TW202004530A - Responsibility analysis system for car accident and method thereof using a large number of surveillance videos to determine whether a car accident has occurred - Google Patents

Abstract

The invention relates to a responsibility analysis system for car accidents and a method thereof, and the main structure includes a plurality of monitoring devices for recording road condition surveillance video, a plurality of car accident identification modules for identifying the occurrence of a car accident, a cloud identification system for evaluating the severity of a car accident, a monitoring center for sending a car accident broadcast message, at least one search device for searching a sending destination of the car accident broadcast message, and an application for a user to view data related to the car accident, that is, car accident analysis software. In this way, a large number of surveillance videos are used to determine whether a car accident has occurred. When a car accident occurs, the relevant units are notified according to the car accident status, and after the application of the interested party is connected through the search device, the surveillance video is transmitted for inspection and retention. As such, there is no need for taking photos as evidences and reporting assistance, so the cars can be quickly moved to avoid affecting the traffic.

Description

Responsibility analysis system and method for car accident

The present invention is to provide a system and method for responsibility analysis of car accidents, in particular, it can automatically collect evidence on the scene of car accidents and keep them for record, so as to clarify the blame and quickly restore the road conditions afterwards, while effectively reducing the manpower of police and other accidents. The analysis system and method of the rate of responsibility for car accidents.

According to the fact that Taiwan has a dense population and heavy use of vehicles such as automobiles and locomotives, the frequency of car accidents is quite frequent. The attribution of the responsibility for the accident in the late stage of the car accident is a headache for both parties and traffic control units. However, at present, only the "Taipei City Traffic Control Center" website of the Transportation Bureau of the Taipei City Government can provide free viewing of street video footage and can instantly check whether traffic accidents have affected traffic conditions. However, the information on the website is The traffic control unit can only play a role after receiving the message or after the public reports it, which is too passive.

At present, although the driving recorder for vehicle loading is helpful for the identification of traffic accidents, the general public has limited clarification of the responsibilities of vehicle accidents and limited understanding of traffic laws. They can only call the police and wait for the police to deal with the scene, or send them to the mediation committees of various towns and cities for mediation. Or it can be sent to accident appraisal units at all levels for appraisal, or even judicial proceedings to clarify legal liabilities. However, when an accident actually occurs, the two parties will each have their own arguments on the interpretation and application of regulations due to differences in their cognition, and when interpreting afterwards, if only the photo and distance information are used, it may be because of different shooting angles, There are obstacles that block the sight and cause misjudgment of the reader. In addition, the traffic accident scene is usually on a busy road or a fast road, and the accident scene range of many vehicles may be too large. In addition to the need to occupy the accident road section and affect the traffic, the time-consuming measurement and forensics It will also expose the surveyor to a dangerous situation and cause a second accident.

And according to the I502558 "Traffic Accident Monitoring and Tracking System" announced by the Patent Search System of the Republic of China, it is to disclose a network-type digital image monitoring system applied to a large number of cameras, digital image recognition, and digital audio recognition functions with radio equipment, using The results of image recognition and audio recognition can detect the characteristics of traffic accidents and visualize the location of traffic accidents, so that the police can go to the location where the accident occurred. However, although this technology can automatically identify traffic accidents, it still has to wait for the police to deal with it, and cannot quickly solve the problem of car accidents.

Therefore, how to use simple and fast ways to solve the problem of car accidents and quickly restore the road to normal conditions is the one where the inventors of the present invention and related manufacturers engaged in this industry are eager to study the direction of improvement.

Therefore, in view of the above-mentioned deficiencies, the inventor of the present invention has collected relevant information, evaluated and considered through multiple parties, and based on years of experience accumulated in this industry, through continuous trial and modification, began to design this kind of automatic collection of car accident scenes Evidence and keep the record for the purpose of clarifying the blame and quick recovery of road conditions afterwards, and effectively reduce the police and other manpower, reduce the accident rate of the second accident of the car accident accident analysis system and method patent inventor.

The main purpose of the present invention is to use artificial intelligence to automatically identify the occurrence of car accidents and the status of car accidents, and to automatically send information such as monitoring video and audio to relevant personnel and units.

In order to achieve the above object, the main structure of the present invention includes: a plurality of monitoring devices for recording and monitoring audio and video, and a plurality of car accident recognition modules respectively provided in each of the monitoring devices, and each of the car accident recognition modules includes a device for instantaneous return occurrence High decibel message to the sound interpretation unit of the car accident identification module, a cloud identification system to link the car accident identification modules and to evaluate the severity of the car accident status, an information link to the cloud identification system and to send at least one car accident A monitoring center for broadcast messages, at least one information linking the monitoring center for searching for the sending target of the car accident broadcast message, and the search device includes at least one positioning module, and at least one electronic tag recognition on the side of the road The module and at least one car accident analysis software installed in the driver's mobile electronic device and the information link to the search device are for receiving and displaying the car accident broadcast message.

The invention mainly installs a monitoring device (which can also be set at the roadside) in all directions of the intersection, and records the monitoring image at any time for 24 hours. When the car accident recognition module receives the high decibel message of the sound interpretation unit, it immediately recognizes the monitoring within the predetermined time Video and audio to determine whether a car accident has occurred. When it is judged that there is a car accident, the severity of the car accident is evaluated by the cloud identification system or the personnel of the monitoring center assist in the judgment, and the control monitoring center will notify the disaster relief unit, and at the same time, use the positioning module and electronic label identification module of the search device Team to find out the satellite positioning position of the eTag and its mobile electronic device on the party’s car, and the surveillance video related to the accident is transmitted by the monitoring center to the party’s car accident analysis software. In this way, at the moment of a car accident, all the procedures of the standard car accident handling process are automatically completed to eliminate the problem of insufficient evidence search and unclear responsibility, and save the time for dispute notification, so that the parties can directly move the car while keeping the road clear 3. Avoid secondary accidents.

With the above technology, it is possible to break through the problems of incomplete evidence collection, clarification of suspicions, waste of human resources, and traffic safety hazards in conventional car accident incident handling methods, and achieve the practical progress of the above advantages.

1. 1b, 1A, 1B ‧‧‧ monitoring device

2‧‧‧Car accident identification module

21‧‧‧Sound Interpretation Unit

3. 3a, 3c ‧‧‧ cloud identification system

31a‧‧‧Audio and video recording module

32c‧‧‧Fire identification module

33c‧‧‧person security identification module

4, 4a, 4b ‧‧‧ monitoring center

41b‧‧‧Distance calculation module

5, 5b‧‧‧ search device

51, 51b‧‧‧positioning module

52、52b‧‧‧Electronic tag recognition module

6. 6a‧‧‧Car accident analysis software

61a‧‧‧ Settlement Unit

7‧‧‧Mobile electronic device

81‧‧‧Car

811‧‧‧ Rearview mirror

812‧‧‧Electronic label

82‧‧‧ Locomotive

821‧‧‧ Knight

9‧‧‧ Disaster relief unit

The first figure is a structural block diagram of a preferred embodiment of the present invention.

The second figure is a distribution diagram of the monitoring device of the preferred embodiment of the present invention.

The third figure is a schematic diagram of the system relationship of the preferred embodiment of the present invention.

The fourth figure is a block flow diagram of a preferred embodiment of the present invention.

The fifth figure is a schematic diagram of APP installation according to a preferred embodiment of the present invention.

The sixth figure is a schematic diagram (1) of monitoring audio and video according to a preferred embodiment of the present invention.

The seventh figure is a schematic diagram (2) of monitoring audio and video according to a preferred embodiment of the present invention.

The eighth figure is a block diagram of a car accident recognition block according to a preferred embodiment of the present invention.

The ninth figure is a schematic diagram of a traffic accident notification according to a preferred embodiment of the present invention.

The tenth figure is a schematic diagram of a search operation according to a preferred embodiment of the present invention.

Figure 11 is a schematic diagram of a broadcast message according to a preferred embodiment of the present invention.

Fig. 12 is a block diagram of another preferred embodiment of the present invention.

Figure 13 is a reconciled schematic diagram of yet another preferred embodiment of the present invention.

Figure 14 is a decision diagram of the notification process according to yet another preferred embodiment of the present invention.

Fig. 15 is a structural block diagram of another preferred embodiment of the present invention.

Figure 16 is a schematic diagram of a warning broadcast according to yet another preferred embodiment of the present invention.

Figure 17 is a schematic diagram of a situation recognition of a car accident according to another preferred embodiment of the present invention.

In order to achieve the above objectives and effects, the technical means and structure adopted by the present invention, the drawings and details of the preferred embodiments of the present invention are described in detail below. Their features and functions are as follows, so that they fully understand.

Please refer to the first to fourth figures, which are block diagrams to block flow diagrams of the preferred embodiment of the present invention. It can be clearly seen from the figures that the present invention includes: a plurality of monitoring devices 1 for recording monitoring A plurality of car accident recognition modules 2 are provided in each of the monitoring devices 1 to identify whether a car accident has occurred, and each car accident recognition module 2 includes a sound interpretation unit 21 for instantaneous return transmission High-decibel information to the car accident identification module 2; an information link to the cloud identification system 3 of the car accident identification modules 2 for evaluating the severity of the car accident status; an information link to the monitoring center 4 of the cloud identification system 3, At least one car accident broadcast message is sent according to the car accident situation; at least one information link to the search device 5 of the monitoring center 4 is used to search for the send target of the car accident broadcast message, and the search device 5 includes at least one positioning module 51, And at least one electronic tag recognition module 52 provided on one side of the road; and at least one car accident analysis software 6 linked to the search device 5 is provided in at least one driver's mobile electronic device 7 for reception and display The car accident broadcast message.

The main steps of the liability analysis method for car accidents of the present invention include: (a) recording surveillance audio and video on road conditions using a plurality of monitoring devices; (b) using an audio interpretation unit to transmit instantaneous high-decibel messages to each vehicle in the event of a car accident The car accident identification module; (c) Based on the monitored audio and video, a plurality of car accident identification modules identify whether a car accident has occurred; (d) Based on the monitored audio and video and the high decibel message, a cloud identification system evaluates the severity of the car accident status; (e) According to the situation of the car accident, at least one car accident broadcast message is sent by a monitoring center; (f) Using the positioning module and the electronic tag recognition module of a search device to search for the sending target of the car accident broadcast message; (g) enable at least A driver's mobile electronic device receives and displays the car accident broadcast message through a car accident analysis software.

Through the above description, we can understand the structure of this technology, and according to the corresponding cooperation of this structure, we can automatically collect evidence on the scene of the car accident and keep it for record, so as to clarify the blame and quickly restore the road conditions afterwards, and have the manpower to reduce police affairs. , Reduce the incidence of secondary accidents and other advantages, and a detailed explanation will be explained below.

Please also refer to the first to eleventh figures, which are the structure block diagram to the broadcast message schematic diagram of the preferred embodiment of the present invention. When the above components are configured, it can be clearly seen from the figure that the system Mainly cooperate with street cameras (surveillance device 1), artificial intelligence recognition system (car accident recognition module 2 and cloud recognition system 3), and APP (car accident analysis software 6) installed by users on mobile phones (mobile electronic devices 7) Analysis and integration. Among them, the monitoring device 1 can be equipped with a car accident identification module 2 in an existing camera on the road, or a new camera including the car accident identification module 2 can be used. The erection of the monitoring device 1 determines the number of erections according to the size of the intersection, for example, Two-lane roads are equipped with 2 units at each intersection, a total of 8 units, and four-lane roads are equipped with 3 to 4 units and a total of 12 to 16 units at each intersection. The surveillance video recorded by monitoring device 1 directly uses the internal car accident recognition module 2 Identify whether there is a car accident, and can be sent to the cloud recognition system 3 for further identification of the car accident situation, or directly sent to the monitoring center 4, the monitoring center 4 can use the split screen to simultaneously display the surveillance video of multiple intersections, or a single intersection The multi-angle monitoring of audio and video, and when the artificial intelligence cannot judge, the background operator judges; and the judgment basis of the car accident identification module 2 and the cloud identification system 3 is mainly based on the monitoring of abnormal traffic congestion in the audio and video, and the integrity of the car body shape Situational judgments such as sex, abnormal stay on the road, and matching with the big data data of historical car accident data, so it can more and more accurately judge the occurrence of car accidents and the severity of car accident conditions; as for the car accident analysis software 6 (such as (Figure 5) can be recommended to download and install through the mobile electronic device 7 when the public obtains a driver's license, replaces a driver/driver's license, or inspects a car, so as to increase the coverage of the present invention.

In actual use, the monitoring device 1 is used to record 24 hours of surveillance images at each intersection (step a). At the moment of a car accident, a high-decibel collision sound or a high-decibel screaming sound will generally be produced, so if the sound is interpreted When the unit 21 receives a loud sound, it immediately determines whether it is the collision sound or scream. If it is determined that the sound may be caused by a car accident, a high-decibel message is generated and sent to the car accident recognition module 2 (step b), and When the car accident recognition module 2 receives the high-decibel message from the sound interpretation unit 21, it immediately recognizes the surveillance video within a predetermined time (such as within 5 minutes) to determine whether a car accident has occurred, thereby monitoring the first time the car accident occurs For video and audio collection, even if the audio interpretation unit 21 does not receive or interpret errors, it can still use the aforementioned identification technology to identify multiple angles of video and audio simultaneously based on a large number of monitoring (step c).

For example, as shown in the sixth and seventh figures, they respectively show the surveillance audio and video obtained from different angles using different monitoring devices 1A, 1B at the same accident scene. From the sixth figure, it can be seen that the cause of the accident is the collision of the car 81 Locomotive 82, but from the seventh picture, we can see the image of the rear view mirror 811 of the car 81 crashed by the locomotive 82. Therefore, the fact of the car accident should be that the locomotive 82 hit the car 81. By using multi-angle surveillance video to analyze the responsibility of the car accident, At the same time, the fact that the car accident occurred can be judged by the traces of the impact, the fallen locomotive 82 and the knight 821.

When it is judged that there is a car accident, the cloud recognition system 3 evaluates the severity of the car accident and controls the monitoring center 4 to immediately notify the disaster relief unit 9 (such as a police station) (step d), as shown in the ninth figure, at this time File the surveillance video and audio file of the scene of the car accident within the predetermined time and send it to the disaster relief unit 9 (step e). The above-mentioned identification technology is implemented by a digital image and audio identification system, and can be assisted by the operator of the monitoring center 4 to increase the accuracy of artificial intelligence determination in each learning. At the same time, the positioning module 51 and the electronic tag recognition module 52 of the search device 5 are used to find the satellite tag position of the electronic tag (eTag) 812 and the mobile electronic device 7 on the person's car (step f). The search action of the search device 5 is mainly based on the phone number, license plate number and other information registered by the person in the car accident analysis software 6 for the positioning module 51 to use a satellite positioning system to search for the mobile phone position of the phone number and the identification of the power supply label The module 52 reads the electronic tag (eTag) 812 provided on the vehicle body. The electronic tag recognition module 52 of this embodiment is provided in the monitoring device 1 and is detected as the detection range marked in the tenth figure. In this way, it is possible to find out the car accident analysis software 6 corresponding to the person concerned according to the vehicle that caused the accident.

Therefore, as shown in the eleventh figure, after confirming the occurrence of a car accident, the surveillance center 4 will immediately transmit the relevant surveillance images to the party responsible for the accident, so that the party can use the car accident analysis software 6 to view the information for automatic evidence search and clarify the responsibility for the accident (Step g). In this way, at the moment of a car accident, all the procedures of the standard car accident handling process are automatically completed to eliminate the problem of insufficient evidence search and unclear responsibility, and save the time for dispute notification, so that the parties can directly move the car while keeping the road clear 3. Avoid secondary accidents.

Please also refer to the twelfth to fourteenth figures, which are the block diagrams of the structure of another preferred embodiment of the present invention to the decision diagram of the notification process. As can be clearly seen from the figures, this embodiment is as described above. The embodiments are similar, and only the cloud identification system 3a includes an audio and video recording module 31a for backing up the monitored audio and video, and the car accident analysis software 6a includes an information link to the reconciliation unit 61a of the monitoring center 4a. A reconciliation message is generated and attached to the car accident broadcast message. In this way, after the parties have reviewed the surveillance video, they can use the reconciliation unit 61a to reconcile, to attach the generated reconciliation information to the car accident broadcast message, and synchronously pass it to the police. In this way, the police does not have to go to the scene to process, and To save manpower, of course, the reconciliation conditions can be recorded in the reconciliation information. At the same time, the related surveillance audio and video of the car accident will also be backed up in the audio and video recording module 31a for inspection, to avoid unilateral reliance.

Please also refer to the fifteenth to sixteenth figures, which are structural block diagrams of another preferred embodiment of the present invention to the notification flow decision diagram. It can be clearly seen from the figures that this embodiment is as described above. The embodiments are similar, and only the monitoring center 4b includes an information linking the distance calculation module 41b of the positioning module 51b, which is used to limit the distance range of the car accident broadcast message. While determining the occurrence of a car accident, according to the predetermined range set by the distance calculation module 41b (such as within one kilometer), the predetermined range can be calculated by using the positional distance between the monitoring device 1b and the electronic tag recognition module 52b, and cooperate with the monitoring center 4b And the search device 5b sends a car accident broadcast message to the car owner within a predetermined range, thereby notifying the driver in the vicinity of a car accident in front of him, so that he can bypass. In this embodiment, the electronic tag identification module 52b can also directly use an electronic toll collection system (Electronic Toll Collection, ETC) for identification.

Please also refer to the seventeenth figure, which is a schematic diagram of the identification of a car accident situation according to another preferred embodiment of the present invention. It can be clearly seen from the figure that this embodiment is similar to the above embodiment, only in the cloud The identification system 3c includes a selective information linking fire identification module 32c of the fire fighting unit, and a selective information linking medical personnel identification module 33c. In this way, the cloud identification system 3c can use the fire identification module 32c to monitor the fire and smoke at the scene of the car accident to determine whether there is a possibility of fire, and whether to report the fire unit, and use a personnel safety identification module 33c Detect the injuries of personnel, such as heavy bleeding, coma, etc., to determine whether to notify the medical unit. In addition, it is also possible to determine whether the towing site needs to be notified by the degree of damage to the body shape.

However, the above is only the preferred embodiment of the present invention, and it does not limit the patent scope of the present invention. Therefore, all the simple modifications and equivalent structural changes caused by the description and drawings of the present invention should be the same. It is included in the patent scope of the present invention and is conjoined with Chen Ming.

To sum up, when the system and method for liability analysis of car accident accidents of the present invention are used, in order to indeed achieve their efficacy and purpose, the present invention is truly an invention with excellent practicability, which is in accordance with the application requirements of invention patents, To file an application in accordance with the law, I hope that the review committee will grant the invention as soon as possible to protect the inventor’s hard invention. If there is any doubt in the review committee of the Jun Bureau, please send me a letter and give instructions. The inventor will try his best to cooperate and feel virtuous.

1‧‧‧Monitoring device

2‧‧‧Car accident identification module

21‧‧‧Sound Interpretation Unit

3‧‧‧Cloud identification system

4‧‧‧Monitoring Center

5‧‧‧Search device

51‧‧‧Positioning module

52‧‧‧Electronic tag recognition module

6‧‧‧ car accident analysis software

7‧‧‧Mobile electronic device

Claims (10)

A car accident accident responsibility analysis system mainly includes: a plurality of monitoring devices for recording surveillance audio and video; a plurality of car accident identification modules respectively provided in each of the monitoring devices recognizes whether a car accident has occurred according to the monitoring video and audio, and each car accident The identification module includes a sound interpretation unit, which is used to return the instantaneous high-decibel information to the car accident identification module; a cloud identification system that connects the information to the car accident identification modules is used to assess the severity of the car accident; The information link to the monitoring center of the cloud identification system sends at least one car accident broadcast message according to the car accident status; at least one information link to the monitoring center search device is used to search for the sending target of the car accident broadcast message, and the search device includes At least one positioning module, and at least one electronic tag recognition module provided on one side of the road; and at least one car accident analysis software that links the search device to at least one driver's mobile electronic device for reception And display the broadcast message of the car accident. For example, the car accident accident responsibility analysis system described in item 1 of the patent application scope, wherein the cloud identification system includes an audio and video recording module for backing up the monitored audio and video. As described in item 1 of the scope of patent application, the responsibility analysis system for car accident accidents, wherein the monitoring center includes a distance calculation module that links the positioning module with information to limit the distance range of the car accident broadcast message. The accident analysis responsibility system as described in item 1 of the patent application scope, wherein the cloud identification system includes a fire identification module with a selective information link to the fire unit, and a safety identification module with a selective information link to the medical unit . The accident analysis responsibility system as described in item 1 of the patent application scope, wherein the accident analysis software includes an information link to the reconciliation unit of the monitoring center, which generates a settlement information and attaches it to the accident broadcast message. A method for analysing the responsibility of a car accident, the steps of which include: (a) recording and monitoring audio and video on road conditions using a plurality of monitoring devices; (b) when a car accident occurs, a sound interpretation unit is used to transmit instantaneous high decibel information to each car accident identification Modules; (c) Based on the monitored audio and video, a plurality of car accident identification modules identify whether a car accident has occurred; (d) Based on the monitored audio and video and the high decibel message, a cloud recognition system assesses the severity of the car accident status; (e) According to the situation of the car accident, a monitoring center sends at least one car accident broadcast message; (f) uses a positioning device of a search device and an electronic label recognition module to search for the transmission target of the car accident broadcast message; (g) enables at least one driver The mobile electronic device receives and displays the car accident broadcast message through a car accident analysis software. The method for analyzing the liability for accidents caused by a car accident as described in item 6 of the patent application scope further includes the step (c1) when the car accident identification module confirms that a car accident has occurred, a video recording module is used to back up the monitored video. The method for analyzing the responsibility for accidents caused by a car accident as described in item 6 of the patent application scope, wherein a distance calculation module can be used to limit the distance range of the car accident broadcast message before step (e) is performed. The method for analyzing the responsibility for accidents caused by car accidents as described in item 6 of the patent application scope, which further includes step (d1). The cloud identification system uses a fire identification module to determine whether to notify the fire unit, and a personnel safety identification module to determine whether to report Medical unit. The method for analyzing the responsibility for accidents caused by cars as described in item 6 of the patent application scope, which further includes step (h). The driver of the car accident event uses the reconciliation unit of the car accident analysis software to generate a reconciliation information and append it to the car accident broadcast message.

Images