CN111353369A - Application method and system of high-order video of urban roadside parking in assisting criminal investigation - Google Patents

Application method and system of high-order video of urban roadside parking in assisting criminal investigation Download PDF

Info

Publication number
CN111353369A
CN111353369A CN201910984244.0A CN201910984244A CN111353369A CN 111353369 A CN111353369 A CN 111353369A CN 201910984244 A CN201910984244 A CN 201910984244A CN 111353369 A CN111353369 A CN 111353369A
Authority
CN
China
Prior art keywords
vehicle
information
driver
license plate
berth
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910984244.0A
Other languages
Chinese (zh)
Other versions
CN111353369B (en
Inventor
闫军
杨怀恒
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Intelligent Interconnection Technologies Co ltd
Original Assignee
Intelligent Interconnection Technologies Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Intelligent Interconnection Technologies Co ltd filed Critical Intelligent Interconnection Technologies Co ltd
Priority to CN201910984244.0A priority Critical patent/CN111353369B/en
Publication of CN111353369A publication Critical patent/CN111353369A/en
Application granted granted Critical
Publication of CN111353369B publication Critical patent/CN111353369B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/53—Querying
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161—Detection; Localisation; Normalisation
    • G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/60—Type of objects
    • G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/625—License plates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/08—Detecting or categorising vehicles

Abstract

The invention discloses an application system and a method of an urban roadside parking high-level video system in auxiliary criminal investigation, which particularly comprise the steps of obtaining action information of a vehicle driving in/out of a roadside parking space, collecting a high-level video image of the parking space according to the action information, receiving the high-level video image, obtaining a characteristic picture of the vehicle driving in/out of the parking space and a characteristic picture of a vehicle driver in a getting-off/getting-on process, identifying the characteristic picture of the vehicle driving in/out of the parking space and the characteristic picture of the vehicle driver in the getting-off/getting-on process, and confirming people-vehicle related event information; and processing the human-vehicle related event information in real time, and integrating a human-vehicle linkage database according to a real-time processing result. By adopting the system and the method for applying the roadside parking high-order video to assisting criminal investigation, all license plate information, vehicle characteristic information and driver information of driving-in/driving-out roadside parking can be stored in a standard and efficient manner so as to be used for assisting investigation work.

Description

Application method and system of high-order video of urban roadside parking in assisting criminal investigation
Technical Field
The invention relates to the field of urban roadside parking management and computer-assisted criminal investigation application, in particular to an application method of an urban roadside parking high-level video in assisted criminal investigation.
Background
Roadside parking management is parking management using the fields on both sides of a road on which traffic is made on the ground. With the rapid development of urban economy and the continuous improvement of the living standard of people, the quantity of urban motor vehicles is kept and rapidly increased, and the importance of the management problem of static traffic in cities is increasingly highlighted. The roadside parking management system based on the high-level video is adopted by more and more urban management departments due to the advantages of high parking identification accuracy rate, difficult damage after installation, comprehensive and clear captured video, no need of manual operation on site, standard and convenient charging and the like. At present, after the system is on line, practice proves that the system has outstanding effects in the aspects of charge regulation, congestion relief, parking space tension solving and the like.
In fact, in some areas, the investigation work always has the factors that the long-term security in China is influenced by too single auxiliary means, low criminal investigation work efficiency and the like. In view of this, there is a need to develop a new criminal investigation auxiliary tool, especially a technical scheme with less investment, fast implementation and good effect, and with the help of all other technical devices and technical methods in social life, which are beneficial to investigation, efforts are made to improve the criminal investigation work efficiency and the investigation success rate, and the good pursuit of the peaceful and happy industry of the masses in China is continuously satisfied.
Disclosure of Invention
Aiming at the problems that the existing criminal investigation auxiliary means are too few, in order to enrich the criminal investigation auxiliary technology, the intelligent device can be an important auxiliary means for criminal investigation work by utilizing the existing intelligent devices such as the high-level videos of the roadside parking and the like without investing too much capital and manpower, and has important significance for the work of evidence obtaining, investigation, searching and the like.
In order to achieve the purpose, the invention provides an application method of an urban roadside parking high-order video in assisting criminal investigation, which comprises the following steps:
acquiring action information of a vehicle driving-in/out roadside berth, acquiring a high-order video image of the berth according to the action information,
receiving the high-order video image, acquiring the characteristic picture of the vehicle entering/exiting the berth and the characteristic picture of the vehicle driver in the process of getting off/on,
associating the characteristic information of the vehicles entering/exiting the berth with the characteristic information of the driver in the process of getting off/on the vehicle, and confirming the information of the human-vehicle associated events;
processing the human-vehicle related event information in real time, integrating a human-vehicle linkage database according to the real-time processing result,
and searching the integrated information of the man-vehicle linkage database according to the instruction, and outputting a search result.
As a further improvement of the present invention, after the obtaining of the characteristic information of the parked vehicle, the method further includes obtaining the time and location information of the vehicle being parked according to the characteristic information of the parked vehicle.
As a further improvement of the present invention, after the obtaining of the time and place of the vehicle entering, the method further includes obtaining characteristic information of a process of getting off a vehicle of a driver driving the vehicle within a set time threshold, and associating the vehicle characteristic information, the time and place of the vehicle entering, and the characteristic information of the driver.
The method is further improved by acquiring the action information of the berth at the driving road side of the vehicle and acquiring the high-order video image of the berth according to the action information;
and detecting whether the vehicle is pressed into the parking space or not, and if the vehicle is driven into the parking space, acquiring the action information of the vehicle driving into the roadside parking space.
As a further improvement of the invention, the method comprises the steps of acquiring action information of a berth at the driving road side of a vehicle, and acquiring a high-order video image of the berth according to the action information;
acquiring the special display image in the process of driving the vehicle into the parking space every several seconds, identifying the special display image in the process of driving the vehicle into the parking space, and confirming the vehicle entrance time;
and after the vehicle berth is confirmed, continuously acquiring a plurality of frames of image information to obtain a high-order video image of the vehicle in the berth.
The method is further improved by acquiring the action information of the vehicle driving out of the roadside berth and acquiring the high-order video image of the berth according to the action information;
continuously collecting the state information of a suspected driver passing by the vehicle, when the collected suspected driver is in contact with the vehicle and within a set time threshold, pressing the vehicle line out of the parking space,
confirming that the suspected driver is the driver driving the vehicle, determining the time and place information of the vehicle leaving the parking space, and associating the time and place information of the vehicle leaving the parking space with the driver.
As a further improvement of the invention, the information of the human-vehicle related events is processed in real time, and the human-vehicle linkage database is integrated according to the real-time processing result,
acquiring video information of multiple entering/exiting berths of the same vehicle, acquiring entering/exiting berth pictures of the vehicle and pictures containing biological characteristic information of a driver in the process of getting off/on the vehicle,
training a plurality of entrance/exit berth pictures of the same vehicle and pictures containing biological characteristic information of a driver in the process of getting off/on the vehicle to obtain a corresponding vehicle information model and a driver biological characteristic model;
and establishing a man-vehicle linkage database according to the vehicle information model and the driver biological characteristic model.
As a further improvement of the present invention,
the steps of obtaining the corresponding vehicle information model and the driver biometric model are,
marking a plurality of driving-in/driving-out parking position pictures and pictures containing biological characteristic information of a driver in a getting-off/getting-on process, and adding license plate numbers, vehicle brand information, driver faces and driver gait feature marks;
a plurality of pictures with marks are deeply trained,
establishing a vehicle information model according to the vehicle brand information;
and establishing a biological characteristic model of the driver according to the facial characteristics and the gait characteristics of the driver.
As a further improvement of the invention, if the same license plate number has a plurality of vehicle information models and/or a plurality of driver biological characteristic models, the license plate number is taken as a main key, and the plurality of vehicle information models and/or the plurality of driver biological characteristic models are respectively added to the man-vehicle linkage database.
As a further improvement of the invention, a plurality of pictures containing the images of the entrance/exit berth and the biological characteristic information of the driver in the getting-off process are trained by a deep learning method based on the convolutional neural network, and a corresponding vehicle information model and a corresponding biological characteristic model of the driver are obtained.
As a further improvement of the invention, after the human-vehicle linkage database is established according to the vehicle information model and the driver biological characteristic model, the method also comprises the following steps: and according to the license plate number, associating the corresponding vehicle information model with the corresponding biological characteristic model of the driver, and establishing a vehicle record.
As a further improvement of the invention, according to the license plate number, the corresponding vehicle information model and the corresponding driver biological characteristic model are associated, and a vehicle record is established:
and if one license plate number corresponds to a plurality of pieces of vehicle brand information, vehicle type information and corresponding driver facial feature and gait feature information, respectively taking the license plate number as a main key, associating a plurality of vehicle information models and biological feature models to form a plurality of vehicle records.
As a further improvement of the invention, the high-order video image is received and processed in real time, the human-vehicle linkage database is integrated according to the real-time processing result,
acquiring high-level video information of the entrance/exit berth of the vehicle in real time,
marking the entering/exiting berth image, obtaining the corresponding license plate number of the vehicle and the biological characteristic information of the driver,
and respectively comparing the license plate number, the biological characteristic information of the driver and the data in the man-vehicle linkage database to obtain comparison results.
As a further improvement of the invention, if only the license plate number of the vehicle entering/exiting the berth exists in the people-vehicle linkage database,
and supplementing the biological characteristic information of the driver corresponding to the vehicle into a biological characteristic model of the driver corresponding to the vehicle entering/exiting the berth in the man-vehicle linkage database.
As a further improvement of the invention, if the license plate number of the vehicle entering/exiting the berth and the biological characteristic information of the driver in the process of getting off the vehicle do not exist in the people-vehicle linkage database, the license plate number is taken as a main key, and a record containing the vehicle information model in the vehicle record and the biological characteristic model of the driver is newly added in the people-vehicle linkage database.
As a further improvement of the invention, the specific steps of searching the integrated human-vehicle linkage database information according to the instruction and outputting the search result comprise,
a retrieval instruction is input according to the retrieval information,
and confirming whether the retrieval information is matched with one or more items in the man-vehicle linkage database information one by one according to the retrieval instruction, and outputting a retrieval result.
As a further improvement of the present invention,
confirming whether the retrieval information is matched with one or more items in the people and vehicle linkage database information one by one according to the retrieval instruction, and outputting the retrieval result comprises the following steps:
inputting face information and/or gait information, searching whether the information exists in the man-vehicle linkage database, and if so, outputting license plate numbers and vehicle information corresponding to the face information and/or the gait information.
As a further improvement of the present invention, the step of confirming one by one whether the search information matches one or more items in the information of the human-vehicle linkage database according to the search instruction, and the step of outputting the search result further comprises:
inputting license plate number and/or vehicle information, searching whether the information exists in the people-vehicle linkage number library, and if so, outputting face information corresponding to the license plate number and the vehicle information.
An application system of roadside parking high-order video in assisting criminal investigation comprises an image acquisition device, a processor and a retrieval terminal, wherein the image acquisition device, the processor and the retrieval terminal are in communication connection;
the image acquisition equipment is used for acquiring the action information of the berth at the driving/driving-in/driving-out side of the vehicle and acquiring a high-order video image of the berth according to the action information,
the processor is used for receiving the high-order video image, acquiring the characteristic picture of the vehicle entering/exiting the berth and the characteristic picture of the vehicle driver in the process of getting off/on,
the processor is also used for correlating the characteristic information of the vehicles entering/exiting the berth and the characteristic information of the driver in the process of getting off/on the vehicle, confirming the information of the human-vehicle correlation events,
the processor is also used for processing the human-vehicle related event information, integrating the human-vehicle linkage database according to the real-time processing result,
and the retrieval terminal is used for retrieving the integrated human-vehicle linkage database information according to the instruction and outputting a retrieval result.
As a further improvement of the present invention, the processor is further configured to obtain the time and location information of the vehicle being parked according to the characteristic information of the parked vehicle.
As a further improvement of the present invention, the processor is further configured to acquire characteristic information of an alighting process of a driver driving the vehicle within a set time threshold, and associate the vehicle characteristic information, the time when the vehicle is parked, and the location information with the characteristic information of the driver.
As a further improvement of the present invention, the vehicle parking system further includes a front-end device, and the front-end device is further configured to detect whether the vehicle enters the parking space, and if the vehicle enters the parking space, obtain motion information of the vehicle entering/exiting roadside parking space.
As a further improvement of the present invention, the image capturing apparatus includes a capturing module and a vehicle identification module;
the acquisition module is used for acquiring the specially displayed pictures every few seconds in the vehicle entrance process;
the vehicle identification module is used for identifying a special display picture in the process of driving a vehicle into a parking space and confirming the vehicle entrance time;
the acquisition module is further used for continuously acquiring a plurality of frames of image information after the vehicle berth is confirmed, and acquiring a high-order video image of the vehicle in the berth.
As a further improvement of the invention, the acquisition module is also used for continuously acquiring the status information of suspected drivers passing by the vehicle,
the processor as a further improvement of the invention comprises an identification module, an association module, a labeling module, a training module and an integration module;
when the collected suspected driver is in contact with the vehicle and the vehicle line is pushed out of the parking space within a set time threshold, confirming that the suspected driver is the driver driving the vehicle,
the vehicle identification module is used for determining the time and place information of the vehicle exiting from the berth and identifying the vehicle information;
the correlation module is used for correlating the time and place information of the vehicle driving out/into the parking space with the biological characteristic information of the driver in the process of getting on/off the vehicle to obtain the human-vehicle correlation event information;
the labeling module is used for labeling a plurality of driving-in/driving-out parking position pictures and pictures containing biological characteristic information of a driver in a driving-in/driving-out process, and adding license plate numbers, vehicle brand information, driver faces and driver gait characteristic marks;
the training module is used for deeply training a plurality of pictures with marks;
the integration module is used for establishing a vehicle information model according to the vehicle brand information and establishing a biological characteristic model of the driver according to the facial characteristics and the gait characteristics of the driver.
As a further improvement of the present invention, if there are multiple vehicle information models and/or multiple driver biometric models for the same license plate number, the integration module is further configured to add the multiple vehicle information models and/or the multiple driver biometric models to the people-vehicle linkage database, respectively, with the license plate number as the primary key.
As a further improvement of the present invention, the training module is further configured to train multiple images containing the images of the entering/exiting berth and the biological feature information of the driver in the getting-off process of the same vehicle through a deep learning method based on a convolutional neural network, and obtain a corresponding vehicle information model and a corresponding biological feature model of the driver.
As a further improvement of the present invention, the integration module is further configured to associate, according to the license plate number, vehicle brand information and vehicle type information corresponding to the license plate number, and corresponding driver facial feature and gait feature information, and establish a vehicle record.
As a further improvement of the present invention, if one license plate number corresponds to a plurality of pieces of brand information, model information, and corresponding facial features and gait feature information of the driver, the integration module is further configured to associate a plurality of vehicle information models and biometric feature models with the license plate number as a main key, respectively, to form a plurality of vehicle records.
As a further improvement of the invention, the processor further comprises a comparison module, and the comparison module is used for respectively comparing the acquired license plate number and the biological characteristic information of the driver with data in the people-vehicle linkage database to acquire a comparison result.
As a further improvement of the present invention, if only the license plate number of the vehicle entering/exiting the parking lot exists in the people-vehicle linkage database, the integration module is further configured to add the biological characteristic information of the driver corresponding to the vehicle into the biological characteristic model of the driver corresponding to the vehicle entering/exiting the parking lot in the people-vehicle linkage database.
As a further improvement of the invention, if the number plate of the vehicle entering/exiting the parking space and the biological characteristic information of the driver in the process of getting off the vehicle do not exist in the people-vehicle linkage database, the integration module is also used for taking the number plate as a main key, and a record containing the vehicle information model in the vehicle record and the biological characteristic model of the driver is newly added in the people-vehicle linkage database.
As a further improvement of the present invention, the retrieval terminal comprises an input module and a retrieval module; the input module is used for inputting a retrieval instruction according to the retrieval information;
the retrieval module is used for confirming whether the retrieval information is matched with one or more items in the man-vehicle linkage database information one by one according to the retrieval instruction and outputting the retrieval result.
As a further improvement of the present invention, the retrieval terminal further comprises an output module;
when the face information and/or the gait information are input, the information is searched to exist in the man-vehicle linkage database, and the output module is further used for outputting the license plate number and the vehicle information corresponding to the face information and/or the gait information.
As a further improvement of the invention, when the license plate number and/or the vehicle information is input, the information is searched to exist in the people-vehicle linkage database, and the output module is also used for outputting the face information corresponding to the license plate number and the vehicle information.
As a further improvement of the invention, the system also comprises a human-vehicle linkage database which is used for storing a plurality of vehicle records.
By adopting the system and the method for applying the roadside parking high-level video to the assistance of criminal investigation, all license plate information, vehicle characteristic information and driver information of driving-in/driving-out roadside parking can be stored in a normative and efficient manner based on the high-level video of the urban roadside parking management system and the human-vehicle linkage database created by training, so as to be used for the assistance of investigation work. When the system is used, certain input conditions can be accurately and efficiently searched according to data in the database, so that identity information and characteristic information of suspicious personnel and suspicious vehicles can be quickly searched, and great help is provided for improving the case handling efficiency.
Drawings
FIG. 1 is a schematic flow chart of an application method of roadside parking high-level video in assisting criminal investigation according to the present invention;
FIG. 2 is a schematic flow chart of an application method of roadside parking high-level video in assisting criminal investigation according to the present invention;
FIG. 3 is a schematic diagram of an application system of roadside parking high-level video in assisting criminal investigation according to the present invention;
FIG. 4 is a schematic diagram of a roadside parking high-level video establishing a pedestrian-vehicle linkage database in an auxiliary criminal investigation application method according to the present invention;
fig. 5 is an application scene diagram of the application method of the roadside parking high-order video in assisting criminal investigation according to the invention.
Detailed Description
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Aiming at the problem that the existing criminal investigation auxiliary means are few, in order to enrich the criminal investigation auxiliary technology, the invention improves and utilizes the existing intelligent equipment such as the roadside parking high-level video and the like, and can enable the equipment to become the important auxiliary means of criminal investigation work without investing too much capital and manpower, and can be used for obtaining evidence, investigation, search and the like.
S1, acquiring action information of a vehicle driving in/out of a roadside berth, and acquiring a high-order video image of the berth according to the action information, wherein the specific berth action generally comprises the moment that a vehicle is pressed to enter the berth or the moment that the vehicle has a tendency to enter the berth, and at the moment, the vehicle can be judged to possibly enter the berth; the event which is about to occur at the berth can be recorded by the high-order video system according to the action information.
The method comprises the steps of obtaining action information of a berth at the side where a vehicle enters and acquiring a high-order video image of the berth according to the action information of entering and exiting the berth;
and detecting whether the vehicle is pressed into the parking space or not, and if the vehicle is driven into the parking space, acquiring the action information of the vehicle driving into the roadside parking space. The specific detection mode is as follows: the vehicle parking space video acquisition system can be determined by geomagnetic induction equipment, infrared induction equipment or acquisition of a high-level video about to enter a parking space, when a vehicle is about to drive into the parking space, the geomagnetic induction equipment or the infrared induction equipment arranged at an entrance of a parking lot or buried underground at the entrance of the parking lot senses that the vehicle passes by, signals can be transmitted to the high-level video equipment, the high-level video can be acquired by the high-level equipment, and when the vehicle is about to enter the parking space, the vehicle runs at a low speed, so that a clear vehicle video can be acquired conveniently;
specifically, when a vehicle runs to a monitoring area (generally, an area near an entrance or an exit of a parking lot or a parking space), a video of a vehicle running process can be collected at intervals, wherein the video comprises a video of the vehicle approaching the parking space, and when no vehicle enters a roadside parking lot, it is required to ensure that at least one frame of image is collected every 5 seconds, so that the total amount of garbage data is kept; when the situation that the vehicle drives out of the parking space is monitored, collecting the vehicle until the vehicle drives out from a previous time period according to the driving-out occurrence time; when the situation that a vehicle drives into the parking space is automatically monitored, collecting the vehicle for a certain time period from the time of driving; when the vehicle is monitored to enter or exit, the acquisition frequency is increased to at least every 6 frames/second, so that the images of the vehicle entering or exiting the parking space and the images containing the characteristics of the face, the gait and the like of a vehicle driver are guaranteed to be captured, and later-stage evidence collection is facilitated.
S2, receiving the high-order video image, and acquiring the characteristic picture of the vehicle driving in/out of the parking space and the characteristic picture of the vehicle driver in the process of getting off/on;
the method comprises the steps that a vehicle is always in driving before entering a parking space, when the vehicle drives into the parking space and stops, characteristic information of the vehicle is obtained and is not limited to vehicle type information, license plate numbers, vehicle colors and the like, and when the vehicle stops within a certain time threshold (generally 0-4 min), a driver can get off the vehicle, and at the moment, collected videos at least comprise vehicle characteristic information (not limited to vehicle type information, license plate numbers, vehicle colors and the like), biological characteristic information of the driver and characteristic information (including a driver getting-off action picture) related to the vehicle driven by the driver;
the method comprises the steps of continuously collecting state information of a suspected driver passing by a vehicle before the vehicle leaves a parking space, confirming the suspected driver as the driver driving the vehicle when the collected suspected driver is in contact with the vehicle and the suspected driver has a boarding action and presses a line to leave the parking space within a set time threshold, and determining time and place information when the vehicle leaves the parking space, wherein the collected information comprises biological characteristic information of the driver, characteristic information of the vehicle and characteristic information (including a boarding action picture of the driver) related to the vehicle driven by the driver.
S3, identifying the characteristic information of the vehicle entering/exiting the berth and the characteristic information of the driver in the process of getting off/on the vehicle, and confirming the information of the human-vehicle related events; the people-vehicle related event information is event information obtained by summarizing the collected characteristic information of the vehicle, the biological characteristic information of a driver driving the vehicle and the characteristic information related to the people and the vehicle in the processes of getting on and off of the driver, so that people and the vehicle are related with high precision, and generally, the summarized event information can be formed by taking the license plate number of the vehicle as a main key so as to ensure the accuracy of people-vehicle related events.
S4, processing the human-vehicle related event information in real time, and integrating the human-vehicle linkage database according to the real-time processing result;
specifically, before processing the high-order video image collected in real time, a human-vehicle linkage database needs to be established, as shown in fig. 3, which is a schematic diagram of establishing a human-vehicle database,
specifically, a man-vehicle linkage database is established by collecting video images of a vehicle before and after entering and exiting a berth for multiple times; and obtaining a man-vehicle linkage database according to the obtained information;
but in order to ensure the reliability of data, the video information of the same vehicle entering/exiting the berth for multiple times can be collected in different sections,
acquiring video information of multiple entering/exiting berths of the same vehicle, acquiring entering/exiting berth pictures of the vehicle and pictures containing biological characteristic information of a driver in the process of getting off/on the vehicle,
training a plurality of entrance/exit berth pictures of the same vehicle and pictures containing biological characteristic information of a driver in the process of getting off/on the vehicle to obtain a corresponding vehicle information model and a driver biological characteristic model;
and establishing a man-vehicle linkage database according to the vehicle information model and the driver biological characteristic model.
Specifically, the steps of obtaining the corresponding vehicle information model and the driver biometric model are as follows:
marking a plurality of driving-in/driving-out parking position pictures and pictures containing biological characteristic information of a driver in a getting-off/getting-on process, and adding license plate numbers, vehicle brand information, driver faces and driver gait feature marks; obtaining a plurality of pictures with various marks, and deeply training the plurality of pictures with the marks;
establishing a vehicle information model according to the trained vehicle brand information, vehicle colors, license plate numbers and the like, wherein the specific vehicle information model is not limited to the license plate number model, the vehicle brand model and the like;
and establishing a biological feature model of the driver according to the facial features and the gait features of the driver, wherein the biological feature model is not limited to comprise a facial feature model and a gait feature model and is mainly used for later forensics as the key features of the driver.
Specifically, in implementation, the facial features and the gait features of the driver are very important features in the human-vehicle related events, generally, the human-vehicle related events can be confirmed after the vehicle information is highly related to the facial features and the gait features of the driver, but when the facial features of the driver cannot be clearly acquired due to special weather reasons or poor light at night, the gait features of the driver are very important features for perfecting the human-vehicle related events.
In the steps, a plurality of pictures containing the entering/exiting berth pictures and the biological characteristic information of the driver in the getting-off process are trained by a deep learning method based on a convolutional neural network, and a corresponding vehicle information model and a corresponding biological characteristic model of the driver are obtained.
After the people and vehicle linkage database is established, the information of the people and vehicle linkage database needs to be updated irregularly, as further improvement of the information of the people and vehicle linkage database, the information of the people and vehicle linkage database needs to be carded and updated according to the license plate number,
the method comprises the following specific steps: and according to the license plate number, associating a vehicle brand model and a vehicle type information model corresponding to the license plate number, and establishing a vehicle record according to the corresponding driver face characteristic model and gait characteristic model.
And if one license plate number corresponds to a plurality of vehicle brand information models, vehicle type information models and corresponding driver facial feature models and gait feature models, associating the plurality of vehicle information models and the biological feature models by taking the license plate number as a main key respectively to form a plurality of vehicle records.
When a plurality of vehicle brand models exist in one license plate number, the information needs to be marked, and a fake-licensed vehicle owner can be struck;
when a license plate number has a plurality of biological feature models, each biological feature can be respectively marked, for example, a driver 1, a driver 2 and the like are respectively marked, when a vehicle is stolen, a suspect can be locked, for example, when the vehicle is found to be lost, the last driver can be locked, a suspect target is determined, and case tracking is facilitated.
Further, in order to ensure that the human-vehicle linkage database is up-to-date, the database needs to be improved at irregular time, and the information of the database needs to be updated as long as a vehicle stops; specifically, receiving the high-order video image and performing real-time processing, and integrating the human-vehicle linkage database according to a real-time processing result specifically further comprises:
acquiring image information of the vehicle entering/exiting berth in real time,
marking the entering/exiting berth image, obtaining the corresponding license plate number of the vehicle and the biological characteristic information of the driver,
and respectively comparing the license plate number, the biological characteristic information of the driver and the data in the man-vehicle linkage database to obtain comparison results.
According to the comparison result, it is determined whether the man-vehicle linkage database needs to be updated,
if only the license plate number of the vehicle entering/exiting the berth exists in the people-vehicle linkage database,
and acquiring a plurality of pictures containing the biological characteristic information of the driver corresponding to the vehicle, training the pictures to obtain a biological characteristic model of the driver, and supplementing the biological characteristic model into a human-vehicle linkage database.
If the number plate of the vehicle entering/leaving the berth and the biological characteristic information of the driver in the process of getting off do not exist in the people-vehicle linkage database,
then obtaining a plurality of high-order video pictures of the vehicle entering/exiting the berth and training to obtain a vehicle information model and a driver biological characteristic model, and taking the license plate number as a main key to newly add a record containing the vehicle information model and the driver biological characteristic model in the vehicle record in the people-vehicle linkage database.
If the data obtained in real time are stored in the human-vehicle linkage database, the record can be ignored and is not updated.
The biological characteristics of the driver comprise gait characteristics of the driver and facial characteristics of the driver, the vehicle information model comprises vehicle type information, license plate numbers, vehicle brands and the like, the accuracy is the highest in order to guarantee that information of the people-vehicle linkage database is required to be updated for multiple times, the high conformity of vehicle owners and vehicle information is guaranteed, and criminal investigation is facilitated.
Another practical way is that if a driver is confirmed to have two vehicle information models in criminal investigation, it is also possible that the vehicle owner changes the vehicle, and if the driver is confirmed to change the vehicle, the original record under the license plate number in the people-vehicle linkage database is deleted, and the changed vehicle record is newly added.
S5, retrieving the integrated information of the man-vehicle linkage database according to the instruction, and outputting a retrieval result;
the method comprises the specific steps of,
a retrieval instruction is input according to the retrieval information,
and confirming whether the retrieval information is matched with one or more items in the man-vehicle linkage database information one by one according to the retrieval instruction, and outputting a retrieval result.
When the police acquires the facial features of the suspect in a detection case, matching corresponding vehicle information according to the facial features, if the vehicle exists, calling the vehicle information, confirming the place where the vehicle frequently moves, and confirming the track of the suspect according to the movement track of the vehicle; assisting the police to do a case.
When the facial features of the suspect are not definitely acquired or the facial features are blocked to cause that the clear facial features cannot be acquired, matching corresponding vehicle information through the gait features, if the vehicle exists, calling the vehicle information, confirming the place where the vehicle frequently moves, and confirming the track of the suspect according to the movement track of the vehicle; meanwhile, fuzzy matching can be carried out on the fuzzy facial features and the facial feature model recorded by the license plate number, and the identity of the suspect can be further confirmed.
If the driver confirmed by the facial features or the gait features has a plurality of vehicles, the vehicle information can be called one by one, the tracks of the vehicles are confirmed respectively, and police are assisted to solve the case.
In the case of vehicle loss, when only the license plate information of a suspect is acquired, corresponding information of a driver recorded by the license plate can be matched, if a plurality of drivers are recorded by the license plate, the driver can be excluded by matching facial features, other drivers driving the vehicle and corresponding driving tracks are further confirmed, and the vehicle is searched by the vehicle owner.
In another criminal investigation case, a suspect can abandon a vehicle for a dive after a crime, so that an police can only acquire vehicle information, when the criminal investigation is assisted, an information record is acquired by inputting a license plate number, vehicle type information is matched according to the license plate number, if only one vehicle and only one facial feature of a driver exist in a vehicle information model matched with the license plate number, a suspect is a vehicle owner with a high probability, a driver driving the vehicle is searched, the vehicle and the movement track of the driver are called, and the case is directly investigated.
If the suspect is not the owner, the license plate number is matched with the vehicle information models in the people-vehicle linkage database, if different vehicle models are recorded by the same license plate number, one vehicle can be determined as a fake-licensed vehicle, and corresponding drivers are matched according to different vehicle information models respectively to detect cases.
If the people-vehicle linkage database is used for assisting criminal investigation, any morphological characteristics of the vehicle comprise license plate number, vehicle type information, vehicle brand and the like; and biological characteristics of a driver, including facial characteristics, gait characteristics and the like, can be used as key characteristics for assisting criminal investigation, and a criminal investigation line cable chain can be completed by associating any two or more characteristics, so that a favorable guarantee is provided for criminal investigation.
As shown in fig. 4, the invention also discloses a system for executing the method, wherein the system comprises a front-end device 8, an image acquisition device 4, a processor 5, a retrieval terminal 7 and a man-vehicle linkage database 6;
the image acquisition equipment 8 acquires action information of a vehicle driving in/out of a roadside berth, and acquires a high-order video image of the berth according to the action information, the image acquisition equipment preferably adopts an array camera, the array camera generally comprises a plurality of cameras with different focal lengths, the cameras are arranged in parallel and respectively shoot images at the far end and the near end, and the image acquisition equipment not only can acquire images of the vehicle driving in and out of the berth, but also can acquire biological characteristics and gait characteristic images of a driver related to the vehicle and can be used as powerful evidence in criminal investigation;
the image acquisition device 8 can also select a gun-ball linkage device, which is a linkage device comprising a gun-shaped camera and a spherical camera, can accurately acquire images of vehicles entering and leaving the berth, but has a little shortage in the acquisition of biological characteristics of drivers, but can make up the shortage through multiple times of algorithm processing, the image acquisition device is mainly used for acquiring videos and images in various formats, after the images are acquired by the image acquisition device, the videos and images are transmitted to the processor through a wired or wireless network, the processor receives the high-level video images, acquires characteristic pictures of vehicles entering/leaving the berth and characteristic pictures of the drivers of the vehicles in the process of getting off/on the berth, and associates the characteristic information of the vehicles entering/leaving the berth and the characteristic information of the drivers in the process of getting off/on the berth, confirming the information of the events related to the people and the vehicles,
also used for processing the human-vehicle related event information, integrating the human-vehicle linkage database according to the real-time processing result,
the specific processing process comprises data marking, deep learning and the like, and specific marking of license plate information, vehicle brand and model, driver face, driver gait and the like. The deep learning training refers to training the video image samples of the vehicles and the drivers by a deep learning method based on a convolutional neural network to obtain a vehicle brand model, a face feature model of the drivers and a gait feature model.
The retrieval terminal is used for retrieving the integrated human-vehicle linkage database information according to the instruction and outputting a retrieval result, and the retrieval terminal is generally arranged at the criminal investigation retrieval terminal, is in communication connection with the human-vehicle linkage database and is used for calling the human-vehicle information in the human-vehicle linkage database according to the retrieval instruction.
The system further comprises front-end equipment, wherein the front-end equipment is further used for detecting whether the vehicle enters the parking space or not, and acquiring action information of the vehicle entering/exiting the roadside parking space if the vehicle enters the parking space, and the specific front-end equipment can be a single gun type video camera, a single ball type video camera, a plurality of ball type video cameras, a plurality of gun type video cameras or a parallel array camera (comprising a plurality of cameras which are arranged in one equipment shell), is mainly used for judging whether the vehicle enters the parking space or not, and can assist in judgment through devices such as a geomagnetic sensor and an infrared sensor.
Specifically, the acquisition equipment comprises an acquisition module; the acquisition module is used for acquiring the specially displayed pictures every few seconds in the vehicle entrance process; the acquisition module is also used for acquiring a biological characteristic information picture of a driver corresponding to the vehicle in the process of getting off/on the vehicle.
In the invention, information such as a driver biological characteristic model and a vehicle information model is used as key information searched in a criminal investigation case.
In the invention, the processor comprises an identification module, an association module, a marking module, a training module and an integration module;
when the collected suspected driver is in contact with the vehicle and the vehicle line is pushed out of the parking space within a set time threshold, confirming that the suspected driver is the driver driving the vehicle,
the vehicle identification module is used for determining the time and place information of the vehicle exiting from the berth and identifying the vehicle information;
the correlation module is used for correlating the time and place information of the vehicle exiting from the berth with the biological characteristic information of the driver in the boarding process to obtain the human-vehicle correlation event information;
the marking module is used for marking a plurality of driving-in/driving-out parking position pictures and pictures containing biological characteristic information of a driver in a getting-off process, and adding license plate numbers, vehicle brand information, driver faces and driver gait feature marks;
the training module is used for training a plurality of sample pictures with marks;
the integration module is used for establishing a vehicle information model according to the vehicle brand information and establishing a biological characteristic model of the driver according to the facial characteristics and the gait characteristics of the driver.
If the same license plate number has a plurality of vehicle information models and/or a plurality of driver biological characteristic models, the integration module is also used for respectively adding the plurality of vehicle information models and/or the plurality of driver biological characteristic models to the man-vehicle linkage database by taking the license plate number as a main key.
The training module is also used for training a plurality of pictures containing the entrance/exit berth pictures and the biological characteristic information of the driver in the getting-off process of the same vehicle through a deep learning method based on the convolutional neural network, and acquiring a corresponding vehicle information model and a corresponding biological characteristic model of the driver.
The integration module is also used for associating vehicle brand information and vehicle type information corresponding to the license plate number and corresponding driver facial feature and gait feature information according to the license plate number, and establishing a vehicle record.
If one license plate number corresponds to a plurality of pieces of vehicle brand information, vehicle type information, and corresponding facial features and gait feature information of a driver, the integration module is further used for respectively associating a plurality of vehicle information models and biological feature models by taking the license plate number as a main key to form a plurality of vehicle records.
Further, the processor also comprises a comparison module, and the comparison module is used for respectively comparing the acquired license plate number and the biological characteristic information of the driver with data in the people-vehicle linkage database to acquire a comparison result.
And if only the license plate number of the vehicle entering/exiting the parking space exists in the people-vehicle linkage database, the integration module is also used for supplementing the biological characteristic information of the driver corresponding to the vehicle into the biological characteristic model of the driver corresponding to the vehicle entering/exiting the parking space in the people-vehicle linkage database.
If the license plate number of the vehicle entering/leaving the parking space and the biological characteristic information of the driver in the process of getting off do not exist in the people-vehicle linkage database, the integration module is also used for taking the license plate number as a main key, and a record containing the vehicle information model in the vehicle record and the biological characteristic model of the driver is newly added in the people-vehicle linkage database.
The retrieval terminal comprises an input module and a retrieval module; the input module is used for inputting a retrieval instruction according to the retrieval information;
the retrieval module is used for confirming whether the retrieval information is matched with one or more items in the man-vehicle linkage database information one by one according to the retrieval instruction and outputting the retrieval result.
The retrieval terminal also comprises an output module;
when the face information and/or the gait information are input, the information is searched to exist in the man-vehicle linkage database, and the output module is further used for outputting the license plate number and the vehicle information corresponding to the face information and/or the gait information.
When the license plate number and/or the vehicle information are/is input, the information is searched for in the people-vehicle linkage database, and the output module is further used for outputting the face information corresponding to the license plate number and the vehicle information.
The high-order video in the invention refers to: the video collected by the image collecting equipment hung at the high position of the monitoring rod is adopted in the existing roadside road occupying parking or closed parking lot parking.
It should be noted that, as can be understood by those skilled in the art, deep learning is one kind of machine learning, and machine learning is a necessary path for implementing artificial intelligence. The concept of deep learning is derived from the research of artificial neural networks, and a multi-layer perceptron comprising a plurality of hidden layers is a deep learning structure. Deep learning forms a more abstract class or feature of high-level representation properties by combining low-level features to discover a distributed feature representation of the data. In the scheme, the specific steps of labeling and training the sample by the deep learning method are not repeated.
The embodiment of the invention provides an application method of a high-order video of roadside parking in an auxiliary criminal investigation, which is only a preferred embodiment of the invention.
It should be understood that the specific order or hierarchy of steps in the processes disclosed is an example of exemplary approaches. Based upon design preferences, it is understood that the specific order or hierarchy of steps in the processes may be rearranged without departing from the scope of the present disclosure. The accompanying method claims present elements of the various steps in a sample order, and are not intended to be limited to the specific order or hierarchy presented.
In the foregoing detailed description, various features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the claimed embodiments of the subject matter require more features than are expressly recited in each claim. Rather, as the following claims reflect, invention lies in less than all features of a single disclosed embodiment. Thus, the following claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. To those skilled in the art; various modifications to these embodiments will be readily apparent, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
What has been described above includes examples of one or more embodiments. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the aforementioned embodiments, but one of ordinary skill in the art may recognize that many further combinations and permutations of various embodiments are possible. Accordingly, the embodiments described herein are intended to embrace all such alterations, modifications and variations that fall within the scope of the appended claims. Furthermore, to the extent that the term "includes" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprising" as "comprising" is interpreted when employed as a transitional word in a claim. Furthermore, any use of the term "or" in the specification of the claims is intended to mean a "non-exclusive or".
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (37)

1. An application method of an urban roadside parking high-order video in assisting criminal investigation is characterized by comprising the following steps:
acquiring action information of a vehicle driving-in/driving-out roadside berth, acquiring a high-order video image of the berth according to the action information,
receiving the high-order video image, acquiring the characteristic picture of the vehicle entering/exiting the berth and the characteristic picture of the vehicle driver in the process of getting off/on,
identifying characteristic pictures of vehicles entering/exiting the berth and characteristic pictures of the vehicle driver in the process of getting off/on the vehicle, and confirming the information of the human-vehicle related events;
processing the human-vehicle related event information in real time, integrating a human-vehicle linkage database according to the real-time processing result,
and searching the integrated information of the man-vehicle linkage database according to the instruction, and outputting a search result.
2. The method for applying high-level video of urban roadside parking according to claim 1, wherein,
acquiring action information of a berth at the side of a vehicle entering the road, and acquiring a high-order video image of the berth according to the action information;
and detecting whether the vehicle is pressed into the parking space or not, and if the vehicle is driven into the parking space, acquiring the action information of the vehicle driving into the roadside parking space.
3. The method for applying high-level video of urban roadside parking according to claim 2, wherein,
and acquiring the time and place information of the vehicle entering the parking space according to the characteristic image of the vehicle entering the parking space after acquiring the characteristic image of the vehicle entering the parking space.
4. The method for applying high-level video of urban roadside parking according to claim 3, wherein,
the method further comprises the steps of obtaining characteristic information of a driver getting-off process of driving the vehicle within a set time threshold after obtaining the time and the place of the vehicle, and associating the vehicle characteristic information, the time and the place of the vehicle with the characteristic information of the driver.
5. The method for applying high-level video of urban roadside parking according to claim 4, wherein,
acquiring action information of a vehicle driving out of a roadside berth, and acquiring a high-order video image of the berth according to the action information;
continuously collecting the state information of a suspected driver passing by the vehicle, when the collected suspected driver is in contact with the vehicle and within a set time threshold, pressing the vehicle line out of the parking space,
confirming that the suspected driver is the driver driving the vehicle, determining the time and place information of the vehicle leaving the parking space, and associating the time and place information of the vehicle leaving the parking space with the characteristic information of the driver.
6. The method for applying high-level video of urban roadside parking according to claim 5, wherein,
the method comprises the steps of obtaining action information of a vehicle driving-in/driving-out roadside berth, and establishing a man-vehicle linkage database after acquiring a high-order video image of the berth according to the action information.
7. The method for applying high-level video of urban roadside parking according to claim 6, wherein,
the steps of establishing the man-vehicle linkage database are as follows: acquiring high-order video information of multiple entering/exiting berths of the same vehicle, acquiring entering/exiting berth pictures of the vehicle and pictures containing biological characteristic information of a driver in the process of getting off/on the vehicle,
training a plurality of entrance/exit berth pictures of the same vehicle and pictures containing biological characteristic information of a driver in the process of getting off/on the vehicle to obtain a corresponding vehicle information model and a driver biological characteristic model;
and establishing a man-vehicle linkage database according to the vehicle information model and the driver biological characteristic model.
8. The method for applying high-level video of urban roadside parking according to claim 7, wherein,
the steps of obtaining the corresponding vehicle information model and the driver biometric model are,
marking a plurality of driving-in/driving-out parking position pictures and pictures containing biological characteristic information of a driver in a getting-off/getting-on process, and adding license plate numbers, vehicle brand information, driver faces and driver gait feature marks;
a plurality of pictures with marks are deeply trained,
establishing a vehicle information model according to the vehicle brand information;
and establishing a biological characteristic model of the driver according to the facial characteristics and the gait characteristics of the driver.
9. The method for applying high-level video of urban roadside parking according to claim 8, wherein,
and if the same license plate number has a plurality of vehicle information models and/or a plurality of driver biological characteristic models, respectively adding the plurality of vehicle information models and/or the plurality of driver biological characteristic models to the man-vehicle linkage database by taking the license plate number as a main key.
10. The method for applying high-level video of urban roadside parking according to claim 8, wherein,
and training a plurality of pictures containing the entrance/exit berth pictures and the biological characteristic information of the driver in the getting-off process of the same vehicle by a deep learning method based on a convolutional neural network to obtain a corresponding vehicle information model and a corresponding biological characteristic model of the driver.
11. The method for applying the high-level video of urban roadside parking according to claim 10, wherein the method comprises the following steps:
after a human-vehicle linkage database is established according to the vehicle information model and the driver biological characteristic model, the method further comprises the following steps: and according to the license plate number, associating the corresponding vehicle information model with the corresponding biological characteristic model of the driver, and establishing a vehicle record.
12. The method for applying high-level video of urban roadside parking according to claim 11, wherein,
and if one license plate number corresponds to a plurality of pieces of vehicle brand information, vehicle type information and corresponding driver facial feature and gait feature information, respectively taking the license plate number as a main key, associating a plurality of vehicle information models and biological feature models to form a plurality of vehicle records.
13. The method for applying high-level video of urban roadside parking according to claim 12, wherein,
processing the information of the human-vehicle related events in real time, and integrating a human-vehicle linkage database according to a real-time processing result;
acquiring high-level video information of the entrance/exit berth of the vehicle in real time,
marking the entering/exiting berth image, obtaining the corresponding license plate number of the vehicle and the biological characteristic information of the driver,
and respectively comparing the license plate number, the biological characteristic information of the driver and the data in the man-vehicle linkage database to obtain comparison results.
14. The method for applying high-level video of urban roadside parking according to claim 13, wherein,
if only the license plate number of the vehicle entering/exiting the berth exists in the people-vehicle linkage database,
and supplementing the biological characteristic information of the driver corresponding to the vehicle into a biological characteristic model of the driver corresponding to the vehicle entering/exiting the berth in the man-vehicle linkage database.
15. The method for applying high-level video of urban roadside parking according to claim 14, wherein,
if the number plate of the vehicle entering/leaving the parking space and the biological characteristic information of the driver in the process of getting off/on the vehicle do not exist in the people-vehicle linkage database, the number plate is used as a main key, and a record containing the vehicle information model in the vehicle record and the biological characteristic model of the driver is newly added in the people-vehicle linkage database.
16. The method for applying the high-level video of urban roadside parking according to any one of claims 1 to 15 in assisting criminal investigation,
the specific steps of searching the integrated information of the man-vehicle linkage database according to the instruction and outputting the search result comprise,
a retrieval instruction is input according to the retrieval information,
and confirming whether the retrieval information is matched with one or more items in the man-vehicle linkage database information one by one according to the retrieval instruction, and outputting a retrieval result.
17. The method for applying high-level video of urban roadside parking according to claim 16, wherein,
confirming whether the retrieval information is matched with one or more items in the people and vehicle linkage database information one by one according to the retrieval instruction, and outputting the retrieval result comprises the following steps:
inputting face information and/or gait information, searching whether the information exists in the man-vehicle linkage database, and if so, outputting license plate numbers and vehicle information corresponding to the face information and/or the gait information.
18. The method for applying high-level videos of urban roadside parking according to claim 17 in assisting criminal investigation,
confirming one by one whether the retrieval information is matched with one or more items in the people and vehicle linkage database information according to the retrieval instruction, and outputting the retrieval result further comprises:
inputting license plate number and/or vehicle information, searching whether the information exists in the people-vehicle linkage number library, and if so, outputting face information corresponding to the license plate number and the vehicle information.
19. The application system of the roadside parking high-order video in the auxiliary criminal investigation is characterized by comprising an image acquisition device, a processor and a retrieval terminal, wherein the image acquisition device, the processor and the retrieval terminal are in communication connection;
the image acquisition equipment is used for acquiring the action information of the berth at the driving/driving-in/driving-out side of the vehicle and acquiring a high-order video image of the berth according to the action information,
the processor is used for receiving the high-order video image, acquiring the characteristic picture of the vehicle entering/exiting the berth and the characteristic picture of the vehicle driver in the process of getting off/on,
the processor is also used for identifying the characteristic picture of the vehicle entering/exiting the berth and the characteristic picture of the vehicle driver in the process of getting off/on, confirming the information of the human-vehicle related events,
the processor is also used for processing the human-vehicle related event information in real time and integrating the human-vehicle linkage database according to a real-time processing result;
and the retrieval terminal is used for retrieving the integrated human-vehicle linkage database information according to the instruction and outputting a retrieval result.
20. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 19,
the vehicle parking system further comprises front-end equipment, wherein the front-end equipment is also used for detecting whether the vehicle enters the parking space or not, and acquiring the action information of the vehicle driving in/out of the roadside parking space if the vehicle enters the parking space.
21. The system for applying roadside parking high-level video to assisting criminal investigation according to claim 20, wherein the processor is further configured to obtain the time and place information of the vehicle entering the position according to the characteristic information of the vehicle entering the position.
22. The system for applying roadside parking high-level video to assisting criminal investigation according to claim 21, wherein the processor is further configured to obtain characteristic information of an alighting process of a driver driving the vehicle within a set time threshold, and associate the vehicle characteristic information, vehicle in-position time and location information with the characteristic information of the driver.
23. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 22,
the image acquisition equipment is also used for acquiring special display pictures of the vehicles entering the field every few seconds; and after the vehicle berth is confirmed, continuously acquiring a plurality of frames of image information to obtain a high-order video image of the vehicle in the berth.
24. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 23,
the acquisition module is also used for continuously acquiring the state information of suspected drivers passing by the vehicle.
25. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 24,
the processor comprises an identification module, an association module, a labeling module, a training module and an integration module;
when the collected suspected driver is in contact with the vehicle and the vehicle line is pushed out of the parking space within a set time threshold, confirming that the suspected driver is the driver driving the vehicle,
the vehicle identification module is used for determining the time and place information of the vehicle exiting from the berth and identifying the vehicle information;
the correlation module is used for correlating the time and place information of the vehicle driving out/into the parking space with the biological characteristic information of the driver in the process of getting on/off the vehicle to obtain the human-vehicle correlation event information;
the labeling module is used for labeling a plurality of driving-in/driving-out parking position pictures and pictures containing biological characteristic information of a driver in a driving-in/driving-out process, and adding license plate numbers, vehicle brand information, driver faces and driver gait characteristic marks;
the training module is used for deeply training a plurality of pictures with marks;
the integration module is used for establishing a vehicle information model according to the vehicle brand information and establishing a biological characteristic model of the driver according to the facial characteristics and the gait characteristics of the driver.
26. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 25,
if the same license plate number has a plurality of vehicle information models and/or a plurality of driver biological characteristic models, the integration module is also used for respectively adding the plurality of vehicle information models and/or the plurality of driver biological characteristic models to the man-vehicle linkage database by taking the license plate number as a main key.
27. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 26,
the training module is also used for training a plurality of pictures containing the entrance/exit berth pictures and the biological characteristic information of the driver in the getting-off process of the same vehicle through a deep learning method based on the convolutional neural network, and acquiring a corresponding vehicle information model and a corresponding biological characteristic model of the driver.
28. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 27,
the integration module is also used for associating vehicle brand information and vehicle type information corresponding to the license plate number and corresponding driver facial feature and gait feature information according to the license plate number, and establishing a vehicle record.
29. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 28,
if one license plate number corresponds to a plurality of pieces of vehicle brand information, vehicle type information, and corresponding facial features and gait feature information of a driver, the integration module is further used for respectively associating a plurality of vehicle information models and biological feature models by taking the license plate number as a main key to form a plurality of vehicle records.
30. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 29,
the processor also comprises a comparison module, and the comparison module is used for respectively comparing the acquired license plate number and the biological characteristic information of the driver with data in the man-car linkage database to acquire a comparison result.
31. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 29,
and if only the license plate number of the vehicle entering/exiting the parking space exists in the people-vehicle linkage database, the integration module is also used for supplementing the biological characteristic information of the driver corresponding to the vehicle into the biological characteristic model of the driver corresponding to the vehicle entering/exiting the parking space in the people-vehicle linkage database.
32. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 31,
if the license plate number of the vehicle entering/leaving the parking space and the biological characteristic information of the driver in the process of getting off do not exist in the people-vehicle linkage database, the integration module is also used for taking the license plate number as a main key, and a record containing the vehicle information model in the vehicle record and the biological characteristic model of the driver is newly added in the people-vehicle linkage database.
33. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 19,
the retrieval terminal comprises an input module and a retrieval module; the input module is used for inputting a retrieval instruction according to the retrieval information;
the retrieval module is used for confirming whether the retrieval information is matched with one or more items in the man-vehicle linkage database information one by one according to the retrieval instruction and outputting the retrieval result.
34. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 33,
the retrieval terminal also comprises an output module;
when the face information and/or the gait information are input, the information is searched to exist in the man-vehicle linkage database, and the output module is further used for outputting the license plate number and the vehicle information corresponding to the face information and/or the gait information.
35. The system for applying roadside parking high-level video in assisting criminal investigation according to claim 34,
when the license plate number and/or the vehicle information are/is input, the information is searched for in the people-vehicle linkage database, and the output module is further used for outputting the face information corresponding to the license plate number and the vehicle information.
36. The system for applying roadside parking high-order video in assisting criminal investigation of any one of claims 19 to 35, wherein the collecting device comprises an array camera or a gun-and-ball linked camera.
37. The system for applying roadside parking high-order video in assisting criminal investigation according to any one of claims 19 to 35, wherein the system further comprises a people-vehicle linkage database for storing a plurality of vehicle records.
CN201910984244.0A 2019-10-16 2019-10-16 Application method and system of urban road side parking high-level video in auxiliary criminal investigation Active CN111353369B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910984244.0A CN111353369B (en) 2019-10-16 2019-10-16 Application method and system of urban road side parking high-level video in auxiliary criminal investigation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910984244.0A CN111353369B (en) 2019-10-16 2019-10-16 Application method and system of urban road side parking high-level video in auxiliary criminal investigation

Publications (2)

Publication Number Publication Date
CN111353369A true CN111353369A (en) 2020-06-30
CN111353369B CN111353369B (en) 2023-11-10

Family

ID=71196949

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910984244.0A Active CN111353369B (en) 2019-10-16 2019-10-16 Application method and system of urban road side parking high-level video in auxiliary criminal investigation

Country Status (1)

Country Link
CN (1) CN111353369B (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111862425A (en) * 2020-07-07 2020-10-30 深圳市凯达尔科技实业有限公司 Parking lot management system and method based on 3D visual dynamic monitoring
CN112766206A (en) * 2021-01-28 2021-05-07 深圳市捷顺科技实业股份有限公司 High-order video vehicle detection method and device, electronic equipment and storage medium
CN113205690A (en) * 2021-04-09 2021-08-03 超级视线科技有限公司 Roadside parking departure event judgment method and system based on multiple dimensions
CN113449624A (en) * 2021-06-22 2021-09-28 超级视线科技有限公司 Method and device for determining vehicle behavior based on pedestrian re-recognition
CN114141053A (en) * 2021-12-07 2022-03-04 无锡加视诚智能科技有限公司 High-position gun and ball linkage device and using method
CN115147911A (en) * 2022-08-22 2022-10-04 山东海博科技信息系统股份有限公司 Smart city file information integration method, system and terminal
WO2022213523A1 (en) * 2021-04-09 2022-10-13 超级视线科技有限公司 Method and system for determining roadside parking entrance event on basis of a plurality of dimensions

Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102426786A (en) * 2011-11-15 2012-04-25 无锡港湾网络科技有限公司 Intelligent video analyzing system and method for automatically identifying fake plate vehicle
CN107133563A (en) * 2017-03-17 2017-09-05 深圳市能信安科技股份有限公司 A kind of video analytic system and method based on police field
CN107256394A (en) * 2017-06-09 2017-10-17 北京深瞐科技有限公司 Driver information and information of vehicles checking method, device and system
CN107633687A (en) * 2017-09-08 2018-01-26 绵阳西真科技有限公司 A kind of vehicle identification method and system for parking lot
CN107945521A (en) * 2017-11-20 2018-04-20 青岛比特信息技术有限公司 A kind of rifle ball linkage trackside is parked detecting system and method
CN108154708A (en) * 2018-01-19 2018-06-12 北京悦畅科技有限公司 Parking lot management method, server, video camera and terminal device
CN108288307A (en) * 2018-01-19 2018-07-17 北京悦畅科技有限公司 A kind of parking lot management method, server, video camera and terminal device
CN108427718A (en) * 2018-02-07 2018-08-21 江苏大学 A kind of vehicle traveling information temporal index and big data analysis method
CN108470381A (en) * 2018-03-08 2018-08-31 沈志强 Roadside Parking management method
CN108765976A (en) * 2018-06-21 2018-11-06 智慧互通科技有限公司 The parallel parking information of trackside manages system and method
CN108805008A (en) * 2018-04-19 2018-11-13 江苏理工学院 A kind of community's vehicle security system based on deep learning
CN109147068A (en) * 2018-09-12 2019-01-04 西安艾润物联网技术服务有限责任公司 Unlicensed vehicles management method, device and computer readable storage medium
CN109446921A (en) * 2018-10-08 2019-03-08 武汉爱迪科技股份有限公司 A kind of police service comprehensive monitoring method based on parking lot monitoring
CN109712255A (en) * 2019-02-27 2019-05-03 北京猎户智芯科技有限公司 A kind of Car park payment evidence-obtaining system and method based on dynamic picture synthesis
CN109815848A (en) * 2018-12-29 2019-05-28 深圳市天彦通信股份有限公司 Parking lot safety pre-warning system and method

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102426786A (en) * 2011-11-15 2012-04-25 无锡港湾网络科技有限公司 Intelligent video analyzing system and method for automatically identifying fake plate vehicle
CN107133563A (en) * 2017-03-17 2017-09-05 深圳市能信安科技股份有限公司 A kind of video analytic system and method based on police field
CN107256394A (en) * 2017-06-09 2017-10-17 北京深瞐科技有限公司 Driver information and information of vehicles checking method, device and system
CN107633687A (en) * 2017-09-08 2018-01-26 绵阳西真科技有限公司 A kind of vehicle identification method and system for parking lot
CN107945521A (en) * 2017-11-20 2018-04-20 青岛比特信息技术有限公司 A kind of rifle ball linkage trackside is parked detecting system and method
CN108288307A (en) * 2018-01-19 2018-07-17 北京悦畅科技有限公司 A kind of parking lot management method, server, video camera and terminal device
CN108154708A (en) * 2018-01-19 2018-06-12 北京悦畅科技有限公司 Parking lot management method, server, video camera and terminal device
CN108427718A (en) * 2018-02-07 2018-08-21 江苏大学 A kind of vehicle traveling information temporal index and big data analysis method
CN108470381A (en) * 2018-03-08 2018-08-31 沈志强 Roadside Parking management method
CN108805008A (en) * 2018-04-19 2018-11-13 江苏理工学院 A kind of community's vehicle security system based on deep learning
CN108765976A (en) * 2018-06-21 2018-11-06 智慧互通科技有限公司 The parallel parking information of trackside manages system and method
CN109147068A (en) * 2018-09-12 2019-01-04 西安艾润物联网技术服务有限责任公司 Unlicensed vehicles management method, device and computer readable storage medium
CN109446921A (en) * 2018-10-08 2019-03-08 武汉爱迪科技股份有限公司 A kind of police service comprehensive monitoring method based on parking lot monitoring
CN109815848A (en) * 2018-12-29 2019-05-28 深圳市天彦通信股份有限公司 Parking lot safety pre-warning system and method
CN109712255A (en) * 2019-02-27 2019-05-03 北京猎户智芯科技有限公司 A kind of Car park payment evidence-obtaining system and method based on dynamic picture synthesis

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111862425A (en) * 2020-07-07 2020-10-30 深圳市凯达尔科技实业有限公司 Parking lot management system and method based on 3D visual dynamic monitoring
CN111862425B (en) * 2020-07-07 2022-05-13 深圳市凯达尔科技实业有限公司 Parking lot management system and method based on 3D visual dynamic monitoring
CN112766206A (en) * 2021-01-28 2021-05-07 深圳市捷顺科技实业股份有限公司 High-order video vehicle detection method and device, electronic equipment and storage medium
CN113205690A (en) * 2021-04-09 2021-08-03 超级视线科技有限公司 Roadside parking departure event judgment method and system based on multiple dimensions
WO2022213523A1 (en) * 2021-04-09 2022-10-13 超级视线科技有限公司 Method and system for determining roadside parking entrance event on basis of a plurality of dimensions
CN113449624A (en) * 2021-06-22 2021-09-28 超级视线科技有限公司 Method and device for determining vehicle behavior based on pedestrian re-recognition
CN114141053A (en) * 2021-12-07 2022-03-04 无锡加视诚智能科技有限公司 High-position gun and ball linkage device and using method
CN115147911A (en) * 2022-08-22 2022-10-04 山东海博科技信息系统股份有限公司 Smart city file information integration method, system and terminal

Also Published As

Publication number Publication date
CN111353369B (en) 2023-11-10

Similar Documents

Publication Publication Date Title
CN111353369A (en) Application method and system of high-order video of urban roadside parking in assisting criminal investigation
CN110008298B (en) Parking multidimensional information perception application system and method
CN101848377B (en) Device and method for intelligent linkage of multi-video recording device based on cloud computing and mass video searching
CN107633687A (en) A kind of vehicle identification method and system for parking lot
CN110738857B (en) Vehicle violation evidence obtaining method, device and equipment
CN108986472B (en) Method and device for monitoring vehicle turning round
CN106448160B (en) Target person tracking method combining vehicle running track and monitoring video data
CN105741560A (en) Vehicle identification method used for gas station
CN107067730B (en) Network appointment vehicle-man-vehicle inconsistency monitoring method based on bayonet equipment
CN105046966A (en) System and method for automatically detecting illegal parking behaviors in drop-off areas
CN104766479A (en) Automobile identity recognition method and device based on ultrahigh frequency radio frequency and video image dual-recognition matching
CN108710827B (en) A kind of micro- police service inspection in community and information automatic analysis system and method
CN110619277A (en) Multi-community intelligent deployment and control method and system
KR102181355B1 (en) Vehicle search system based artificial intelligence
CN111445703A (en) Intelligent parking lot system and method based on image recognition
CN114898297A (en) Non-motor vehicle illegal behavior determination method based on target detection and target tracking
CN204856897U (en) It is detection device violating regulations in abscission zone territory that motor vehicle stops promptly
CN111508090A (en) Recognition and charging system and method for unmanned vehicles to enter and exit parking lot
CN112949439A (en) Method and system for monitoring invasion of personnel in key area of oil tank truck
WO2016107302A1 (en) Method and system for inspecting moving vehicles
CN107180535A (en) The joyride behavior automatic identification equipment and method of a kind of automatic sound detection based on deep learning
CN114140719A (en) AI traffic video analysis technology
CN110837753A (en) Collective and separate model for human-vehicle object identification and control and use method thereof
CN112330960A (en) Multidimensional data acquisition system
CN111461124A (en) Large data-based shielded license plate recognition method and device and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB02 Change of applicant information
CB02 Change of applicant information

Address after: 075000 ten building, phase 1, airport economic and Technological Development Zone, Zhangjiakou, Hebei

Applicant after: Smart intercommunication Technology Co.,Ltd.

Address before: 075000 ten building, phase 1, airport economic and Technological Development Zone, Zhangjiakou, Hebei

Applicant before: INTELLIGENT INTER CONNECTION TECHNOLOGY Co.,Ltd.

GR01 Patent grant
GR01 Patent grant