TW201822168A - Vehicle moving direction predicting system and method using digital image recognition in combination with moving trace computation technology with application of big data computation technology - Google Patents

Abstract

The present invention provides a vehicle moving direction predicting system and method, which uses digital image recognition in combination with moving trace computation technology, with application of big data computation technology, to provide an accurate moving direction predicting system. The present invention has three units including a road condition weight control unit, time and traffic condition weight control unit, and vehicle moving trace weight control unit, which carry out the calculation of weights according to the environmental factors including the person, event, time, place, and object of a vehicle and uses the statistic result of previous vehicle moving trace data as a basis for prediction, also involving the instantaneously traffic condition, traffic flow, and road condition, to provide a dynamic and instantaneous prediction of a potential moving direction of a target vehicle, in order to facilitate police to track a suspect vehicle.

Description

   Vehicle driving direction prediction system and method   

The invention is a system and method for predicting a vehicle's driving direction by extracting environmental factors such as people, events, times, places, and things for weight control to analyze and judge.

As the traffic flow grows year by year and traffic accidents are increasing, the police use smart intersection image monitoring and license plate recognition systems to assist in handling cases, find stolen license plates, the Police Department has reported increasing demand for seized vehicles, accident escape cars and rounded up vehicles. In the past, although the vehicle real-time image trajectory tracking system can provide direction prediction for the reference of the police, the prediction result is still not accurate. Therefore, in order to effectively improve the way of service deployment and police use when rounding up criminal vehicles, how can we use intelligent The alarm vehicle interception system is integrated with the integration of the Yunlong trajectory system and the police vehicle positioning dispatch system to achieve real-time grasp of the location of the criminal vehicle and the nearby police resources. With reference to the prediction results of the vehicle driving direction pre-analysis system, it is necessary to further appropriately deploy service deployment to make the same The police resources can achieve the best hunting efficiency, which is a subject for all parties to study hard.

The invention provides a vehicle driving direction prediction system, including: an intersection surveillance camera module, which captures an instant traffic image of an intersection and license plate information of a target vehicle, wherein the intersection surveillance camera module includes a plurality of surveillance cameras . A smart image management module is based on the intersection traffic image and license plate information to determine the flow velocity of a vehicle at an intersection, and records the customary travel time and route corresponding to the license plate information of the target vehicle. The smart image management module also includes . A historical trajectory analysis unit records the time and route of the habitual travel corresponding to the license plate information of the target vehicle. A real-time road condition analysis unit is used to analyze the current vehicle flow rate through the intersection traffic image. An environmental factor extraction unit generates an environmental variable based on an environmental and road condition data, wherein the environmental variable is a road type, a target vehicle type and a driving situation, real-time traffic flow and a habitual path. A vehicle driving direction prediction calculation module is used to predict the driving direction of a target vehicle through environmental variables, the current vehicle speed, the time and route used to travel. A network communication interface module provides transmission between intersection surveillance camera modules, intelligent image management modules, historical trajectory analysis units, real-time road condition analysis units, environmental factor extraction units, and vehicle driving direction prediction calculation modules.

The vehicle driving direction prediction calculation module calculates the probability of the target vehicle passing through the intersection according to the customary travel time and route corresponding to the license plate information of the target vehicle through a line of vehicle trajectory weight control units.

The vehicle driving direction prediction calculation module calculates the probability of the target vehicle passing through the intersection through a time and road condition weight control unit based on the current vehicle flow rate.

The vehicle driving direction prediction calculation module uses a road condition weight control unit to calculate the probability of the target vehicle passing through the intersection according to the road width and speed limit of the intersection.

The invention provides a method for predicting a vehicle's driving direction. The steps include: an intersection monitoring camera module captures an instant traffic image of an intersection and a license plate information of a target vehicle and sends the information to an intelligent image management module; the intelligent image management module; The image management module judges the flow rate of a vehicle at an intersection and the license plate information of the target vehicle through the traffic image and license plate information at the intersection, including the license plate, model, and color of the target vehicle, and stores the customary travel corresponding to the license plate information of the target vehicle. Time and route; An instantaneous road condition analysis unit is used to analyze the current vehicle flow rate, a historical trajectory analysis unit is used to analyze the time and route of the habitual travel of the target vehicle, and an environmental factor extraction unit generates an environmental variable based on environmental and road condition data; And a vehicle driving direction prediction calculation module performs prediction of the driving direction of the target vehicle through environmental variables, current vehicle flow velocity, time and route used to travel.

The invention proposes a vehicle driving direction prediction system, which uses the vehicle's driving environment to mutate parameters and refer to the vehicle's past habitually selected path. After weight calculation, a more intelligent prediction result can be obtained, which helps the police to be more informed during the round-up process. Accurately grasp the route of the criminal vehicle and successfully complete the arrest. Compared with the prior art, the present invention is characterized in that it can extract environmental factors such as people, events, times, places and things for weight control. The environmental factors are briefly described as follows:

1. The "person" factor refers to the driving habits of the target vehicle. Based on the historical track data of the past days and times recorded in the database, based on the past data, analyze the driving habits of the target vehicle to find out the choice of past habits. Route as a reference for predicting the direction of travel of the vehicle.

2. The "event" factor refers to the driving situation of the target vehicle, the current escaped vehicle being chased by the police and the problem vehicle notified for detection. The driving route chosen by the two is based on past experience and statistics. There are some differences. , Will also affect the prediction result of the vehicle driving direction.

3. The "time" factor refers to the travel time of the target vehicle. The real-time traffic volume is a factor that affects the prediction of the vehicle's driving direction. The rush hour traffic volume on the main road is relatively large. The current escape vehicles are less likely to choose traffic jam sections. In addition, the status of the sign will also indirectly affect the prediction of the direction of the vehicle. When a vehicle on the road stops at the red light according to the sign, it is like a traffic jam to the target vehicle. If the intersection can turn left or right, the target vehicle chooses The chance of turning left and right will increase.

4. The "ground" factor refers to the types of roads that the target vehicle may choose to drive at the intersection, such as: main roads, branch roads, lanes, one-way streets, tailless lanes, etc. Generally speaking, driving Vehicles are more likely to choose straight main roads than other roads.

5. The "material" factor refers to the type of target vehicle (such as large cars, small cars, treadmills, etc.). Vehicle types can be used to filter inaccessible road sections, exclude unreasonable predictions, and improve predictions. The accuracy of the results. For example, large cars cannot enter narrow alley roads and single-lane roads with high traffic volume, and the possibility of reverse driving is very low.

The above detailed description is a specific description of a feasible embodiment of the present invention, but this embodiment is not intended to limit the patent scope of the present invention. Any equivalent implementation or change that does not depart from the technical spirit of the present invention should be included in Within the scope of the patent in this case.

To sum up, this case is not only innovative in terms of space type, but also enhances the above-mentioned multiple effects over conventional items. It should have fully met the requirements for statutory invention patents that are novel and progressive. Apply for it in accordance with the law and ask your office for approval. This invention patent application is designed to encourage inventions, and it is a matter of virtue.

100‧‧‧Intersection surveillance camera module

200‧‧‧Smart Image Management Module

300‧‧‧History Track Analysis Unit

400‧‧‧Real-time traffic analysis unit

500‧‧‧Environmental factor extraction unit

600‧‧‧Vehicle driving direction prediction calculation module

700‧‧‧ network communication interface module

110‧‧‧ Surveillance Camera

610‧‧‧travel track weight control unit

620‧‧‧time and road weight control unit

630‧‧‧ road condition weight control unit

S201 ~ S204‧‧‧step flow

FIG. 1 is a structural diagram of a vehicle driving direction prediction system of the present invention.

FIG. 2 is a flowchart of a method for predicting a driving direction of a vehicle according to the present invention.

In order for the reviewing committee members to understand the technical features, contents and advantages of the present invention and the effects that can be achieved, the present invention is described in detail with the accompanying drawings in the form of embodiments, and the drawings used therein are The main purpose is only for the purpose of illustration and supplementary description. It may not be the actual proportion and precise configuration after the implementation of the invention. Therefore, the attached drawings should not be interpreted and limited to the scope of rights of the present invention in actual implementation. He Xianming.

Please refer to FIG. 1. As shown in the figure, it is a structural diagram of a vehicle driving direction prediction system according to the present invention, which includes an intersection monitoring camera module 100, an intelligent image management module 200, a historical trajectory analysis unit 300, and a real-time road condition analysis unit. 400, an environmental factor extraction unit 500, a vehicle driving direction prediction calculation module 600, and a network communication interface module 700, among which each module is transmitted through the network communication interface module 700, and the use of the intersection monitoring camera module 100 is more The surveillance cameras 110 transmit the image to the intelligent image management module 200 through the network communication interface module 700. The intelligent image management module 200 identifies the characteristics of the target vehicle (such as license plate, vehicle model, and vehicle color), and recognizes The results are stored in a database. The recognition result is calculated by the historical trajectory analysis unit 300 and the real-time road condition analysis unit 400. The environmental factor extraction unit 500 can extract the road type, target vehicle type and driving situation, real-time traffic volume, and habitual path of the target vehicle from the analysis results. The environmental factors are calculated by the trajectory weight control unit 610, time and road condition weight control unit 620, and road condition weight control unit 630, and then weighted by the vehicle driving direction prediction calculation module 600. Finally, the target vehicle can be obtained. Direction prediction results.

Please refer to FIG. 2. As shown in the figure, the flowchart of the method for predicting the direction of vehicle travel includes the following steps: S201: The intersection monitoring camera module captures the real-time traffic image of the intersection and the license plate information of the target vehicle and sends it to the intelligent image management. Module; S202: The intelligent image management module judges the vehicle flow rate at the intersection and the license plate information of the target vehicle through the traffic image and license plate information at the intersection, and stores the time and route of the customary travel corresponding to the license plate information of the target vehicle; S203 : Using the real-time road condition analysis unit to analyze the current vehicle flow rate, the historical trajectory analysis unit is used to analyze the time and route of the habitual travel of the target vehicle, and the environmental factor extraction unit generates environmental variables based on the environment and road condition data; and S204: the vehicle direction prediction calculation The module uses the environment variables, the current vehicle flow rate, the time and route used to travel to predict the driving direction of the target vehicle.

Take a practical case to illustrate, the intersection of Zhonghua East Road and Xiaodong Road is located in the East District of Tainan City, adjacent to the Dawan Interchange. Zhonghua East Road is the main arterial road linking the north-south ring road connecting Yongkang District and Tainan District in Tainan City. The east-west main arterial road in Shidong District; suppose the police are hunting for a currently escaping vehicle. This target vehicle is about to pass from Zhonghua East Road and Xiaodong Intersection from south to north. The sample space S = {left, straight, right} for the direction of the vehicle. P (left) is the probability of the vehicle turning to the left, P (right) is the probability of the vehicle turning to the right, and P (straight) is the probability of the vehicle going straight, and the weight is calculated according to the following formula: The driving situation adjustment factor fs, if fs = current escape, increase the weight of the large road. According to statistics, the turning probability of vehicles passing this intersection from south to north is P (left) = 9%, P (straight) = 72%, and P (right) = 19%. The hunting time is 18:30 in the afternoon on weekdays. The rush hour of off-duty traffic, the historical data statistics show that the vehicle steering probability for this period is P (left) = 7%, P (straight) = 80%, P (right) = 13%, based on the driving trajectory of this target vehicle in the past According to the recorded statistical results, it was found that the route selected by the target vehicle from south to north via this intersection was P (left) = 6%, P (straight) = 42%, P (right) = 52%, and passed the weight control unit. After calculation, the prediction results are as follows:

To sum up, this case is not only innovative in terms of technical ideas, but also has many of the above-mentioned effects that are not used by traditional methods. It has fully met the requirements of statutory invention patents that are novel and progressive. To approve this invention patent application, to encourage invention, to the utmost convenience.

Claims (12)

    A vehicle driving direction prediction system includes: an intersection monitoring camera module, which captures an instant traffic image of an intersection and license plate information of a target vehicle; an intelligent image management module, which uses the intersection traffic image and The license plate information determines the flow velocity of a vehicle at the intersection, and records the time and route of customary travel corresponding to the license plate information of the target vehicle. The intelligent image management module further includes: a historical track analysis unit, which records The time and route of the habitual travel of the target vehicle corresponding to the license plate information; a real-time road condition analysis unit that analyzes the current vehicle flow rate through the traffic image at the intersection; an environmental factor extraction unit that generates data based on an environment and road condition data An environment variable; a vehicle driving direction prediction calculation module for predicting the driving direction of the target vehicle through the environmental variable, the current vehicle flow velocity, the time and route used to travel; and a network communication interface module, Provide the intersection surveillance camera module, the intelligent image management module, the history Trace analysis unit, the real-time traffic analysis unit, the extracting unit and the environmental factors traveling direction of the vehicle between the transmission of the prediction calculation module.         The vehicle driving direction prediction system according to item 1 of the patent application scope, wherein the intersection surveillance camera module includes a plurality of surveillance cameras.         The vehicle driving direction prediction system according to item 1 of the scope of the patent application, wherein the environmental variables are road types, target vehicle types and driving situations, real-time traffic flow, and customary paths.         The vehicle driving direction prediction system according to item 1 of the scope of the patent application, wherein the vehicle driving direction prediction calculation module is based on a line of trajectory weight control units, according to the time of customary travel corresponding to the license plate information of the target vehicle and Route, calculate the probability of the target vehicle passing through the intersection.         The vehicle driving direction prediction system according to item 1 of the patent application scope, wherein the vehicle driving direction prediction calculation module calculates the probability of the target vehicle passing the intersection through a time and road condition weight control unit based on the current vehicle flow rate.         The vehicle driving direction prediction system according to item 1 of the scope of the patent application, wherein the vehicle driving direction prediction calculation module calculates a target vehicle passing through the intersection through a road condition weight control unit according to the road width and speed limit of the intersection The probability.         A vehicle driving direction prediction method includes the steps of: an intersection monitoring camera module captures an instant traffic image of an intersection and a license plate information of a target vehicle and sends it to a smart image management module; the smart image management The module judges the flow velocity of a vehicle at the intersection and the license plate information of the target vehicle through the traffic image and the license plate information of the intersection, and stores the time and route of the customary travel corresponding to the license plate information of the target vehicle. The real-time road condition analysis unit analyzes the current vehicle flow rate, a historical trajectory analysis unit is divided into the habitual travel time and route corresponding to the target vehicle, an environmental factor extraction unit generates an environmental variable based on an environment and road condition data; and a vehicle travels The direction prediction calculation module performs prediction of the driving direction of the target vehicle through the environment variable, the current flow velocity of the vehicle, the time and route used to travel.         The method for predicting the direction of travel of a vehicle as described in item 7 of the scope of patent application, wherein the license plate information includes the license plate, model, and color of the target vehicle.         The method for predicting the driving direction of a vehicle as described in item 7 of the scope of the patent application, wherein the environmental variables are road types, target vehicle types and driving situations, real-time traffic flow, and customary paths.         The method for predicting the direction of travel of a vehicle according to item 7 in the scope of the patent application, wherein the calculation module of the direction of travel prediction of the vehicle is based on a line of trajectory weight control units, according to the time of customary travel corresponding to the license plate information of the target vehicle and Route, calculate the probability of the target vehicle passing through the intersection.         The method for predicting the driving direction of a vehicle according to item 7 in the scope of the patent application, wherein the vehicle driving direction prediction calculation module calculates the probability of the target vehicle passing the intersection through a time and road condition weight control unit based on the current vehicle flow rate.         The method for predicting the driving direction of a vehicle according to item 7 in the scope of the patent application, wherein the vehicle driving direction prediction calculation module calculates the target vehicle passing through the intersection through a road condition weight control unit according to the road width and speed limit of the intersection The probability.