Traffic warning system, computer-readable medium, and method of operation thereof

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 for operating the same. [Background technology]

[0002] As the number and types of vehicles increase, the probability of traffic accidents is also increasing year by year. Clearly, in addition to the continuous advancements in transportation technology in power management systems, improving safety while driving has become a crucial issue. For example, bicycle routes contain many objects that riders must pay attention to, such as traffic signs, pedestrians, and obstacles. Riders need to pay attention to the road environment as a basis for their next driving actions.

[0003] In modern society, traffic accidents frequently occur when passengers fail to pay attention to road conditions. Due to fatigue or distraction, passengers 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 ahead.

[0004] In U.S. Patent No. 10668971, a 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 a predetermined safety threshold. The rating is uploaded to a remote server along with video evidence. The device may have a highly distinctive light indicator / effect (also functioning as a bicycle rear light) so that the driver can only react by recognizing the device and overtaking if it is safe to do so.

[0005] The invention disclosed 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, which is used to capture and process images of the road conditions in front of the intelligent vehicle in order to acquire road condition information ahead, the rear millimeter-wave radar is used to enable the user to acquire information on obstacles behind the intelligent vehicle, the rear obstacle information includes the speed of the obstacles, a data processing module is used to process the road condition information and the rear obstacle information, and a rear early warning processing unit is provided, which is used to acquire the collision time according to the relative speed in the y-direction between the rear obstacle and the intelligent vehicle, and issues a warning when the collision time reaches a rear collision threshold, thereby disclosing a collision early warning system. The early warning function ensures the safe driving of the intelligent vehicle in accordance with this scheme, significantly reduces the possibility of the vehicle colliding, and improves the safety of driving the intelligent vehicle. [Overview of the Initiative]

[0006] One embodiment of the present invention provides a traffic warning system comprising a visual sensor, a radar sensor, and a processor. The visual sensor is configured to detect visual sensing data for each of at least one object. The radar sensor is configured to detect radar sensing data for each of at least one object. The processor is configured to obtain a radar confidence level for the radar sensing data by analyzing the radar sensing data, obtain a visual confidence level for the visual sensing data by analyzing the visual sensing data, and, if both the radar confidence level and the visual confidence level exceed a preset value, obtain the position and velocity of the at least one object by analyzing the visual sensing data and the radar sensing data.

[0007] In another embodiment of the present invention, a method of operating a traffic warning system includes the following steps: detecting, by a vision sensor, visual sensing data of each of at least one object; detecting, by a radar sensor, radar sensing data of each of at least one object; obtaining, by a processor, a radar confidence level of the radar sensing data by analyzing the radar sensing data; obtaining, by a processor, a vision confidence level of the visual sensing data by analyzing the visual sensing data; and when both the radar confidence level and the vision confidence level exceed a preset value, obtaining, by the processor, the position and velocity of at least one object by analyzing the visual sensing data and the radar sensing data.

[0008] In one embodiment of the present invention, a computer-readable medium is provided. The non-transitory computer-readable medium stores instructions executable by at least one processor of a traffic warning system, the instructions being for causing the traffic warning system to detect respective visual sensing data of at least one object, detect respective radar sensing data of the at least one object, obtain a radar confidence level of the radar sensing by analyzing the radar sensing data, obtain a vision confidence level of the visual sensing data by analyzing the visual sensing data, and when both the radar confidence level and the vision confidence level exceed a preset value, obtain the position and velocity of the at least one object by analyzing the visual sensing data and the radar sensing data.

[0009] Many objects, features, and advantages of the present invention will become readily apparent to those of ordinary skill in the art upon reading the following detailed description of the embodiments of the present invention in conjunction with the accompanying drawings. It should be understood, however, that the various embodiments and drawings used herein are for illustrative purposes only and are not to be construed as limiting the present invention.

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

Brief Description of the Drawings

[0011] [Figure 1] It is a schematic functional block diagram of a traffic warning system according to an embodiment of the present invention. [Figure 2A] It shows a flowchart of an operation method of the traffic warning system of FIG. 1. [Figure 2B] It shows a flowchart of an operation method of the traffic warning system of FIG. 1. [Figure 3] It is a schematic diagram of object recognition in different lanes by a vision-based algorithm in the present disclosure. [Figure 4] It is a schematic diagram showing the relative relationship between the relative speed (horizontal axis) of an object with respect to the traffic warning system and the corresponding stopping distance (vertical axis). [Figure 5] It is a schematic diagram of a collision zone and a threat zone in an embodiment. [Figure 6] It is a schematic diagram showing the relative relationship between the relative speed (horizontal axis) of an object with respect to the traffic warning system and the corresponding stopping distance (vertical axis) according to another embodiment.

Modes for Carrying Out the Invention

[0012] The present invention integrates two types of sensors, such as a "radar or microwave sensor" and a "vision sensor", and while leveraging the advantages of both, a traffic warning system that can significantly reduce the power consumption required for the operation of the system by means of an implementable operation method and control logic, which can improve the driving safety of passengers, avoid potential risks posed by vehicles approaching from behind, or timely output warnings to avoid threats in the blind spots of passengers. In particular, it provides a type of traffic warning system applicable to various mobile lightweight transportation vehicles, such as bicycles, electric bicycles (e-bikes), and other multi-wheeled personal mobility means. Such urgent detection results can be detected throughout the system, and since the information that should be reflected first can be timely provided to the passengers, beneficial effects such as power saving, low interference, and accurate warnings can be achieved.

[0013] Referring to FIG. 1, FIG. 1 shows a schematic diagram of the functional blocks of a traffic warning system 100 according to an embodiment of the present invention. The traffic warning system 100 can be mounted on a vehicle (such as a bicycle, an electric bicycle, or other multi-wheeled personal mobility means, etc.).

[0014] As shown in FIG. 1, the traffic warning system 100 includes a vision sensor 110, a radar sensor 120, and a processor 130. The vision sensor 110 is configured to detect the vision sensing data D of each of at least one object 10. V The radar sensor 120 is configured to detect the radar sensing data D of each of at least one object 10. R The processor 130 analyzes the radar sensing data D to obtain the radar confidence level L of the radar sensing data. R The processor 130 analyzes the vision sensing data D to obtain the vision confidence level L of the vision sensing data D. R When both the radar confidence level L and the vision confidence level L exceed a preset value, the vision sensing data D V When both the radar confidence level L and the vision confidence level L exceed a preset value, the vision sensing data D V When both the radar confidence level L and the vision confidence level L exceed a preset value, the vision sensing data D V When both the radar confidence level L and the vision confidence level L exceed a preset value, the vision sensing data D R When both the radar confidence level L and the vision confidence level L exceed a preset value, the vision sensing data D V When both the radar confidence level L and the vision confidence level L exceed a preset value, the vision sensing data DV and radar sensing data D R The system is configured to obtain the position and velocity of object 10 by analyzing the data. In one embodiment, the accuracy of object detection can be improved by using the vision sensor 110 and the radar sensor 120 in combination.

[0015] In this specification, “visual sensor” can be implemented by an imaging device, such as a camera (e.g., a GoPro camera mounted on a helmet, 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 objects 10 using an AI (artificial intelligence) model. The AI ​​model can be acquired in advance by various appropriate machine learning techniques, such as deep learning. Objects 10 are, for example, obstacles, vehicles, pedestrians, etc. The visual sensor 110 has the characteristics of (1) having a close and near sensing range and reducing blind spots within a field of view (FOV) area of ​​length 30m and width 144°, and (2) visual detection is 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 is available in a variety of specifications, with or without lane recognition functionality. Generally, the power consumption of visual sensors (e.g., 10W) ​​is much higher than that of radar sensors, limiting their applicability. In this embodiment, the traffic warning system can integrate the radar sensor and the visual sensor to minimize power consumption and further enhance so-called "smart" functions (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, by using, for example, 24GHz millimeter waves, has the characteristic of further extending coverage beyond a field of view of at least 100 meters (m) in length and 8 m in width.

[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 status of the sensors according to different traffic scenarios or environmental conditions, and to process the data sensed or generated by the sensors. The processor 130 is capable of controlling the visual sensor 110 and the radar sensor 120 to implement the operation method of the traffic warning system 100. The processor 130 processes the visual sensing data D V Based on this, the position and size of object 10 (object frame) are obtained, and radar sensing data D R Based on this, it is possible to obtain the position and velocity of object 10.

[0018] Referring to Figures 2A and 2B, Figures 2A and 2B show sequential flowcharts corresponding to different situations in which the traffic warning system 100 of Figure 1 operates.

[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, while the vision sensor 110 may remain OFF to conserve power.

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

[0021] In step S130A, the radar sensor 120 remains in the ON state, while the vision sensor 110 remains in the OFF state to conserve power.

[0022] In step S130B, the visual sensor 110 is turned on at each sampling interval (e.g., 2 to 5 seconds) to detect whether or not an object 10 appears around the traffic warning system 100. In one embodiment, the visual sensor 110 may be turned on for only a validity period (e.g., 2 to 5 seconds) at each sampling interval. In one embodiment, this validity 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 improve or enhance the reliability of the detection result. The visual sensor 110 receives the visual sensing data D of at least one object 10. V It is detectable.

[0024] In steps S140B and S140C, the processor 130 performs a fusion procedure. In the fusion procedure, the processor 130 processes radar sensing data D R By analyzing (for example, based on reflected signal intensity, flight time, etc.), radar sensing data D R Radar confidence level L R To obtain and visual sensing data D V By analyzing (for example, based on captured images and image processing results, etc.), visual sensing data D V Visual confidence level L V It is possible to obtain the radar confidence level L. R This can be acquired, for example, using millimeter-wave radar, and the visual confidence level L V This can be acquired using, for example, CMOS (complementary metal-oxide-semiconductor) sensor cameras, QVGA (quarter video graphics array) sensor cameras, and / or infrared sensor cameras.

[0025] In one embodiment, only one of the radar sensor or the visual sensor may be in operation, providing more reliable detection information in specific situations. In the former case, visual detection accuracy is significantly reduced in bad weather and / or nighttime conditions, so the radar sensor alone can be used to detect vehicles, further improving this adverse condition. In the latter case, if an approaching object is outside the detection range of the radar sensor, the visual sensor can detect vehicles in the radar's blind spot, further improving this missing condition. Thus, the two types of sensors can be operated adaptively depending on different situations.

[0026] In step S140D, the processor 130 sets the radar confidence level L R and visual confidence level L V Determine whether both of the following exceed a preset value. Radar confidence level L R and visual confidence level L V If both exceed the preset values, the process proceeds to step S150A. Radar confidence level L R and visual confidence level L V is a pre-set value The following is In that case, the process is step S160 A Proceed to the next step. In one embodiment, this preset value may be in the range of, for example, 70% to 90% (for example, 80%), but may be smaller or larger. 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 processes the visual sensing data D V and radar sensing data D R By analyzing this data, it is possible to obtain the position (e.g., relative position) and velocity (e.g., relative velocity) of object 10.

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

[0029] In step S160A, the processor 130 sets the radar confidence level L R Visual confidence level L V Determine whether it is higher than or equal to Radar Confidence Level L. R Visual confidence level L V If it is higher than that, the process proceeds to step S160B. Radar confidence level L R Visual confidence level L V If it is not higher than (for example, visual confidence level L) V Radar confidence level L R If it is higher than that, the process proceeds to step S160C.

[0030] In step S160B, the processor 130 processes radar sensing data D R The number of objects 10 approaching the traffic warning system is determined based on (for example, position and speed).

[0031] In step S160C, the processor 130 processes the visual sensing data D V (For example, location) and radar sensing data D R The number of objects 10 approaching the traffic warning system is determined based on (for example, speed or estimated speed).

[0032] In step S170, the processor 130 determines whether the number of objects 10 is less than or equal to a preset number. If the number of objects 10 is less than or equal to the preset number, the process proceeds to step S180A. If the number of objects 10 exceeds the preset number, the process proceeds to step S180B. In one embodiment, the preset number is, for example, 5, or a number less than or greater than 5.

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

[0035] Referring to Figure 3, Figure 3 shows a schematic diagram of object recognition in different traffic lanes using the vision-based algorithm described herein.

[0036] First, the processor 130 may acquire object frame 11 (or vehicle identification frame) information depending on the size of the approaching object. Next, the processor 130 may calculate the size of pixels based on the type of object. Then, based on the calculated pixel size, the processor 130 may calculate the position of the object relative to the occupant (for example, the position of the object is represented by distance distH and / or distance distW), and the object position value of object 10 may be a reference value for different types of vehicles. Distance distW is the distance between object 10 and the edge of the lane, and distance distH is the distance between the traffic warning system 100 and 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 2m, the processor 130 of the traffic warning system 100 determines that the approaching object is likely to be in the same lane because it is behind the occupant. If distW is 2m or more, the processor 130 determines that the approaching object is likely to be in a different lane.

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

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

[0039] If an approaching object 10 is coming from behind the occupant, the object 10 is recognized as being in the collision zone, and the occupant is notified of either a warning or a danger message, depending on how much reaction time remains. Furthermore, while an object with potential danger is in the threat zone, the occupant will be notified in warning mode. The collision zone and threat zone here are for illustrative purposes only.

[0040] As shown in Figures 2A, 2B, and 4, if the number of vehicles is less than or equal to a preset number (e.g., 5 vehicles), the process proceeds to step 180A. If the number of vehicles exceeds the preset number, the process proceeds to step 180B. In step 180A, the processor outputs a warning signal according to the different modes described above (i.e., normal mode, warning mode, and / or danger mode). In step 180B, the processor detects the top five dangerous vehicles in the "preset number" of dangerous vehicles (e.g., the top 5 dangerous vehicles) and outputs a warning signal S1 in a different mode accordingly. Dangerous vehicles are defined as vehicles that are very close (or continuously close) to the traffic warning system 100 and are likely to collide in a very short time.

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

[0042] As shown in Figure 5, the area where the traffic warning system 100 is located is defined as a collision zone CZ. The 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, with a first collision width W between the reference line R1 and the first collision boundary CZ1. CZ1 The second collision width W is defined between the reference line R1 and the second collision boundary CZ2. CZ2 The first collision width W is defined. In one embodiment, the first collision width W is defined. CZ1 This ranges from, for example, 1 meter to 3 meters (for example, 2 meters), and the second collision width W CZ2 This ranges from, for example, 1 meter to 3 meters (for example, 2 meters). First collision width W CZ1 and the second collision width W CZ2 These may be the same or different. If 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 Figure 4.

[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 It has a threat width W TZ This range can be, for example, 1 meter to 3 meters (for example, 2 meters). If an approaching object is recognized as being 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 Figure 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 diverse traffic conditions and different modes, and can adaptively output warning signals accordingly.

[0044] As shown in Figure 5, the radar sensing area R of the radar sensor 120 R In comparison, the visual sensing range R of the visual sensor 110 V The visual sensor 110 has a wider sensing angle to reduce and eliminate blind spots near the traffic warning system 100. V In comparison, the radar sensing area R of the radar sensor 120 R It has a longer sensing length to expand the detectable area far from the traffic warning system 100. Furthermore, in one embodiment, the radar sensing area R of the radar sensor 120 R The visual sensing length DS of the visual sensor 110 V Longer radar detection length than DS R It has a radar detection length DS. R This can be in the range of, for example, 90 meters to 110 meters (for example, 100 meters), and the visual perception length DS V This range is, for example, 20 to 40 meters (for example, 30 meters).

[0045] Table 1 below shows schematic diagrams of several modes in different embodiments of the present disclosure. In Mode 1, the radar sensor is always powered ON (or kept powered ON), and the vision sensor is periodically powered ON (e.g., step S130B in Figure 2A), thereby reducing power consumption. In Mode 2, when the radar sensor detects a fast-approaching object, the vision sensor is immediately powered ON due to a potentially hazardous event (e.g., step S140A in Figure 2A). In Mode 3, the radar sensor is always powered ON, and the vision sensor is powered OFF in adverse weather conditions (e.g., heavy rain or dense fog) or at a vision confidence level L V The power will turn off if the value frequently falls below the range of 50% to 80%. In Mode 4, if an object behind the occupant remains stationary for more than a few seconds (e.g., 2 to 5 seconds), the radar sensor will be temporarily or periodically turned off, and the vision sensor will also be turned off.

[0046] [Table 1]

[0047] The disclosure herein of a traffic warning system and a computer-readable medium storing executable instructions for performing an algorithm or method, which are processed by at least one processor of the traffic warning system, are some exemplary embodiments that include at least the following features:

[0048] 1. To leverage the advantages of both radar and vision sensing technologies, one of them is adaptively turned ON or OFF in a timely manner based on low power consumption conditions; echo signals from radar and image data from vision are preprocessed separately; a data grid is used to map the positional information of objects detected by different sensors; and data fusion is performed on measured or estimated data (such as vehicle speed (moving speed or relative speed) or brake reaction time, object size, etc.) with the quality (reliability) of that data, or dynamic predictions are made for approaching vehicles based on more reliable data.

[0049] 2. Determination of the reliability of sensing data: Radar confidence 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 how to handle the situation when all of these values ​​are too low.

[0050] 3. If the number of detected moving objects is too high (for example, if there are 5, 10, or more approaching vehicles), it may be considered to prioritize the identification of the top 5 (or other) potentially threatening vehicles (but not limited to vehicle type, e.g., bicycles, motorcycles, cars, trucks, vans, etc.) to reduce power consumption and improve computational efficiency. For example, the threat level of an approaching vehicle may be determined based on the relationship between relative vehicle speed and brake response time / distance.

[0051] 4. The positions of the above-mentioned objects can be distinguished by different lanes, relative distances (far, near, etc.), and can be displayed (or indicated) in a human-machine interface, which can adaptively provide warning information under different circumstances.

[0052] While the present invention has been described in relation to embodiments considered most practical and preferred at present, it should be understood that the present invention is not limited to the disclosed embodiments. Rather, the present invention is intended to cover a variety of modifications and similar configurations that fall within the essence and scope of the appended claims, and the claims should be interpreted most broadly to encompass all such modifications and similar configurations.

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