative velocity of the object 10 are positive).

[0029] In step S160A, processor 130 calculates the radar confidence level L R is the visual confidence level L V Determine whether the radar reliability level L is higher than R is the visual confidence level L V If the radar confidence level L is higher than the threshold, the process proceeds to step S160B. R is the visual confidence level L V is not higher than (for example, visual confidence level L V is the radar reliability level L R If the value is higher than , the process proceeds to step S160C.

[0030] In step S160B, the processor 130 calculates the radar sensing data D R Based on (eg, location and speed), the number of objects 10 approaching the traffic warning system is determined.

[0031] In step S160C, the processor 130 calculates the visual sensing data D V (e.g., location) and radar sensing data D R Based on (eg, speed or speed estimates), the number of objects 10 approaching the traffic warning system is determined.

[0032] In step S170, the processor 130 determines whether the number of objects 10 is equal to or less than a preset number. If the number of objects 10 is equal to or less than the preset number, the process proceeds to step S180A. If the number of objects 10 is greater than the preset number, the process proceeds to step S180B. In one embodiment, the preset number is, for example, five, or a number less than or greater than five.

[0033] In step S180A, the processor 130 outputs a warning signal S1, which may be, for example, a sound, a vibration, an image, or a light.

[0034] In step S180B, the processor 130 outputs a warning signal S1 for the object 10 closest to the traffic warning system 100. For example, if the distance between the object 10 closest to the traffic warning system 100 and the traffic warning system 100 is equal to or less than a preset distance, the processor 130 outputs the warning signal S1.

[0035] Referring to FIG. 3, FIG. 3 shows a schematic diagram of object recognition in different traffic lanes by a vision-based algorithm in the present disclosure.

[0036] First, the processor 130 may obtain object frame 11 (or vehicle identification frame) information according to the size of the approaching object. Next, the processor 130 may calculate pixel sizes (size pixels) based on various object types. Next, the processor 130 may calculate the object's position relative to the occupant (e.g., the object's position is represented by a distance distH and / or a distance distW) based on the calculated pixel sizes, and the object position value of the object 10 may be a reference value for different types of vehicles. The distance distW is the distance between the object 10 and the edge of the lane, and the distance distH is the distance between the traffic warning system 100 and the object 10. Next, the value of distW is configured to determine whether the detected vehicle is in the same lane as the occupant. For example, if distW is less than 2 m, the processor 130 of the traffic warning system 100 determines that the approaching object is likely to be located behind the occupant and therefore in the same lane. If distW is 2 m or more, the processor 130 determines that the approaching object is likely to be located in another lane.

[0037] Referring to Figure 4, Figure 4 is a schematic diagram illustrating the relationship between the relative speed (horizontal axis) of an object 10 with respect to the traffic warning system 100 and the stopping distance (vertical axis) corresponding to a particular relative speed. The stopping distance is the distance required or estimated for an approaching vehicle to go from a moving state to a complete stop for different relative speeds when approaching an occupant vehicle equipped with the traffic warning system 100.

[0038] As shown in Figure 4, when compared with the warning mode, the stopping distance in the danger mode is shorter than that in the warning mode for the same relative speed. Also, when compared with the normal mode, the stopping distance in the danger mode is shorter than that in the normal mode for the same relative speed. From the above observation, the danger mode implies a higher urgency than the warning mode, and therefore, if the detected situation reaches or exceeds the level of the danger mode, warning the occupants should be given top priority.

[0039] If an approaching object 10 approaches the occupant from behind, the object 10 will be recognized as being in a collision zone and the occupant will be notified with either a warning or a danger message depending on the amount of reaction time remaining. Additionally, while a potentially dangerous object is located in a threat zone, the occupant will be notified in a warning mode. The collision zone and threat zones herein are for illustrative or exemplary purposes only.

[0040] 2A, 2B, and 4, if the number of vehicles is less than or equal to a preset number (e.g., five), the process proceeds to step 180A. If the number of vehicles is greater than the preset number, the process proceeds to step 180B. In step 180A, the processor outputs a warning signal according to the different modes (i.e., normal mode, warning mode, and / or danger mode) described above. In step 180B, the processor detects the top of the "preset number" of dangerous vehicles (e.g., the top five dangerous vehicles) and outputs the warning signal S1 in different modes accordingly. A dangerous vehicle refers to a vehicle that is very close (or continuously close) to the traffic warning system 100 and may collide with it in a very short time.

[0041] Referring to Figures 5 and 6, Figure 5 shows a schematic diagram of a collision zone CZ and a threat zone TZ according to one embodiment, and Figure 6 shows a schematic diagram of the relative velocity (horizontal axis) of an object 10 to a traffic warning system 100 and the corresponding stopping distance (vertical axis) according to another embodiment.

[0042] As shown in Figure 5, the area in which the traffic warning system 100 is located is defined as a collision zone CZ. A reference line R1 passes through the traffic warning system 100, and the collision zone CZ has a first collision boundary CZ1 and a second collision boundary CZ2. A first collision width W is defined between the reference line R1 and the first collision boundary CZ1. CZ1 is defined, and a second collision width W is defined between the reference line R1 and the second collision boundary CZ2. CZ2 In one embodiment, a first collision width W CZ1 is, for example, in the range of 1 meter to 3 meters (for example, 2 meters), and the second collision width W CZ2 is in the range of, for example, 1 meter to 3 meters (for example, 2 meters). CZ1 and second collision width W CZ2 When an approaching object is recognized as being located in the collision zone CZ, the mode can be determined by the processor 130 using the relative relationship shown in FIG.

[0043] As shown in Figure 5, the threat zone TZ is located outside the collision zone CZ. Furthermore, the threat zone TZ is connected to and adjacent to the first collision boundary CZ1 and the second collision boundary CZ2, respectively. The threat zone TZ has a threat width W TZ and the threat width W TZ may be in the range of, for example, 1 meter to 3 meters (e.g., 2 meters). If the approaching object is recognized to be located in the threat zone TZ (outside the collision zone CZ), the processor 130 can determine the mode using the relative relationship, as shown in FIG. 6. Furthermore, if the approaching vehicle is located within the collision zone CZ, the urgency is higher. The traffic warning system 100 can operate appropriately based on various traffic conditions and different modes, and adaptively output warning signals accordingly.

[0044] As shown in FIG. 5, the radar sensing region R R , the visual sensing range R of the visual sensor 110 Vhas a wider sensing angle to reduce or eliminate blind spots around the traffic warning system 100. The visual sensing range R of the visual sensor 110 V , the radar sensing region R of the radar sensor 120 is R has a longer sensing length to extend the detectable area farther away from the traffic warning system 100. Furthermore, in one embodiment, the radar sensing area R of the radar sensor 120 R is the visual sensing length DS of the visual sensor 110 V Radar detection length DS longer than R In one embodiment, the radar sensing length DS R may be in the range of, for example, 90 meters to 110 meters (for example, 100 meters), and the visual sensing length DS V is, for example, in the range of 20 meters to 40 meters (for example, 30 meters).

[0045] Table 1 below shows a schematic diagram of multiple modes in different embodiments of the present disclosure. In Mode 1, the radar sensor is always powered on (or kept powered on), and the visual sensor is periodically powered on (e.g., step S130B in FIG. 2A), thereby reducing power consumption. In Mode 2, when the radar sensor detects a fast-approaching object, the visual sensor is immediately powered on due to a potentially dangerous event (e.g., step S140A in FIG. 2A). In Mode 3, the radar sensor is always powered on, and the visual sensor is powered off in bad weather (e.g., heavy rain or dense fog) or when the visual confidence level L V The power is turned off if the value frequently falls below a value in the range of 50% to 80%. In mode 4, if an object behind the passenger is stationary for more than a few seconds (for example, 2 to 5 seconds), the power to the radar sensor is turned off temporarily or periodically, and the power to the visual sensor is also turned off.

[0046] [Table 1]

[0047] Disclosed herein are some example embodiments of a traffic warning system and a computer readable medium having executable instructions stored thereon for executing an algorithm or method that is processed by at least one processor of the traffic warning system, the embodiments including at least the following features:

[0048] 1. To utilize the advantages of both radar and visual sensing technologies, adaptively turn one of them on or off in a timely manner based on low power consumption states, separately preprocess the echo signals from radar and image data from vision, use a data grid to map the location information of objects detected by different sensors, and perform data fusion of measured or estimated data (such as vehicle speed (moving speed or relative speed), brake reaction time, object size, etc.) based on the quality (reliability) of the data, or perform dynamic prediction of approaching vehicles based on more reliable data.

[0049] 2. Determining the reliability of sensing data: Radar reliability level L in different scenarios of this disclosure R and visual confidence level L V The data fusion process and the visual confidence level L V and radar confidence level L R We will consider what to do if either of these is too low.

[0050] 3. If the number of detected dynamic objects is too large (e.g., five, ten, or more approaching vehicles), prioritizing the identification of the top five (or other value) potentially threatening vehicles (but not limited to vehicle types (e.g., bicycles, motorcycles, cars, trucks, vans, etc.)) may be considered to reduce power consumption and improve computational efficiency. For example, the threat level of an approaching vehicle may be determined according to the relationship between relative vehicle speed and braking response time / distance.

[0051] 4. The location of the object can be distinguished by different lanes, relative distance (far, near, etc.), and can be displayed (or indicated) in the human-machine interface, and the human-machine interface can provide warning information adaptively under different circumstances.

[0052] While the present invention has been described in terms of what are presently considered to be the most practical and preferred embodiments, it is to be understood that the invention is not necessarily limited to the disclosed embodiments. Rather, the invention is intended to cover various modifications and similar arrangements included within the spirit and scope of the appended claims, which claims should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements.

Claims

1. A traffic warning system, wherein the traffic warning system is A visual sensor configured to detect visual sensing data for each of at least one object, A radar sensor configured to detect radar sensing data for each of the at least one of the aforementioned objects, A processor electrically connected to the visual sensor and the radar sensor. Equipped with, The aforementioned processor, By analyzing the aforementioned radar sensing data, the radar confidence level of the radar sensing data is obtained. By analyzing the aforementioned visual sensing data, the visual confidence level of the visual sensing data is obtained. If both the radar confidence level and the visual confidence level exceed a preset value, the system is configured to obtain the position and velocity of at least one object by analyzing the visual sensing data and the radar sensing data. The aforementioned processor, A traffic warning system configured to determine the number of at least one object approaching the traffic warning system based on the visual sensing data and the radar sensing data, when the radar confidence level and the visual confidence level are less than or equal to the preset value, and the visual confidence level is higher than the radar confidence level.

2. The traffic warning system according to claim 1, wherein the preset value is in the range of 70% to 90%.

3. The aforementioned processor, The traffic warning system according to claim 1, configured to determine the number of at least one object approaching the traffic warning system if both the radar confidence level and the visual confidence level exceed the preset value.

4. The aforementioned processor, The traffic warning system according to claim 3, configured to output a warning signal when the number of at least one of the aforementioned objects is less than or equal to a preset number.

5. The aforementioned processor, The traffic warning system according to claim 3, wherein if the number of at least one object exceeds a preset number, the system is configured to output a warning signal for the object closest to the traffic warning system.

6. The aforementioned processor, The traffic warning system according to claim 1, wherein the radar confidence level and the visual confidence level are less than or equal to the preset value, and the radar confidence level is higher than the visual confidence level, the system is configured to determine the number of at least one object approaching the traffic warning system based on the radar sensing data.

7. The aforementioned processor, The traffic warning system according to claim 6, configured to output a warning signal when the number of at least one of the aforementioned objects is less than or equal to a preset number.

8. The aforementioned processor, The traffic warning system according to claim 6, wherein if the number of at least one object exceeds a preset number, the system is configured to output a warning signal for the object closest to the traffic warning system.

9. The aforementioned processor, The traffic warning system according to claim 1, configured to output a warning signal when the number of at least one of the aforementioned objects is less than or equal to a preset number.

10. The aforementioned processor, The traffic warning system according to claim 1, wherein if the number of at least one object exceeds a preset number, the system is configured to output a warning signal for the object closest to the traffic warning system.

11. The aforementioned processor, Determine whether or not the aforementioned at least one object is detected. If no object is detected at all, the radar sensor remains ON and the visual sensor remains OFF. The traffic warning system according to claim 1, configured as follows.

12. The aforementioned processor, Determine whether or not the aforementioned at least one object is detected. If the radar sensor detects at least one object, the visual sensor is turned ON. The traffic warning system according to claim 1, configured as follows.

13. The aforementioned processor, The traffic warning system according to claim 1, configured to periodically turn on the aforementioned visual sensor.

14. A method for operating a traffic warning system, wherein the method is: The visual sensor detects the visual sensing data of at least one object, The radar sensor detects radar sensing data for each of the at least one object, The processor analyzes the radar sensing data to obtain the radar confidence level of the radar sensing data, The processor analyzes the visual sensing data to obtain the visual confidence level of the visual sensing data, The processor includes, if both the radar confidence level and the visual confidence level exceed a preset value, analyzing the visual sensing data and the radar sensing data to obtain the position and velocity of at least one object. A method for operating a traffic warning system, further comprising the processor determining, based on the visual sensing data and the radar sensing data, the number of at least one object approaching the traffic warning system when the radar confidence level and the visual confidence level are less than or equal to the preset value, and the visual confidence level is higher than the radar confidence level.

15. The operating method according to claim 14, wherein the preset value is in the range of 70% to 90%.

16. The operating method according to claim 14, further comprising the processor determining the number of at least one object approaching the traffic warning system if both the radar confidence level and the visual confidence level exceed the preset value.

17. The operating method according to claim 16, further comprising the processor outputting a warning signal if the number of at least one object is less than or equal to the preset number.

18. The operating method according to claim 16, further comprising the processor outputting a warning signal for the object closest to the traffic warning system if the number of at least one object exceeds a preset number.

19. The operating method according to claim 14, further comprising the processor outputting a warning signal if the number of at least one object is less than or equal to a preset number.

20. The operating method according to claim 14, further comprising the processor outputting a warning signal for the object closest to the traffic warning system if the number of at least one object exceeds a preset number.

21. The processor determines whether or not the at least one object is detected, and If no object is detected at all, the processor will keep the radar sensor ON and the visual sensor OFF. The operating method according to claim 14, further comprising:

22. The processor determines whether or not the at least one object is detected, and When the radar sensor detects at least one object, the processor turns on the visual sensor. The operating method according to claim 14, further comprising:

23. The operating method according to claim 14, further comprising periodically turning on the visual sensor.

24. Detect the visual sensing data for at least one object, The radar sensing data of each of the at least one of the aforementioned objects is detected, By analyzing the aforementioned radar sensing data, the radar confidence level of the radar sensing data is obtained. By analyzing the aforementioned visual sensing data, the visual confidence level of the visual sensing data is obtained. If both the radar confidence level and the visual confidence level exceed a preset value, a method for obtaining the position and velocity of at least one object is obtained by analyzing the visual sensing data and the radar sensing data. A non-temporary computer-readable medium storing executable instructions for at least one processor of a traffic warning system to be executed by the traffic warning system, The method described above is If the radar confidence level and the visual confidence level are less than or equal to the preset value, and the visual confidence level is higher than the radar confidence level, the processor further includes determining the number of at least one object approaching the traffic warning system based on the visual sensing data and the radar sensing data. A non-temporary computer-readable medium.

25. The processor determines whether or not the at least one object has been detected, When at least one of the aforementioned objects is detected by the radar sensor, the processor turns on the visual sensor. A non-temporary computer-readable medium according to claim 24, which stores instructions executable by at least one processor of the traffic warning system for causing the traffic warning system to perform the following.

Images