WO2015117528A1

WO2015117528A1 - Car driving record processing method and system

Info

Publication number: WO2015117528A1
Application number: PCT/CN2014/095686
Authority: WO
Inventors: 赵宇
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-07-24
Filing date: 2014-12-30
Publication date: 2015-08-13
Also published as: CN105279814A

Abstract

A car driving record processing method and system. The method comprises the following steps: acquiring car driving information in real time; according to a category of pre-set target feature information, extracting feature information matching with the category of the pre-set target feature information in the car driving information; comparing the feature information with the pre-set target feature information to obtain a matching value; and when the matching value is greater than or equal to a pre-set matching threshold value, acquiring current position information, and uploading the current position information and the car driving information to a server. The car driving record processing method and system enrich the use function of a car driving record processing system in the related art, and improve the resource utilization rate; and meanwhile, by means of the technical solution of the present invention, the monitoring coverage area of a common security and protection system can also be expanded, thereby making up for the deficiency of an insufficient distribution number of fixed cameras of the current common security and protection system, and greatly reducing costs for constructing the common security and protection system.

Description

Driving record processing method and system

Technical field

The invention relates to the technical field of driving records, and in particular to a driving recording method and system.

Background technique

In the related art, the driving record system is mainly composed of two parts, an imaging device and a recording device, for recording image and sound information during the running of the vehicle, and storing the recorded image and sound information in a storage device (such as a hard disk or an SD card). )in. The common trip recording system has only the recording function, and does not have the ability to analyze and process the recorded image and sound information. Among them, the image and sound information recorded outside the vehicle is mainly used to evaluate the traffic accident; the image inside the vehicle is recorded and The sound information is mainly used to monitor the passenger's riding behavior to ensure the safety of the driver's personal and property. However, as a kind of mobile camera device, it is only used as a tool to assist the relevant law enforcement personnel in determining the responsibility of the traffic accident and the safety of the driver and the person and property in real life, and reduces the resource utilization rate of the driving record system.

Summary of the invention

The main object of the present invention is to solve the technical problem that the driving record processing system in the related art has a single function and cannot automatically distinguish the target feature information.

In order to achieve the above objectives, the following technical solutions are adopted:

A driving record processing method includes the following steps:

Get driving information in real time;

Extracting, according to a preset category of the target feature information, feature information that matches the category of the preset target feature information in the driving information;

Comparing the feature information with preset target feature information to obtain a matching value;

When the matching value is greater than or equal to the preset matching threshold, the current location information is acquired, and the current location information and the driving information are uploaded to the server.

Optionally, before the step of acquiring the driving information in real time, the method further includes:

Receiving target feature information pushed by the server and storing it; or,

Sending a request instruction to the server, so that the server responds to the request instruction and pushes the target feature information; receives the target feature information pushed by the server and stores it.

Optionally, the feature information includes facial feature information and/or vocal feature information; or

The feature information includes vehicle license plate feature information.

Optionally, when the feature information includes the face feature information and the vocal feature information, the comparing the feature information with the preset target feature information, and obtaining the matching value includes:

Comparing the face feature information and the vocal feature information with preset target feature information, respectively, to obtain a face matching value and a vocal match value;

And weighting the face matching value and the vocal matching value to obtain the matching value.

Optionally, the method further includes:

When the matching value is greater than or equal to a preset matching threshold, an interaction interface is generated and displayed for the user to perform an alarm based on the interaction interface.

A driving record processing system, comprising a driving information acquiring module, an information extracting module, a data processing module, a position information acquiring module and a communication module, wherein:

The driving information obtaining module is configured to: obtain driving information in real time;

The information extraction module is configured to: extract, according to a category of the preset target feature information, feature information that matches the category of the preset target feature information in the driving information;

The data processing module is configured to: compare the feature information with preset target feature information to obtain a matching value;

The location information obtaining module is configured to: obtain current location information when the matching value is greater than or equal to a preset matching threshold;

The communication module is configured to: upload the current location information and the driving information to a server.

Optionally, the driving record processing system further includes an update module, wherein:

The update module is configured to:

Receiving target feature information pushed by the server and storing it; or,

Sending a request instruction to the server, causing the server to respond to the request instruction and pushing the target feature information, and receiving the target feature information pushed by the server and storing.

The feature information includes vehicle license plate feature information.

Optionally, the data processing module includes a first processing unit and a second processing unit, where:

The first processing unit is configured to: compare the facial feature information and the vocal feature information with preset target feature information, and obtain a face matching value and a vocal matching value;

The second processing unit is configured to perform weighting processing on the face matching value and the vocal matching value to obtain the matching value.

Optionally, the system further includes an interface generation module, wherein:

The interaction interface generation module is configured to: when the matching value is greater than or equal to a preset matching threshold, generate and display an interaction interface for the user to perform an alarm based on the interaction interface.

A computer program comprising program instructions that, when executed by a computer, cause the computer to perform any of the above-described driving record processing methods.

A carrier carrying the computer program.

A driving record processing method provided by an embodiment of the present invention automatically analyzes and processes driving information in the process of acquiring driving information, extracts related feature information from driving information, and combines the characteristic information with a preset target. The feature information is compared to determine whether there is information matching the preset target feature information in the driving information acquired in real time; when the driving information has information matching the preset target feature information, obtaining the current location information Upload the current location information and driving information to the server. Therefore, the embodiment of the invention enriches the use function of the driving record processing system in the related art, and improves the resource utilization rate; meanwhile, the technical solution of the invention can also expand the monitoring coverage of the public security system to make up for the current public security system. The shortage of fixed camera distribution greatly reduces the cost of building a public security system.

BRIEF abstract

1 is a schematic flow chart of a driving record processing method according to a first embodiment of the present invention;

2 is a schematic flow chart of a driving record processing method according to a second embodiment of the present invention;

3 is a schematic flow chart of a driving record processing method according to a third embodiment of the present invention;

4 is a schematic flow chart of a driving record processing method according to a fourth embodiment of the present invention;

FIG. 5 is a schematic diagram showing a refinement process of matching a feature information with a preset target feature information in a driving record processing method according to an embodiment of the present invention;

6 is a detailed flowchart of applying a driving record processing method to a taxi according to an embodiment of the present invention;

7 is a hardware architecture diagram of an operating environment of a driving record processing system according to an embodiment of the present invention;

8 is a schematic diagram of functional modules of a driving record processing system according to a first embodiment of the present invention;

9 is a schematic diagram of functional modules of a driving record processing system according to a second embodiment of the present invention;

10 is a schematic diagram of functional modules of a driving record processing system according to a third embodiment of the present invention;

11 is a refinement function of a data processing module in a driving record processing system according to an embodiment of the present invention; Module schematic.

The implementation, functional features, and advantages of the present invention will be further described in conjunction with the embodiments.

Preferred embodiment of the invention

It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Embodiments of the present invention provide a driving record processing method.

Referring to FIG. 1, in the first embodiment, the driving record processing method includes the following steps:

Step S10, obtaining driving information in real time;

In this step, for example, the driving information is acquired by the camera device, and the camera device may be a camera with a microphone or any other suitable camera. The camera is installed inside or outside the vehicle, and can record the internal condition of the vehicle, or Record the external condition of the vehicle. For example, the camera is installed in a taxi to monitor the passenger behavior of the passengers in the car; for example, the camera is installed in a private car for recording the live information of the road on which the vehicle is traveling. Generally, when the user starts the vehicle, the camera device is turned on, thereby obtaining the driving information of the entire driving process of the vehicle. The driving information of the embodiment may be video data, or may be a combination of video data and audio data, during the driving process. The driving information is written into the buffer of the relevant storage device to improve the storage efficiency of the driving information, and at the same time facilitate the analysis and processing of the driving information.

It is worth mentioning that in order to record driving information for a long time under limited storage space, during the driving process, the previously cached driving information is discarded every preset driving time, for example, the time interval is set to 30 minutes. Taking 30 minutes as a recording time period, the time point at which the driving information is started is used as the starting point of the timing. When the driving time reaches 60 minutes, the driving information cached in the first 30 minutes is discarded; when the driving time reaches 90 minutes, the discarded 30 is discarded. The driving information cached in the period from minute to 60 minutes, and so on, under the premise of preserving the driving information recorded in the latest driving time, reduces the storage capacity requirement of the storage device, thereby reducing the use cost.

Step S20, extracting driving information from the preset according to the category of the preset target feature information. Feature information of category matching of target feature information;

In this step, the target feature information may be human body feature information (such as a face image, a voice print, etc.), or may be other related object feature information (such as a license plate number, a body color, etc.), for example, using the target feature information to the person The identity is identified to determine the target person; or the target feature information is used to identify the vehicle to determine the target vehicle. According to different target feature information, such as the above-mentioned human body feature information, in order to improve the accuracy of the identity recognition, a face image or a voiceprint is generally used as the target feature information, and if the target feature information is selected as a face image, The corresponding category is a human body image; if the target feature information is selected as a voiceprint, the corresponding category is a human voice. It can be understood that the driving information generated by the recording includes a large amount of data. In order to facilitate the subsequent analysis processing, the driving information needs to be filtered, and specific information is extracted therefrom, that is, matching the category of the preset target characteristic information. The feature information, as described above, if the category of the preset target feature information is a human body image, all human body images are extracted from the recorded image of the driving information, and the face image is extracted from each human body image, thereby achieving Reduce the amount of data processing and increase the speed of data processing.

Step S30, comparing the feature information with the preset target feature information to obtain a matching value;

In this step, the feature information extracted in the above step S20 is compared with the preset target feature information to obtain a matching value, which is expressed by a percentage, and may of course be represented by other forms of values, such as decimals and fractions. And determining the correlation between the feature information and the preset target feature information according to the matching value.

For example, in an implementation manner, the preset target feature information is a face image, and correspondingly, the feature information is a face image, and the comparison algorithm needs to be determined before performing the comparison processing, for example, using a face feature point based recognition algorithm. And a face recognition algorithm based on a whole face image recognition algorithm and a template-based recognition algorithm, according to the adopted face recognition algorithm, the preset face image and the real-time acquired face image are performed. Processing, extracting facial image features, and modeling facial features, such as faces composed of eyes, nose, mouth, and chin, and geometric descriptions of these parts and their structural relationships can be used as face recognition An important feature. It can be understood that since the extraction of facial features and face recognition algorithms are known techniques and are well known to those skilled in the art, they are not described again. Extracting the feature data of the acquired face image and matching the feature template of the stored target feature information, that is, the face feature to be recognized and the obtained face feature template (ie, pre- The face feature template is compared, and then a matching value is obtained, and the matching value can be expressed by a percentage. According to the size of the matching value, the degree of similarity between the facial feature to be recognized and the obtained facial feature template can be determined. Thus, the identity information of the face is judged.

Referring to FIG. 5, in order to improve the accuracy of identifying the target person, when the feature information includes the face feature information and the vocal feature information, the foregoing step S30 specifically includes:

Step S301, comparing the face feature information and the vocal feature information with the preset target feature information, respectively, to obtain a face matching value and a vocal match value;

Step S302, performing weighting processing on the face matching value and the vocal matching value to obtain a matching value.

In this embodiment, the face matching value is obtained by the face recognition technology, for example, the face matching value is 60%; similar to the above process of face recognition, the feature extraction is also required before the voiceprint recognition, and then according to The specific algorithm identifies the vocal to be recognized. Since the voiceprint recognition technique is known to those skilled in the art, it will not be described again, for example, the vocal match value obtained by the voiceprint recognition technique is 40%. It is easy to know that the accuracy of determining the identity information by using the voiceprint feature is low. Therefore, when performing data processing, the weight distribution of the identity information should be determined according to the face feature and the voiceprint feature, for example, face matching. The weight of the value is 0.7, and the weight of the vocal match value is 0.3, then the weighted processing results in a final matching value of 54%.

For another example, in another embodiment, the preset target feature information is a license plate number, and correspondingly, the feature information is a license plate number. Usually, the license plate number consists of the motor vehicle registration agency code and number plate number, for example, Guangdong B·123XXX, so the license plate number can be equivalent to a series, and the equivalent series consists of Chinese characters, English letters and numbers. . In this embodiment, the obtained license plate number is defined as a reference sequence, and the license plate number to be identified is defined as a sequence to be identified, and the comparison between the sequence to be identified and the reference sequence has various methods, for example, the sequence to be identified is The reference series is compared horizontally, that is, from the first item of the series, in the order of the order of the series, until the last item of the series, and the number of items to be identified is equal to the number of items in the reference sequence. For example, the reference number is Guangdong B·123456, and the number to be identified is Guangdong B·167356. According to the above horizontal comparison method, the number of identical items in the sequence to be identified and the reference sequence is 5, so the feature information and the preset target are obtained. The matching value obtained after the feature information is compared is 62.5%. Of course, the sequence to be identified can also be discretely compared with the reference sequence, that is, the order of the columns is not considered in the comparison process, and only the number of the same content is counted.

In an emergency, it is often impossible to accurately remember the license plate number of the escaped vehicle, that is, the preset target feature information is not necessarily completely accurate, and in order to improve the accuracy of the preset target feature information, preferably, the preset target The feature information may include a variety of information, in addition to the license plate number, the preset target feature information further includes other vehicle feature information, such as the color of the vehicle, and accordingly, the vehicle color information needs to be extracted from the driving information, usually When comparing the colors of the vehicles, the matching values obtained are generally 0 and 1.

It should be noted that the specific numerical values in this embodiment are for illustrative purposes only, and are not intended to limit the related technical features in the driving record processing method of the embodiment of the present invention.

Step S40: Whether the matching value is greater than or equal to a preset matching threshold;

In this step, the matching value obtained in the above step S30 is compared with a preset matching threshold, for example, the preset matching threshold is 80%, and when the matching value is greater than or equal to 80%, the feature information and the preset are considered. The target feature information is related. For example, the similarity between the face image extracted from the driving information and the preset facial feature template reaches 80%, thereby determining the identity information of the face; for example, extracting from the driving information. The similarity between the license plate number and the preset license plate number reaches 80%, so the vehicle information corresponding to the license plate number is determined. It should be understood that the preset matching threshold of 80% is only an example, and the preset matching threshold size can be flexibly selected in practical applications.

When the matching value is less than the preset matching value, the process jumps to the above step S10; when the matching value is greater than or equal to the preset matching threshold, the process jumps to the following step S50.

Step S50, acquiring current location information;

In this step, the matching value is greater than or equal to the preset matching threshold, so the feature information extracted from the driving information matches the preset target feature information, that is, the target person or the target vehicle is found. At this time, the current location information of the target person or the target vehicle is determined, thereby facilitating the search target person or the target vehicle. In this embodiment, the current location information may be acquired by using any other suitable positioning device such as a car GPS, a communication base station, or the like. It should be understood that related devices (such as a camera and a microphone) that acquire driving information are installed on the user's vehicle. Generally, the target person and the target vehicle and the user's vehicle are all in the same small range, so the current position information of the user's vehicle is taken as The current location information of the target person and the target vehicle. In a preferred embodiment, the current location information may be composed of latitude and longitude coordinates, or may be formed by a specific screen area displayed on the electronic map. Of course, the current location information may also be composed of any other applicable location information. This Do not repeat them.

In step S60, the current location information and the driving information are uploaded to the server.

In this step, the current location information and the driving information are uploaded to the server through a wireless network (such as Wi-Fi, 3G network, 4G network, etc.), for example, the server is a monitoring center of the urban security system, and receives the transmission from the user's vehicle. Current location information and driving information to track suspects and lost vehicles. When the driving information is video data, one frame of the feature information (such as the face of the target person) may be uploaded, or a video in a specific time interval may be uploaded, but the video should contain the feature information; When the information is a combination of video data and audio data, in addition to uploading a recording in a specific time interval, a video of a specific time interval of one frame of the feature information is uploaded. It can be understood that the recording is uploaded to the server only when the feature information relates to the person. The uploaded driving information is used to identify the suspect and the lost vehicle on the one hand, and to assist in determining the geographical location of the suspect and the lost vehicle on the other hand.

In summary, the embodiments of the present invention enrich the use function of the driving record processing system in the related art, and improve the resource utilization rate. At the same time, the technical solution of the present invention can also expand the monitoring coverage of the public security system to make up for the current public. The lack of a fixed number of cameras in the security system greatly reduces the cost of building a public security system.

Referring to FIG. 2, based on the first embodiment, a driving record processing method according to a second embodiment of the present invention is proposed. Before the step S10, the embodiment further includes:

Step S70, receiving target feature information pushed by the server and storing it;

In this step, the target feature information is stored on the server in various forms. For example, publicly wanted criminal information in the public, in order to facilitate public identification, usually a positive photo of the criminal and a recording, so the corresponding target feature information is stored on the server in the form of pictures and recordings; for example, publicly found in the society. The lost vehicle information is usually the license plate number of the lost vehicle, the body color and the vehicle brand, etc., so the corresponding target feature information is stored in the form of text on the server. The server stores the latest target feature information. In this embodiment, the server pushes the target feature information to the user vehicle by means of broadcasting, and the target feature information (that is, the preset target feature information) stored on the user's vehicle needs to be updated. At this time, the target feature information pushed by the server is received and stored, so that the target feature information on the user's vehicle is in the latest state.

It is worth mentioning that, in order to simplify the process of data processing, the server packs and pushes multiple target feature information, and the user vehicle adopts a passive receiving mode, and does not need to target the target feature information contained in the target feature information packet pushed by the server with the preset The target feature information is compared to determine whether an update is needed. When the user vehicle receives the target feature information package and stores it, all the previous target feature information is deleted, thereby saving storage space of the storage device.

Referring to FIG. 3, based on the first embodiment, a driving record processing method according to a third embodiment of the present invention is proposed. In contrast to the second embodiment, the user vehicle actively sends a request instruction to the server according to actual needs, so that the server The user vehicle pushes the target feature information, and the embodiment further includes: before the step S10:

Step S80, sending a request instruction to the server, so that the server responds to the request instruction and pushes the target feature information;

In this step, the user vehicle sends a request instruction to the server at a certain time (for example, 30 minutes), and the server responds to the request instruction and pushes the target feature information after receiving the request instruction.

In step S90, the target feature information pushed by the server is received and stored.

Referring to FIG. 4, based on the first embodiment, a driving record processing method according to a fourth embodiment of the present invention is proposed. After the step S40, the embodiment further includes:

Step S100, generating and displaying an interactive interface;

In this step, when the matching value is greater than or equal to the preset matching threshold, the user vehicle uploads the current location information and the driving information to the server, thereby providing relevant public safety agencies (such as public security organs) with tracing suspects and lost vehicles. Information, at the same time, also generate and display an interactive interface for the user to perform an alarm based on the interactive interface, for example, input a piece of text on the interactive interface, and send the segment text to the alarm center (such as 110 service desk) through the wireless network; For example, by triggering an alarm button (such as a touch screen display) displayed on the interactive interface to dial an alarm, and submitting relevant information to the alarm center in the form of voice. As a result, the discovered target person and the target vehicle can be reported in time.

Taking the taxi as an example, the driving record processing method of the embodiment of the present invention is exemplified.

Referring to Figure 6, the method includes:

In step S201, the photo and recording of the criminal suspect pushed by the server (such as the monitoring center of the public security organ) are received and stored in the SD card, wherein the photo of the criminal suspect is a positive avatar.

In step S202, the camera records the passenger in the taxi, the microphone records the voice of the passenger, and the acquired video data and audio data are stored in the SD card.

Step S203: extracting a facial feature from the video data, and extracting a voiceprint feature from the audio data, for example, the facial feature is a clear facial image, and the voiceprint feature is a human voice after eliminating noise.

Step S204, comparing the facial features of the passenger with the facial features of the criminal suspect, and simultaneously comparing the voiceprint characteristics of the passenger with the voiceprint characteristics of the criminal suspect, respectively obtaining the face matching value and the human voice. The matching value is weighted by the face matching value and the vocal matching value, and finally a matching value is obtained.

In step S205, it is determined whether the matching value is a preset matching threshold (for example, 80%). If the matching value exceeds 80%, the process goes to step S206; otherwise, the process goes to step S202.

Step S206: Acquire current location information, for example, the current location information is obtained by using a vehicle GPS.

In step S207, the current location information and the driving information (photograph and recording of the passenger) are transmitted to the server.

It can be seen that by combining the taxis that are shuttled in various corners of the city with the driving record processing system of the embodiment of the present invention, the monitoring coverage of the public security system is expanded, which makes up for the shortage of the fixed number of fixed cameras in the current public security system, and greatly reduces The cost of building a public security system.

The embodiment of the invention also provides a driving record processing system.

Referring to FIG. 7, an embodiment of the present invention provides a driving recorder for providing an operating environment for a driving record processing system. The driving recorder may include at least one processor 10, such as a CPU, at least one network interface 40, and a user interface 70. The memory 20, the camera 50, the microphone 60, and the display screen 30. The user interface 70 may include an SD card, a hard disk, and the like. The network interface 40 can optionally include a standard wired interface, a wireless interface (such as a WI-FI interface, a 3G network interface, etc.). The memory 20 may be a high speed RAM memory or a non-volatile memory such as at least one disk memory. The memory 20 may optionally be at least one Located at a storage device remote from the aforementioned processor 10. As shown in FIG. 7, the memory 20 as a computer storage medium may include a driving record processing system 200, and may of course include other system modules such as an operating system. The driving recorder is usually installed inside the vehicle, and the camera 50 can be facing the inside of the vehicle, or the camera 50 can be facing the outside of the vehicle, for example, a driving recorder on a taxi is installed, and the camera 50 is facing the inside of the vehicle for recording. The passenger journey of the rear passengers.

It should be noted that the driving record processing system 200 of the embodiment of the present invention can be run on any other suitable user terminal, such as a smart phone, a tablet computer, or the like.

Referring to Figure 8, a first embodiment of the driving record processing system 200 of the present invention is presented. The driving record processing system 200 includes:

The driving information obtaining module 210 is configured to obtain driving information in real time. For example, the driving information is acquired by the camera 50 and the microphone 60. The camera 50 and the microphone 60 are installed inside or outside the vehicle, and the internal condition of the vehicle can be recorded, and the external condition of the vehicle can be recorded. For example, the camera 50 is installed in a taxi for monitoring the riding behavior of the passengers in the vehicle; for example, the camera 50 is installed in a private car for recording the live information of the road on which the vehicle is traveling. Generally, when the user starts the vehicle, the driving recorder is turned on, thereby obtaining driving information of the whole driving process of the vehicle, wherein the driving information of the embodiment may be video data, or may be a combination of video data and audio data, during driving. The driving information is written into the buffer of the relevant storage device to improve the storage efficiency of the driving information, and at the same time facilitate the analysis and processing of the driving information.

The information extraction module 220 is configured to extract, according to the category of the preset target feature information, feature information that matches the category of the preset target feature information in the driving information. The target feature information may be human body feature information (such as a face image, a voiceprint, etc.), or may be other related object feature information (such as a license plate number, a body color, etc.), for example, using the target feature information to identify the person. Identifying to determine the target person; or using the target feature information to identify the vehicle to determine the target vehicle. According to different target feature information, such as the above-mentioned human body feature information, in order to improve the accuracy of the identity recognition, a face image or a voiceprint is generally used as the target feature information, and if the target feature information is selected as a face image, The corresponding category is a human body image; if the target feature information is selected as a voiceprint, the corresponding category is a human voice. It can be understood that the driving information generated by the recording includes a large amount of data. In order to facilitate the subsequent analysis processing, the driving information needs to be filtered, and specific information is extracted therefrom, that is, matching the category of the preset target characteristic information. The feature information, as described above, if the category of the preset target feature information is a human body image, all human body images are extracted from the recorded image of the driving information, and the face image is extracted from each human body image, thereby achieving Reduce the amount of data processing and increase the speed of data processing.

The data processing module 230 is configured to compare the feature information with the preset target feature information to obtain a matching value. Comparing the feature information extracted by the information extraction module 220 with the preset target feature information to obtain a matching value, expressed by a percentage, and of course, may also be represented by other forms of values, such as decimals, scores, etc., according to the The matching value is used to determine the correlation of the feature information with the preset target feature information.

For example, in an implementation manner, the preset target feature information is a face image, and correspondingly, the feature information is a face image, and the data processing module 230 needs to determine an alignment algorithm before performing the comparison process, and the comparison algorithm may be pre- The driving record processing system 200 is written, for example, one of a face recognition algorithm based on a face feature point recognition algorithm, a face image based recognition algorithm, and a template based recognition algorithm, according to a person employed. The face recognition algorithm processes the preset face image and the face image acquired in real time, extracts the face image feature, and models the face feature, for example, the face is composed of eyes, nose, mouth and chin. The geometric description of these parts and the structural relationship between them can be used as an important feature for recognizing human faces. It can be understood that since the extraction of facial features and face recognition algorithms are known techniques and are well known to those skilled in the art, they are not described again. The data processing module 230 matches the acquired feature data of the face image with the feature template of the stored target feature information, that is, the face feature to be recognized and the obtained face feature template (ie, the preset face feature) The template is compared, and then a matching value is obtained, and the matching value can be expressed by a percentage. According to the size of the matching value, the degree of similarity between the facial feature to be recognized and the obtained facial feature template can be determined, thereby The identity information is judged.

Referring to FIG. 11, in order to improve the accuracy of identifying a target person, when the feature information includes the face feature information and the vocal feature information, the data processing module 230 specifically includes:

The first processing unit 231 is configured to compare the facial feature information and the vocal feature information with the preset target feature information to obtain a face matching value and a vocal match value;

The second processing unit 232 is configured to perform weighting processing on the face matching value and the vocal matching value to obtain the matching value.

In this embodiment, the first processing unit 231 obtains a face matching value by using a face recognition technology, for example, a face matching value is 60%; similar to the above process of face recognition, the first processing unit 231 performs voiceprint recognition. Feature extraction is also required before, and then the vocal to be recognized according to a specific algorithm is recognized. Since the voiceprint recognition technology is known to those skilled in the art, it will not be described again, such as obtained by voiceprint recognition technology. The vocal match is 40%. It is easy to know that the accuracy of determining the identity information by using the voiceprint feature is low. Therefore, when performing the data processing, the second processing unit 232 should determine the weight distribution of the importance of determining the identity information according to the face feature and the voiceprint feature. For example, if the weight of the face matching value is 0.7, the weight of the vocal matching value is 0.3, and then the weighting process is performed, and the final matching value is 54%.

For another example, in another embodiment, the preset target feature information is a license plate number, and correspondingly, the feature information is a license plate number. Usually, the license plate number consists of the motor vehicle registration agency code and number plate number, for example, Guangdong B·123XXX, so the license plate number can be equivalent to a series, and the equivalent series consists of Chinese characters, English letters and numbers. . In this embodiment, the obtained license plate number is defined as a reference sequence, and the license plate number to be identified is defined as a sequence to be identified, and the comparison between the sequence to be identified and the reference sequence has various methods, for example, the data processing module 230 The number of columns to be identified is compared horizontally with the reference number column, that is, the first item of the series is compared in the order of the columns, until the last item of the series, and the number of items to be identified is the same as the number of items in the reference sequence. For example, the reference number is Guangdong B·123456, and the number to be identified is Guangdong B·167356. According to the above horizontal comparison method, the number of identical items in the sequence to be identified and the reference sequence is 5, so the feature information and the preset target are obtained. The matching value obtained after the feature information is compared is 62.5%. Of course, the data processing module 230 can also discretely compare the sequence to be identified with the reference sequence, that is, the order of the columns is not considered in the comparison process, and only the number of the same content is counted.

In an emergency, it is often impossible to accurately remember the license plate number of the escaped vehicle, that is, the preset The target feature information is not necessarily completely accurate. In order to improve the accuracy of the preset target feature information, preferably, the preset target feature information may include a plurality of information, in addition to the license plate number, the preset target feature information. It also includes other vehicle characteristic information, such as the color of the vehicle. Accordingly, the vehicle color information is also extracted from the driving information by the information extraction module 220. Generally, the data processing module 230 compares the colors of the vehicle. The matching values obtained are generally 0 and 1.

It should be noted that the specific numerical values in this embodiment are for illustrative purposes only, and are not intended to limit the related technical features in the driving record processing system 200 of the embodiment of the present invention.

The location information obtaining module 240 is configured to acquire current location information when the matching value is greater than or equal to a preset matching threshold. In this embodiment, the location information acquiring module 240 may be a GPS module, and the matching value is greater than or equal to a preset matching threshold. Therefore, the feature information extracted from the driving information matches the preset target feature information, that is, the target person is found. Or the target vehicle. At this time, the location information obtaining module 240 determines the current location information of the target person or the target vehicle, thereby facilitating the search target person or the target vehicle. In this embodiment, the current location information may be any applicable by using a vehicle GPS, a communication base station, or the like. Positioning device acquisition. It should be understood that the driving information acquisition module 210 (such as the camera 50 and the microphone 60) is installed on the user's vehicle. Generally, the target person and the target vehicle and the user vehicle are all in the same small range, so the current position information of the user's vehicle is taken. The current location information of the target person and the target vehicle. In a preferred embodiment, the current location information may be composed of latitude and longitude coordinates, or may be formed by a specific screen area displayed on the electronic map. Of course, the current location information may also be composed of any other applicable location information. This will not be repeated.

The communication module 250 is configured to upload current location information and driving information to the server. Uploading current location information and driving information to a server through a network interface 40 (such as Wi-Fi, 3G network, 4G network, etc.), for example, the server is a monitoring center of the urban security system, by receiving current location information sent from the user's vehicle and Driving information to track suspects and lost vehicles. When the driving information is video data, one frame of the feature information (such as the face of the target person) may be uploaded, or a video in a specific time interval may be uploaded, but the video should contain the feature information; When the information is a combination of video data and audio data, in addition to uploading a recording in a specific time interval, uploading one of the specific time intervals of one frame of the image with the characteristic information Segment video. It can be understood that the recording is uploaded to the server only when the feature information relates to the person. The uploaded driving information is used to identify the suspect and the lost vehicle on the one hand, and to assist in determining the geographical location of the suspect and the lost vehicle on the other hand.

Referring to FIG. 9, a driving record processing system according to a second embodiment of the present invention is proposed based on the first embodiment. In this embodiment, an update module 260 is added to the first embodiment, and the update module 260 is configured to receive a server push. The target feature information is stored and stored; or, for sending a request instruction to the server, so that the server responds to the request instruction and pushes the target feature information; and receives the target feature information pushed by the server and stores it.

The server stores the latest target feature information. In this embodiment, the server pushes the target feature information to the driving recorder by means of broadcasting, and the target feature information (that is, the preset target feature information) stored on the driving recorder. When an update is needed, the update module 260 receives the target feature information pushed by the server and stores it so that the target feature information on the drive recorder is in the latest state.

It is worth mentioning that in order to simplify the process of data processing, the server packs and pushes multiple target feature information, and the driving recorder adopts a passive receiving mode, and does not need to target the target feature information contained in the target feature information packet pushed by the server and the preset. The target feature information is compared to determine whether an update is needed. When the driving recorder receives the target feature information packet and stores it, all the previous target feature information is deleted, thereby saving storage space of the storage device.

In addition, the driving recorder actively sends a request instruction to the server according to actual needs, so that the server pushes the target feature information to the user vehicle.

Referring to FIG. 10, based on the first embodiment, a driving record processing system according to a third embodiment of the present invention is proposed. In this embodiment, an interactive interface generating module 270 is added to the first embodiment, and the matching value is greater than or equal to the preset. The matching interface generating module 270 is configured to generate and display an interactive interface for the user to perform an alarm based on the interactive interface, wherein the interactive interface can be displayed through the display screen 30, and the display screen 30 can be a touch screen or a non-touch screen. . Specifically, in actual application, for example, typing a piece of text on the interactive interface, and sending the piece of text to the alarm center (such as 110 service station) through the wireless network; for example, by triggering an alarm button displayed on the interactive interface (such as The touch screen display) dials the alarm and submits relevant information to the alarm center in the form of voice. As a result, the discovered target person and the target vehicle can be reported in time.

According to the technical solution of the embodiment of the present invention, the use function of the driving record processing system in the related art is enriched, and the resource utilization rate is improved; at the same time, the monitoring coverage of the public security system can be expanded to make up for the number of fixed cameras distributed in the current public security system. The shortcomings have greatly reduced the cost of building a public security system.

The above are only the preferred embodiments of the present invention, and are not intended to limit the scope of the invention, and the equivalent structure or equivalent process transformations made by the description of the present invention and the drawings are directly or indirectly applied to other related technical fields. The same is included in the scope of patent protection of the present invention.

Industrial applicability

The embodiment of the invention enriches the use function of the driving record processing system in the related art, and improves the resource utilization rate; at the same time, the technical coverage of the public security system can be expanded by the technical solution of the invention to make up for the fixed camera distribution of the current public security system. The lack of quantity greatly reduces the cost of building a public security system. Therefore, the present invention has strong industrial applicability.

Claims

A driving record processing method includes the following steps:

Get driving information in real time;

Extracting, according to a preset category of the target feature information, feature information that matches the category of the preset target feature information in the driving information;

Comparing the feature information with preset target feature information to obtain a matching value;

When the matching value is greater than or equal to the preset matching threshold, the current location information is acquired, and the current location information and the driving information are uploaded to the server.
The driving record processing method according to claim 1, wherein before the step of acquiring the driving information in real time, the method further comprises:

Receiving target feature information pushed by the server and storing it; or,

Sending a request instruction to the server, so that the server responds to the request instruction and pushes the target feature information; receives the target feature information pushed by the server and stores it.
A driving record processing method according to claim 1, wherein:

The feature information includes face feature information and/or vocal feature information; or

The feature information includes vehicle license plate feature information.
The driving record processing method according to claim 3, wherein when the feature information includes face feature information and vocal feature information, the comparing the feature information with preset target feature information The steps to match the values include:

Comparing the face feature information and the vocal feature information with preset target feature information, respectively, to obtain a face matching value and a vocal match value;

And weighting the face matching value and the vocal matching value to obtain the matching value.
The driving record processing method according to any one of claims 1 to 4, further comprising:

When the matching value is greater than or equal to a preset matching threshold, an interaction interface is generated and displayed for the user to perform an alarm based on the interaction interface.
A driving record processing system, comprising a driving information acquiring module, an information extracting module, a data processing module, a position information acquiring module and a communication module, wherein:

The driving information obtaining module is configured to: obtain driving information in real time;

The information extraction module is configured to: extract, according to a category of the preset target feature information, feature information that matches the category of the preset target feature information in the driving information;

The data processing module is configured to: compare the feature information with preset target feature information to obtain a matching value;

The location information obtaining module is configured to: obtain current location information when the matching value is greater than or equal to a preset matching threshold;

The communication module is configured to: upload the current location information and the driving information to a server.
The driving record processing system of claim 6 wherein said driving record processing system further comprises an update module, wherein:

The update module is configured to:

Receiving target feature information pushed by the server and storing it; or,

Sending a request instruction to the server, causing the server to respond to the request instruction and pushing the target feature information, and receiving the target feature information pushed by the server and storing.
A driving record processing system according to claim 6, wherein:

The feature information includes face feature information and/or vocal feature information; or

The feature information includes vehicle license plate feature information.
The driving record processing system according to claim 8, wherein said data processing module comprises a first processing unit and a second processing unit, wherein:

The first processing unit is configured to: compare the facial feature information and the vocal feature information with preset target feature information, and obtain a face matching value and a vocal matching value;

The second processing unit is configured to perform weighting processing on the face matching value and the vocal matching value to obtain the matching value.
The driving record processing system according to any one of claims 6 to 9, further comprising an interactive interface generating module, wherein:

The interaction interface generation module is configured to: when the matching value is greater than or equal to a preset matching threshold, generate and display an interaction interface for the user to perform an alarm based on the interaction interface.
A computer program comprising program instructions which, when executed by a computer, cause the computer to perform the driving record processing method of any one of claims 1-5.
A carrier carrying the computer program of claim 11.