WO2020042489A1 - 违法停车案件的鉴别方法、装置和计算机设备 - Google Patents

违法停车案件的鉴别方法、装置和计算机设备 Download PDF

Info

Publication number
WO2020042489A1
WO2020042489A1 PCT/CN2018/123548 CN2018123548W WO2020042489A1 WO 2020042489 A1 WO2020042489 A1 WO 2020042489A1 CN 2018123548 W CN2018123548 W CN 2018123548W WO 2020042489 A1 WO2020042489 A1 WO 2020042489A1
Authority
WO
WIPO (PCT)
Prior art keywords
illegal parking
image
parking
training
key elements
Prior art date
Application number
PCT/CN2018/123548
Other languages
English (en)
French (fr)
Inventor
巢中迪
庄伯金
袁宏进
魏鑫
张玉鑫
肖京
Original Assignee
平安科技（深圳）有限公司
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 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2020042489A1 publication Critical patent/WO2020042489A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
    • G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • 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
    • G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
    • 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

Definitions

  • the present application relates to the field of computer technology, and in particular, to a method, a device, and a computer device for identifying an illegal parking case.
  • the embodiments of the present application provide a method, a device, and a computer device for identifying an illegal parking case, so as to intelligently identify the compliance of the illegal parking case, reduce the cost of manual review, and enable the enforcement of traffic law enforcement personnel. Regulatory supervision.
  • an embodiment of the present application provides a method for identifying an illegal parking case, including: acquiring an image of an illegal parking case; detecting key elements of the illegal parking in the image; and detecting the illegal parking key obtained by the detection. Identifying the information in the elements; determining the illegal parking case based on the detected key position of the illegal parking, the positional relationship between the illegal parking key elements, and the information in the identified illegal parking key elements Types of violations and compliance with the handling of said illegal parking cases.
  • an embodiment of the present application provides an apparatus for identifying an illegal parking case, including: an acquisition module for acquiring an image of the illegal parking case; and a detection module for detecting an illegal parking key in an image obtained by the acquisition module. Elements for detection; an identification module for identifying information in the key elements of illegal parking obtained by the detection module; a determination module for determining the key elements of illegal parking and the key for illegal parking obtained according to the detection module The positional relationship between the elements and the information in the key elements of illegal parking obtained by the identification module, determine the type of violation of the illegal parking case and the compliance of the above illegal parking case processing.
  • an embodiment of the present application provides a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor.
  • the processor executes the computer program, Implement the method described above.
  • an embodiment of the present application provides a computer non-volatile readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the method described above is implemented.
  • the key elements of illegal parking in the above images are detected, the information in the key elements of illegal parking obtained by the detection is identified, and then the key elements of illegal parking obtained by detection, The positional relationship between the key elements of illegal parking and the information obtained from identifying the key elements of illegal parking, determine the types of violations of the above illegal parking cases and the compliance of the above illegal parking cases processing, so that intelligent parking of illegal parking cases can be realized Identification of compliance, reduce the cost of manual review, and can supervise the normative nature of law enforcement by traffic law enforcement personnel.
  • FIG. 1 is a flowchart of an embodiment of a method for identifying an illegal parking case in this application
  • FIG. 2 is a flowchart of another embodiment of a method for identifying an illegal parking case in this application
  • FIG. 3 is a flowchart of another embodiment of a method for identifying an illegal parking case in this application.
  • FIG. 5 is a schematic structural diagram of an embodiment of an identification device for an illegal parking case in this application.
  • FIG. 6 is a schematic structural diagram of another embodiment of an identification device for an illegal parking case in this application.
  • FIG. 7 is a schematic structural diagram of an embodiment of a computer device of the present application.
  • FIG. 1 is a flowchart of an embodiment of a method for identifying an illegal parking case. As shown in FIG. 1, the above method for identifying an illegal parking case may include:
  • Step 101 Obtain an image of an illegal parking case.
  • the images of the above illegal parking cases can include: images of illegal parking scenes taken by traffic law enforcement personnel, and captured snapshots. Images of illegal parking scenes and / or frame images of illegal parking scene videos.
  • the video of the illegal parking scene may be a video of a parking scene of a vehicle taken by a driving recorder on a vehicle, or a video of an illegal parking scene of a bystander (a pedestrian or a driver in an adjacent vehicle). It is a video of the illegal parking scene taken by traffic enforcement personnel, which is not limited in this embodiment.
  • Step 102 Detect the key elements of illegal parking in the image.
  • Step 103 Identify the information in the key elements of illegal parking obtained through detection.
  • the above key elements of illegal parking may include one or a combination of the following: a vehicle, a license plate of the above vehicle, a no-parking sign, and a ticket;
  • the information in the key elements of the illegal parking may include: the license plate number of the license plate, the type of the parking prohibition mark, and the penalty information in the penalty ticket.
  • prohibited parking signs may include: prohibited parking lines, prohibited parking signs, prohibited parking sections, and / or sidewalks.
  • Step 104 Determine the type of violation of the illegal parking case and the combination of the processing of the illegal parking case based on the information obtained from the detection of the key illegal parking element, the positional relationship between the above illegal parking key elements, and the identification of the obtained illegal parking key element. Regulation.
  • the key elements of illegal parking in the above image are detected, the information in the key elements of illegal parking obtained by the detection is identified, and then the illegal parking obtained according to the detection
  • the key elements, the positional relationship between the above key elements of illegal parking, and the information in the key elements of illegal parking obtained through identification determine the type of violation of the above illegal parking cases and the compliance of the above illegal parking cases processing, so that intelligent Identify the compliance of illegal parking cases, reduce the cost of manual review, and supervise the law enforcement of traffic law enforcement personnel.
  • FIG. 2 is a flowchart of another embodiment of a method for identifying an illegal parking case in this application.
  • step 102 may be:
  • Step 201 Perform a normalization process on the size and color distribution of the image.
  • normalizing the size of the image is to process the size of the image according to a predetermined size to make the sizes of the images consistent; to normalize the color distribution of the image is to make the image
  • the colors are in the same distribution to avoid uneven contrast and / or histogram distribution of the above images.
  • Step 202 Use a pre-trained deep neural network model to perform image recognition on the normalized image to obtain a region where the key elements in the normalized image are located and a category of the key elements. Categories include one or a combination of the following: vehicles, license plates, traffic signs, and tickets for those vehicles.
  • the method may further include:
  • Step 203 Use the training image and the annotation file corresponding to the training image to train the training model to obtain a trained deep neural network model.
  • using the training image and the annotation file corresponding to the training image to train the training model may be: inputting the training image and the annotation file corresponding to the training image into the training model, and using a deep neural network algorithm to perform the training model on the training model. Training; wherein the label file corresponding to the training image includes the region where the key elements in the training image are located and the category of the key elements; when the error between the output of the training model and the label file corresponding to the training image is less than a predetermined threshold , A trained deep neural network model is obtained.
  • the size of the predetermined threshold may be set according to system performance and / or implementation requirements during specific implementation. This embodiment does not limit the size of the predetermined threshold.
  • the above-mentioned deep neural network algorithm can use a multi-object detection and recognition algorithm based on deep learning, such as fast region detection based on convolutional neural network (Fast Regions with Convolutional Neural Network; hereinafter referred to as Fast-R-CNN), single-lens multi-box Algorithms such as detection (Single Shot MultiBox Detector; hereinafter referred to as SSD) or only once (You Only Look Once: hereinafter referred to as YOLO) and other algorithms are not limited in this embodiment.
  • fast-R-CNN convolutional neural network
  • SSD Single Shot MultiBox Detector
  • YOLO Only Look Once
  • FIG. 3 is a flowchart of another embodiment of a method for identifying an illegal parking case in this application.
  • the information in the above-mentioned key elements includes one or a combination of the following: The license plate number of the vehicle, the instructions of the above traffic sign, and the penalty information in the above ticket.
  • step 103 may include:
  • Step 301 Recognize the license plate number of the license plate by using a license plate recognition technology, and identify the type of the forbidden parking mark and the penalty information in the penalty ticket through optical character recognition (Optical Character Recognition; OCR).
  • OCR Optical Character Recognition
  • VLPR Vehicle License Recognition Technology
  • vehicle License, Recognition is an application of computer video image recognition technology in vehicle license plate recognition.
  • the license plate recognition technology can extract and recognize the license plate of a vehicle from a complex background.
  • the technology of license plate extraction, image pre-processing, feature extraction, license plate character recognition and other technologies can be used to identify vehicle license plate number and color.
  • the region where the license plate is obtained can be separated from the above image; then, the region where the license plate is located is divided into single characters, and the vertical segmentation method is generally used for character segmentation. Because the projection of characters in the vertical direction is necessarily near the local minimum between characters or in the gap between characters, and this position should meet the character writing format, characters, size restrictions, and some other conditions of the license plate, so the vertical projection method is used. Character segmentation works well.
  • segmented characters can be identified based on template matching algorithms and artificial neural network algorithms.
  • the template-based matching algorithm first binarizes the segmented characters and scales them to the size of the template in the character database, then matches all the templates, and selects the best match as the result.
  • the following introduces the process of identifying the type of the above-mentioned no-parking mark and the penalty information in the above-mentioned ticket through OCR.
  • the following description takes the type of the above-mentioned no-parking mark as an example, and the above-mentioned no-parking sign is identified by OCR
  • the identification of the type of object may be: identification of a no-stop line, a sidewalk, and / or a no-stop section through OCR.
  • the pre-processing mainly includes binarization, noise removal, tilt correction, and the like.
  • binarization is to divide the content of the above image of the no-parking sign into foreground information and background information.
  • the foreground information can be simply defined as black and the background information is white. This is the binary map;
  • Noise removal is based on the characteristics of noise to denoise the image of the above-mentioned no-parking sign
  • the tilt correction is to correct the direction of the image of the parking prohibition sign to avoid the image of the parking prohibition sign from tilting.
  • a layout analysis is performed on the image of the parking prohibition sign, that is, a process of segmenting and / or branching the image of the parking prohibition sign.
  • FIG. 4 is a flowchart of another embodiment of a method for identifying an illegal parking case in this application.
  • step 104 may include:
  • step 401 the integrity of the vehicle obtained by the detection is detected.
  • the vehicle in the key elements of illegal parking obtained by detection it can be checked whether the relevant image of the vehicle includes both the front and the rear. If so, the vehicle can be determined to be complete; and if the relevant image of the vehicle is included If there is only the front or rear of the vehicle, it can be determined that the above vehicles are incomplete.
  • Step 402 After determining that the vehicle is complete, determine a positional relationship between the no-parking mark obtained through detection and the vehicle.
  • Step 403 Determine the violation type of the illegal parking case according to the position relationship between the parking prohibition mark and the vehicle, and identify the type of the obtained parking prohibition mark.
  • prohibited parking signs may include: prohibited parking lines, prohibited parking signs, prohibited parking sections, and / or sidewalks.
  • Step 404 Compare the violation type of the illegal parking case with the punishment information in the identified ticket. According to whether the violation type of the illegal parking case matches the penalty information in the ticket, determine the compliance of the illegal parking case. Sex.
  • the types of violations in the illegal parking case match the penalty information in the ticket with reference to relevant provisions in the Traffic Law. If it matches, it is determined that the handling of the above illegal parking cases is in compliance with the regulations; if it does not match, it is determined that the handling of the above illegal parking cases is not in compliance with the regulations.
  • the key elements of illegal parking obtained are vehicles, the license plates of the above-mentioned vehicles, parking prohibition signs and tickets, and the information in the key elements of illegal parking obtained by detection is further identified.
  • the license plate number obtained by the above license plate is "Jing N *****"
  • the type of the above-mentioned no-parking sign is a no-parking sign
  • the penalty information in the above-mentioned ticket is a fine of 200 yuan.
  • the integrity of the vehicle obtained by the test can be checked to see whether the relevant image of the above vehicle includes both the head and the tail, and if so, it can be determined that the above vehicle is complete;
  • the positional relationship between the no-parking sign obtained from the detection and the above-mentioned vehicle for example: the distance between the above-mentioned vehicle and the no-parking mark is less than 1 meter;
  • the violation type of the illegal parking case is a license plate number of “ ⁇ N ** *** "vehicles park illegally in prohibited parking areas;
  • FIG. 5 is a schematic structural diagram of an embodiment of an identification device for an illegal parking case in this application.
  • the identification device for an illegal parking case in this embodiment can implement the method for identifying an illegal parking case provided in an embodiment of this application.
  • the above discrimination device for illegal parking cases may include: an acquisition module 51, a detection module 52, an identification module 53, and a determination module 54;
  • the obtaining module 51 is configured to obtain an image of an illegal parking case.
  • an illegal parking case there may be multiple sources of the image of the illegal parking case.
  • the image of the above illegal parking case may be Including: images of illegal parking scenes taken by traffic law enforcement personnel, images of illegal parking scenes captured by control and / or frame images of illegal parking scene videos.
  • the video of the illegal parking scene may be a video of a parking scene of a vehicle taken by a driving recorder on a vehicle, or a video of an illegal parking scene of a bystander (a pedestrian or a driver in an adjacent vehicle). It is a video of the illegal parking scene taken by traffic enforcement personnel, which is not limited in this embodiment.
  • a detection module 52 configured to detect key illegal parking elements in an image acquired by the acquisition module 51;
  • the identification module 53 is configured to identify the information in the key elements of illegal parking obtained by the detection module 52.
  • the key elements of the illegal parking may include one or a combination of the following: a vehicle, a license plate of the above vehicle, a prohibited parking sign, and Ticket
  • the information in the key elements of the illegal parking may include: the license plate number of the license plate, the type of the parking prohibition mark, and the penalty information in the penalty ticket.
  • prohibited parking signs may include: prohibited parking lines, prohibited parking signs, prohibited parking sections, and / or sidewalks.
  • a determining module 54 is configured to determine the violations of the illegal parking case based on the information obtained from the detection of the key illegal parking key elements, the positional relationship between the aforementioned illegal parking key elements, and the identification module 53 identifying the obtained illegal parking key elements. Types and compliance with the aforementioned illegal parking cases.
  • the detection module 52 detects the key illegal parking elements in the image, and the identification module 53 performs information on the key illegal parking elements obtained by the detection. Identify and then determine the module 54 to determine the type of violation of the illegal parking case and the illegal parking case based on the information on the critical parking illegal elements obtained from the detection, the positional relationship between the above illegal parking critical elements, and the identified critical parking illegal elements.
  • the compliance of processing can intelligently identify the compliance of illegal parking cases, reduce the cost of manual review, and supervise the standardization of law enforcement by traffic law enforcement personnel.
  • FIG. 6 is a schematic structural diagram of another embodiment of an identification device for an illegal parking case in this application.
  • the detection module 52 is specifically configured to normalize the size and color distribution of the above-mentioned image; using a pre-trained depth
  • the neural network model performs image recognition on the normalized image, and obtains the region where the key elements in the normalized image are located and the category of the key element.
  • the category of the key element includes one or a combination of the following: Vehicles, license plates, traffic signs, and tickets for those vehicles.
  • the detection module 52 performs normalization processing on the size of the image, that is, processes the size of the image according to a predetermined size to make the sizes of the images consistent; the detection module 52 performs normalization processing on the color distribution of the image Is to make the colors of the above images be in the same distribution, and to avoid the unevenness of the contrast and / or histogram distribution of the above images.
  • the above apparatus for identifying a traffic violation case may further include: a training module 55;
  • a training module 55 is configured to use a training image and a label file corresponding to the training image to train a training model to obtain a trained deep neural network model.
  • the training module 55 is specifically configured to input the training image and a label file corresponding to the training image into the training model, and use a deep neural network algorithm to train the training model; the label file corresponding to the training image includes the foregoing The region where the key elements are in the training image and the category of the above key elements; when the error between the output of the training model and the annotation file corresponding to the training image is less than a predetermined threshold, a trained deep neural network model is obtained.
  • the size of the predetermined threshold may be set according to system performance and / or implementation requirements during specific implementation. This embodiment does not limit the size of the predetermined threshold.
  • the above-mentioned deep neural network algorithm may adopt a multi-object detection and recognition algorithm based on deep learning, such as Fast R-CNN, SSD, or YOLO, etc., which is not limited in this embodiment.
  • the identification module 53 is specifically configured to identify the license plate number of the license plate by using a license plate recognition technology, and identify the indication of the traffic sign and the penalty information in the penalty ticket through OCR.
  • the information in the key elements includes one or a combination of the following: : The license plate number of the aforementioned license plate, the instructions of the aforementioned traffic sign, and the penalty information in the aforementioned ticket.
  • VLPR Vehicle License Recognition Technology
  • vehicle License, Recognition is an application of computer video image recognition technology in vehicle license plate recognition.
  • the license plate recognition technology can extract and recognize the license plate of a vehicle from a complex background.
  • the technology of license plate extraction, image pre-processing, feature extraction, license plate character recognition and other technologies can be used to identify vehicle license plate number and color.
  • the recognition module 53 may separate the area where the license plate is obtained from the above-mentioned image; then, the recognition module 53 divides the area where the license plate is located into a single character, and the vertical segmentation method is generally used for character segmentation. Because the projection of characters in the vertical direction is necessarily near the local minimum between characters or in the gap between characters, and this position should meet the character writing format, characters, size restrictions, and some other conditions of the license plate, so the vertical projection method is used. Character segmentation works well.
  • the recognition module 53 can recognize the segmented characters based on a template matching algorithm and an artificial neural network algorithm.
  • the template-based matching algorithm first binarizes the segmented characters and scales them to the size of the template in the character database, then matches all the templates, and selects the best match as the result.
  • the following describes the process of identifying the types of the above-mentioned prohibited parking signs and the penalty information in the above-mentioned ticket through OCR.
  • the following description takes the identification of the types of the above-mentioned prohibited parking signs as an example.
  • the identification module 53 uses OCR
  • the identification of the type of the above-mentioned no-parking sign may be: the recognition module 53 identifies the no-parking line, the sidewalk, and / or the no-parking section through the OCR.
  • the recognition module 53 pre-processes the image of the above-mentioned no-stop sign, and the pre-processing mainly includes binarization, noise removal, tilt correction, and the like.
  • binarization is to divide the content of the image of the above-mentioned no-parking sign into foreground information and background information.
  • the foreground information can be simply defined as black and the background information is white, which is a binary map;
  • Noise removal is based on the characteristics of noise to denoise the image of the above-mentioned no-parking sign
  • the tilt correction is to correct the direction of the image of the parking prohibition sign to avoid the image of the parking prohibition sign from tilting.
  • the identification module 53 performs layout analysis on the image of the parking prohibition sign, that is, a process of dividing the image of the parking prohibition sign into paragraphs and / or branches.
  • the recognition module 53 performs character cutting on the image of the above-mentioned parking stop sign, and then recognizes the cut characters. Finally, the recognition module 53 performs layout restoration on the recognized text, and recognizes the recognition based on the relationship of the specific language context. The results are corrected.
  • the determination module 54 may include: an integrity detection submodule 541, a position relationship determination submodule 542, a violation type determination submodule 543, and a compliance determination submodule 544;
  • the integrity detection submodule 541 is configured to detect the integrity of the vehicle obtained by the detection module 52. Specifically, for the vehicles in the key illegal parking elements obtained by the detection module 52, the integrity detection submodule 541 may Check whether the relevant image of the vehicle includes both the front and the rear. If so, it can be determined that the vehicle is complete; and if the relevant image of the vehicle includes only the front or the rear, it can be determined that the vehicle is incomplete.
  • a position relationship determining submodule configured to determine the position relationship between the forbidden parking mark obtained by the detection module 52 and the vehicle after the integrity detection submodule 541 determines that the vehicle is complete;
  • the violation type determination sub-module 543 is configured to determine the violation type of the illegal parking case according to the position relationship between the parking prohibition mark and the vehicle, and the type of the parking prohibition mark identified by the identification module 53;
  • the types of signs may include: no parking lines, no parking signs, no parking sections, and / or sidewalks.
  • the compliance determination sub-module 544 is used to compare the violation type of the illegal parking case determined by the violation type determination sub-module 543 with the penalty information in the ticket obtained by the identification module 53. According to the foregoing violation type of the illegal parking case, Whether the penalty information in the ticket matches, to determine the compliance of the above illegal parking cases.
  • the compliance determination sub-module 544 compares the types of violations in the illegal parking case with the punishment information in the ticket obtained through detection, and then needs to determine the types of violations in the illegal parking case with the above in combination with relevant provisions in the traffic law. Whether the penalty information in the ticket matches, if it matches, it is determined that the handling of the above illegal parking cases meets the regulations; if it does not match, it is determined that the handling of the above illegal parking cases does not meet the regulations.
  • the detection module 52 detects images of illegal parking cases, and the key elements of illegal parking obtained are vehicles, license plates of the above vehicles, parking prohibition signs, and tickets.
  • the identification module 53 further detects the key elements of illegal parking obtained by detection.
  • the information in the information is identified, the license plate number of the license plate obtained is "Jing N *****", the type of the parking prohibition sign is a parking prohibition sign, and the penalty information in the ticket is a fine of 200 yuan.
  • the integrity detection sub-module 541 can detect the integrity of the vehicle obtained by the inspection, and check whether the relevant image of the vehicle includes both the front and the rear of the vehicle. If so, it can determine that the vehicle is complete;
  • the position relationship determination sub-module 542 determines the position relationship between the detected no-parking mark and the vehicle, for example, the distance between the vehicle and the no-parking mark is less than 1 meter ;
  • the violation type determination submodule 543 may determine that the violation type of the illegal parking case is a license plate number. Vehicles with "Beijing N *****" parking illegally in prohibited parking areas;
  • the compliance determination sub-module 544 compares the types of violations in the above illegal parking cases with the penalty information in the identified ticket.
  • FIG. 7 is a schematic structural diagram of an embodiment of a computer device of the present application.
  • the computer device may include a memory, a processor, and a computer program stored on the memory and executable on the processor.
  • the processor executes the computer program, The method for identifying illegal parking cases provided in the embodiments of the present application can be implemented.
  • the computer device may be a server, such as a cloud server, or an electronic device, such as a smart electronic device such as a smart phone, a smart watch, or a tablet computer. This embodiment does not limit the specific form of the computer device.
  • FIG. 7 shows a block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present application.
  • the computer device 12 shown in FIG. 7 is only an example, and should not impose any limitation on the functions and scope of use of the embodiments of the present application.
  • the computer device 12 is represented in the form of a general-purpose computing device.
  • the components of the computer device 12 may include, but are not limited to, one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).
  • the bus 18 represents one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local area bus using any of a variety of bus structures.
  • these architectures include, but are not limited to, Industry Standard Architecture (hereinafter referred to as ISA) bus, Micro Channel Architecture (hereinafter referred to as MAC) bus, enhanced ISA bus, video electronics Standards Association (Video) Standards Association (hereinafter referred to as: VESA) local bus and Peripheral Component Interconnection (hereinafter referred to as: PCI) bus.
  • ISA Industry Standard Architecture
  • MAC Micro Channel Architecture
  • VESA video electronics Standards Association
  • PCI Peripheral Component Interconnection
  • Computer device 12 typically includes a variety of computer system-readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.
  • the system memory 28 may include a computer system readable medium in the form of volatile memory, such as Random Access Memory (hereinafter referred to as RAM) 30 and / or cache memory 32.
  • Computer device 12 may further include other removable / non-removable, volatile / nonvolatile computer system storage media.
  • the storage system 34 may be used to read and write non-removable, non-volatile magnetic media (not shown in FIG. 7 and is commonly referred to as a "hard drive").
  • a disk drive for reading and writing to a removable non-volatile disk such as a "floppy disk” and a removable non-volatile optical disk (such as a compact disk read-only memory (Compact Disc ReadRead Only Memory (hereinafter referred to as: CD-ROM), digital multi-function read-only optical disc (Digital Video Disc Read Read Only Memory; hereinafter referred to as: DVD-ROM) or other optical media) read and write optical disc drive.
  • CD-ROM Compact Disc ReadRead Only Memory
  • DVD-ROM digital multi-function read-only optical disc
  • each drive may be connected to the bus 18 through one or more data medium interfaces.
  • the memory 28 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of the embodiments of the present application.
  • a program / utility tool 40 having a set (at least one) of program modules 42 may be stored in, for example, the memory 28.
  • Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other programs Modules and program data, each or some combination of these examples may include an implementation of a network environment.
  • the program module 42 generally performs functions and / or methods in the embodiments described in this application.
  • the computer device 12 may also communicate with one or more external devices 14 (such as a keyboard, pointing device, display 24, etc.), may also communicate with one or more devices that enable a user to interact with the computer device 12, and / or with Any device (eg, network card, modem, etc.) that enables the computer device 12 to communicate with one or more other computing devices. This communication can be performed through an input / output (I / O) interface 22.
  • the computer device 12 may also be connected to one or more networks (such as a local area network (hereinafter referred to as LAN), a wide area network (hereinafter referred to as WAN), and / or a public network such as the Internet through the network adapter 20 ) Communication. As shown in FIG.
  • the network adapter 20 communicates with other modules of the computer device 12 through the bus 18. It should be understood that although not shown in FIG. 7, other hardware and / or software modules may be used in conjunction with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tapes Drives and data backup storage systems.
  • the processing unit 16 executes various functional applications and data processing by running a program stored in the system memory 28, for example, implementing a method for identifying an illegal parking case provided by the embodiment of the present application.
  • the embodiment of the present application also provides a computer non-volatile readable storage medium having a computer program stored thereon.
  • the computer program is executed by a processor, the method for identifying an illegal parking case provided by the embodiment of the present application can be implemented.
  • the computer non-volatile storage medium may adopt any combination of one or more computer-readable media.
  • the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium.
  • the computer-readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof.
  • a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in combination with an instruction execution system, apparatus, or device.
  • the computer-readable signal medium may include a data signal in baseband or propagated as part of a carrier wave, which carries a computer-readable program code. Such a propagated data signal may take many forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • the computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, and the computer-readable medium may send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device .
  • Program code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
  • Computer program code for performing the operations of this application may be written in one or more programming languages, or a combination thereof, including programming languages such as Java, Smalltalk, C ++, and also conventional Procedural programming language—such as "C" or similar programming language.
  • the program code can be executed entirely on the user's computer, partly on the user's computer, as an independent software package, partly on the user's computer, partly on a remote computer, or entirely on a remote computer or server.
  • the remote computer can be connected to the user's computer through any kind of network, including a local area network (hereinafter referred to as a LAN) or a wide area network (hereinafter referred to as a WAN), or Connect to an external computer (for example, using an Internet service provider to connect over the Internet).
  • a local area network hereinafter referred to as a LAN
  • a wide area network hereinafter referred to as a WAN
  • Connect to an external computer for example, using an Internet service provider to connect over the Internet.
  • first and second are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as “first” and “second” may explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality” is at least two, for example, two, three, etc., unless it is specifically and specifically defined otherwise.
  • Any process or method description in a flowchart or otherwise described herein can be understood as representing a module, fragment, or portion of code that includes one or more executable instructions for implementing steps of a custom logic function or process
  • the scope of the preferred embodiments of this application includes additional implementations in which the functions may be performed out of the order shown or discussed, including performing the functions in a substantially simultaneous manner or in the reverse order according to the functions involved, which should It is understood by those skilled in the art to which the embodiments of the present application pertain.
  • the word “if” as used herein can be interpreted as “at” or “when” or “responding to determination” or “responding to detection”.
  • the phrases “if determined” or “if detected (the stated condition or event)” can be interpreted as “when determined” or “responded to the determination” or “when detected (the stated condition or event) ) “Or” in response to a detection (statement or event stated) ".
  • terminals involved in the embodiments of the present application may include, but are not limited to, personal computers (hereinafter referred to as PCs), personal digital assistants (hereinafter referred to as PDAs), wireless handheld devices, tablets Computer (Tablet Computer), mobile phone, MP3 player, MP4 player, etc.
  • PCs personal computers
  • PDAs personal digital assistants
  • Tablet Computer tablets Computer
  • mobile phone MP3 player
  • MP4 player etc.
  • the disclosed systems, devices, and methods may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each of the units may exist separately physically, or two or more units may be integrated into one unit.
  • the above integrated unit may be implemented in the form of hardware, or in the form of hardware plus software functional units.
  • the above integrated unit implemented in the form of a software functional unit may be stored in a computer-readable storage medium.
  • the above software functional unit is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) or a processor to execute the methods described in the embodiments of the present application. Some steps.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (hereinafter referred to as ROM), random access memory (Random Access Memory (hereinafter referred to as RAM), magnetic disk or optical disk, etc.) A medium on which program code can be stored.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)

Abstract

一种违法停车案件的鉴别方法、装置和计算机设备，违法停车案件的鉴别方法包括：获取违法停车案件的图像（101）；对图像中的违法停车关键要素进行检测（102）；对检测获得的违法停车关键要素中的信息进行识别（103）；根据检测获得的违法停车关键要素、违法停车关键要素之间的位置关系和识别获得的违法停车关键要素中的信息，确定违法停车案件的违规类型和违法停车案件处理的合规性（104）。可以实现智能化地对违法停车案件的合规性进行鉴别，减少人工审核的成本，并可以对交通执法人员执法的规范性进行监督。

Description

违法停车案件的鉴别方法、装置和计算机设备
本申请要求于2018年8月30日提交中国专利局、申请号为201811004600.X、发明名称为“违法停车案件的鉴别方法、装置和计算机设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，尤其涉及一种违法停车案件的鉴别方法、装置和计算机设备。
背景技术
随着我国国民经济的快速发展，车辆数量的激增，导致了交通需求增长过快而引发的诸如交通阻塞等一系列问题，其中车辆违法停车现象是造成交通阻塞的一个重要因素。
对于车辆的违法停车行为，大多由交通执法人员来判断车辆是否违法停车，但是如何对违法停车案件的合规性进行判断，相关技术中并未提供相应的解决方案。
申请内容
本申请实施例提供了一种违法停车案件的鉴别方法、装置和计算机设备，以实现智能化地对违法停车案件的合规性进行鉴别，减少人工审核的成本，并可以对交通执法人员执法的规范性进行监督。
第一方面，本申请实施例提供了一种违法停车案件的鉴别方法，包括：获取违法停车案件的图像；对所述图像中的违法停车关键要素进行检测；对检测获得的所述违法停车关键要素中的信息进行识别；根据检测获得的所述违法停车关键要素、所述违法停车关键要素之间的位置关系和识别获得的所述违法停车关键要素中的信息，确定所述违法停车案件的违规类型和所述违法停车案件处理的合规性。
第二方面，本申请实施例提供一种违法停车案件的鉴别装置，包括：获取模块，用于获取违法停车案件的图像；检测模块，用于对所述获取模块获取的图像中的违法停车关键要素进行检测；识别模块，用于对所述检测模块检测获得的违法停车关键要素中的信息进行识别；确定模块，用于根据所述检测模块检测获得的违法停车关键要素、所述违法停车关键要素之间的位置关系和所述识别模块识别获得的违法停车关键要素中的信息，确定上述违法停车案件的违规类型和上述违法停车案件处理的合规性。
第三方面，本申请实施例提供一种计算机设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时，实现如上所述的方法。
第四方面，本申请实施例提供一种计算机非易失性可读存储介 质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如上所述的方法。
以上技术方案中，获取违法停车案件的图像之后，对上述图像中的违法停车关键要素进行检测，对检测获得的违法停车关键要素中的信息进行识别，然后根据检测获得的违法停车关键要素、上述违法停车关键要素之间的位置关系和识别获得的违法停车关键要素中的信息，确定上述违法停车案件的违规类型和上述违法停车案件处理的合规性，从而可以实现智能化地对违法停车案件的合规性进行鉴别，减少人工审核的成本，并可以对交通执法人员执法的规范性进行监督。
附图说明
为了更清楚地说明本申请实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为本申请违法停车案件的鉴别方法一个实施例的流程图；
图2为本申请违法停车案件的鉴别方法另一个实施例的流程图；
图3为本申请违法停车案件的鉴别方法再一个实施例的流程图；
图4为本申请违法停车案件的鉴别方法再一个实施例的流程图；
图5为本申请违法停车案件的鉴别装置一个实施例的结构示意图；
图6为本申请违法停车案件的鉴别装置另一个实施例的结构示意图；
图7为本申请计算机设备一个实施例的结构示意图。
具体实施方式
为了更好的理解本申请的技术方案，下面结合附图对本申请实施例进行详细描述。
应当明确，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其它实施例，都属于本申请保护的范围。
在本申请实施例中使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本申请。在本申请实施例和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义。
图1为本申请违法停车案件的鉴别方法一个实施例的流程图，如图1所示，上述违法停车案件的鉴别方法可以包括：
步骤101，获取违法停车案件的图像。
本实施例中，对违法停车案件进行鉴别时，违法停车案件的图像来源可以有多种，举例来说，上述违法停车案件的图像可以包括：交通执法人员拍摄的违法停车现场的图像、布控抓拍的违法停车现场的图像和/或违法停车现场视频的帧图像。
其中，上述违法停车现场视频可以是车辆上的行车记录仪拍摄的车辆停车现场的视频，也可以是旁观者(行人或相邻车辆中的驾乘人员)拍摄的违法停车现场的视频，还可以是交通执法人员拍摄的违法停车现场的视频，本实施例对此不作限定。
步骤102，对上述图像中的违法停车关键要素进行检测。
步骤103，对检测获得的违法停车关键要素中的信息进行识别。
其中，上述违法停车关键要素可以包括以下之一或组合：车辆、上述车辆的车牌、禁止停车标志物和罚单；
上述违法停车关键要素中的信息可以包括：上述车牌的车牌号码、上述禁止停车标志物的类型和上述罚单中的处罚信息。
其中，上述禁止停车标志物的类型可以包括：禁止停车线、禁止停车标识、禁停路段和/或人行道等。
步骤104，根据检测获得的违法停车关键要素、上述违法停车关键要素之间的位置关系和识别获得的违法停车关键要素中的信息，确定上述违法停车案件的违规类型和上述违法停车案件处理的合规性。
上述违法停车案件的鉴别方法中，获取违法停车案件的图像之后，对上述图像中的违法停车关键要素进行检测，对检测获得的违法停车关键要素中的信息进行识别，然后根据检测获得的违法停车关键要素、上述违法停车关键要素之间的位置关系和识别获得的违法停车关键要素中的信息，确定上述违法停车案件的违规类型和上述违法停车案件处理的合规性，从而可以实现智能化地对违法停车案件的合规性进行鉴别，减少人工审核的成本，并可以对交通执法人员执法的规范性进行监督。
图2为本申请违法停车案件的鉴别方法另一个实施例的流程图，如图2所示，本申请图1所示实施例中，步骤102可以为：
步骤201，对上述图像的尺寸和颜色分布进行归一化处理。
具体地，对上述图像的尺寸进行归一化处理，就是将上述图像的尺寸按照预定大小进行处理，使上述图像的尺寸一致；对上述图像的颜色分布进行归一化处理，是为了使上述图像的颜色处于同一分布，避免上述图像的对比度和/或直方图分布不均衡。
步骤202，利用预先训练的深度神经网络模型，对归一化处理后的图像进行图像识别，获得上述归一化处理后的图像中关键要素所在的区域和上述关键要素的类别，上述关键要素的类别包括以下之一或组合：车辆、上述车辆的车牌、交通标志和罚单。
进一步地，步骤202之前，还可以包括：
步骤203，利用训练图像和上述训练图像对应的标注文件，对训练模型进行训练，获得训练好的深度神经网络模型。
具体地，利用训练图像和上述训练图像对应的标注文件，对训练模型进行训练可以为：将上述训练图像和上述训练图像对应的标注文件输入上述训练模型，利用深度神经网络算法对上述训练模型进行训练；其中，上述训练图像对应的标注文件包括上述训练图像中关键要素所在的区域和上述关键要素的类别；当上述训练模型输 出的结果与上述训练图像对应的标注文件之间的误差小于预定阈值时，获得训练好的深度神经网络模型。
上述预定阈值的大小可以在具体实现时，根据系统性能和/或实现需求等自行设定，本实施例对上述预定阈值的大小不作限定。
其中，上述深度神经网络算法可以采用基于深度学习的多目标检测识别算法，例如基于卷积神经网络的快速区域检测(Fast Regions with Convolutional Neural Network；以下简称：Fast R-CNN)、单镜头多盒检测(Single Shot MultiBox Detector；以下简称：SSD)或只看一次(You Only Look Once；以下简称：YOLO)等算法，本实施例对此不作限定。
图3为本申请违法停车案件的鉴别方法再一个实施例的流程图，如图3所示，本申请图1所示实施例中，上述关键要素中的信息包括以下之一或组合：上述车牌的车牌号码、上述交通标志的指示和上述罚单中的处罚信息。这样，步骤103可以包括：
步骤301，通过车牌识别技术对上述车牌的车牌号码进行识别，通过光学字符识别(Optical Character Recognition；以下简称：OCR)对上述禁止停车标志物的类型和上述罚单中的处罚信息进行识别。
其中，车牌识别技术(Vehicle License Plate Recognition；以下简称：VLPR)是计算机视频图像识别技术在车辆牌照识别中的一种应用。车牌识别技术可以将车辆的牌照从复杂背景中提取并识别出来，通过车牌提取、图像预处理、特征提取、车牌字符识别等技术，识别车辆牌号、颜色等信息。
具体地，首先，可以对检测获得的车牌所在区域从上述图像中分离出来；然后，再将上述车牌所在区域分割成单个字符，字符分割一般采用垂直投影法。由于字符在垂直方向上的投影必然在字符间或字符内的间隙处取得局部最小值的附近，并且这个位置应满足车牌的字符书写格式、字符、尺寸限制和一些其他条件，因此利用垂直投影法进行字符分割有较好的效果。
最后，可以基于模板匹配算法和基于人工神经网络算法对分割后的字符进行识别。其中，基于模板匹配算法首先将分割后的字符二值化并将其尺寸大小缩放为字符数据库中模板的大小，然后与所有的模板进行匹配，选择最佳匹配作为结果。基于人工神经网络的算法有两种：一种是先对字符进行特征提取，然后用所获得特征来训练神经网络分配器；另一种方法是直接把图像输入网络，由网络自动实现特征提取直至识别出结果。
下面以通过OCR对上述禁止停车标志物的类型和上述罚单中的处罚信息进行识别的过程进行介绍，下面的介绍以对上述禁止停车标志物的类型进行识别为例，通过OCR对上述禁止停车标志物的类型进行识别可以为：通过OCR对禁止停车线、人行道和/或禁停路段进行识别。
具体地，首先，对上述禁止停车标志物的图像进行预处理，预处理主要包括：二值化、噪声去除和倾斜较正等。
其中，二值化是对上述禁止停车标志物的图像的内容进行划分， 分为前景信息与背景信息，可以简单的定义前景信息为黑色，背景信息为白色，这就是二值化图了；
噪声去除是根据噪声的特征对上述禁止停车标志物的图像进行去噪；
倾斜较正是对上述禁止停车标志物的图像的方向进行校正，避免上述禁止停车标志物的图像倾斜。
然后，对上述禁止停车标志物的图像进行版面分析，也就是将上述禁止停车标志物的图像进行分段落和/或分行的过程。
接下来，对上述禁止停车标志物的图像进行字符切割，然后对切割后的字符进行识别，最后对识别获得的文字进行版面恢复，根据特定的语言上下文的关系，对识别结果进行校正。
图4为本申请违法停车案件的鉴别方法再一个实施例的流程图，如图4所示，本申请图1所示实施例中，步骤104可以包括：
步骤401，对检测获得的车辆的完整性进行检测。
具体地，对于检测获得的违法停车关键要素中的车辆，可以检查上述车辆的相关图像中是否既包括车头，又包括车尾，如果是，则可以确定上述车辆完整；而如果上述车辆的相关图像中只有车头或车尾，那就可以确定上述车辆不完整。
步骤402，在确定上述车辆完整之后，确定检测获得的禁止停车标志物与上述车辆的位置关系。
步骤403，根据上述禁止停车标志物与上述车辆的位置关系，以及识别获得的禁止停车标志物的类型，确定上述违法停车案件的违规类型。
其中，上述禁止停车标志物的类型可以包括：禁止停车线、禁止停车标识、禁停路段和/或人行道等。
步骤404，将上述违法停车案件的违规类型和识别获得的罚单中的处罚信息进行对比，根据上述违法停车案件的违规类型与上述罚单中的处罚信息是否匹配，确定上述违法停车案件处理的合规性。
具体地，将上述违法停车案件的违规类型和识别获得的罚单中的处罚信息进行对比之后，需要结合交通法中的相关规定，确定上述违法停车案件的违规类型与上述罚单中的处罚信息是否匹配，如果匹配，则确定上述违法停车案件的处理符合规定；如果不匹配，则确定上述违法停车案件的处理不符合规定。
举例来说，假设对违法停车案件的图像进行检测，获得的违法停车关键要素为车辆、上述车辆的车牌、禁止停车标志物和罚单，进一步对检测获得的违法停车关键要素中的信息进行识别，获得上述车牌的车牌号码为“京N*****”，上述禁止停车标志物的类型为禁止停车标识，上述罚单中的处罚信息为罚款200元。
首先，可以对检测获得的车辆的完整性进行检测，检查上述车辆的相关图像中是否既包括车头，又包括车尾，如果是，则可以确定上述车辆完整；
在确定上述车辆完整之后，确定检测获得的禁止停车标志物与上述车辆的位置关系，例如：上述车辆与禁止停车标志物之间的距 离小于1米；
进一步地，根据上述禁止停车标志物与上述车辆的位置关系，以及识别获得的上述禁止停车标志物的类型为禁止停车标识，可以确定上述违法停车案件的违规类型为车牌号码为“京N*****”的车辆在禁止停车的区域违法停车；
最后，将上述违法停车案件的违规类型和识别获得的罚单中的处罚信息进行对比，这里就是将“车牌号码为‘京N*****’的车辆在禁止停车的区域违法停车”与“罚款200元”进行对比，结合交通法中的规定，确定上述违法停车案件的违规类型与上述罚单中的处罚信息匹配，进而可以确定上述违法停车案件的处理符合规定。
图5为本申请违法停车案件的鉴别装置一个实施例的结构示意图，本实施例中的违法停车案件的鉴别装置可以实现本申请实施例提供的违法停车案件的鉴别方法。如图5所示，上述违法停车案件的鉴别装置可以包括：获取模块51、检测模块52、识别模块53和确定模块54；
其中，获取模块51，用于获取违法停车案件的图像；本实施例中，对违法停车案件进行鉴别时，违法停车案件的图像来源可以有多种，举例来说，上述违法停车案件的图像可以包括：交通执法人员拍摄的违法停车现场的图像、布控抓拍的违法停车现场的图像和/或违法停车现场视频的帧图像。
其中，上述违法停车现场视频可以是车辆上的行车记录仪拍摄的车辆停车现场的视频，也可以是旁观者(行人或相邻车辆中的驾乘人员)拍摄的违法停车现场的视频，还可以是交通执法人员拍摄的违法停车现场的视频，本实施例对此不作限定。
检测模块52，用于对获取模块51获取的图像中的违法停车关键要素进行检测；
识别模块53，用于对检测模块52检测获得的违法停车关键要素中的信息进行识别；其中，上述违法停车关键要素可以包括以下之一或组合：车辆、上述车辆的车牌、禁止停车标志物和罚单；
上述违法停车关键要素中的信息可以包括：上述车牌的车牌号码、上述禁止停车标志物的类型和上述罚单中的处罚信息。
其中，上述禁止停车标志物的类型可以包括：禁止停车线、禁止停车标识、禁停路段和/或人行道等。
确定模块54，用于根据检测模块52检测获得的违法停车关键要素、上述违法停车关键要素之间的位置关系和识别模块53识别获得的违法停车关键要素中的信息，确定上述违法停车案件的违规类型和上述违法停车案件处理的合规性。
上述违法停车案件的鉴别装置中，获取模块51获取违法停车案件的图像之后，检测模块52对上述图像中的违法停车关键要素进行检测，识别模块53对检测获得的违法停车关键要素中的信息进行识别，然后确定模块54根据检测获得的违法停车关键要素、上述违法停车关键要素之间的位置关系和识别获得的违法停车关键要素中的信息，确定上述违法停车案件的违规类型和上述违法停车案件处理 的合规性，从而可以实现智能化地对违法停车案件的合规性进行鉴别，减少人工审核的成本，并可以对交通执法人员执法的规范性进行监督。
图6为本申请违法停车案件的鉴别装置另一个实施例的结构示意图，本实施例中，检测模块52，具体用于对上述图像的尺寸和颜色分布进行归一化处理；利用预先训练的深度神经网络模型，对归一化处理后的图像进行图像识别，获得上述归一化处理后的图像中关键要素所在的区域和上述关键要素的类别，上述关键要素的类别包括以下之一或组合：车辆、上述车辆的车牌、交通标志和罚单。
具体地，检测模块52对上述图像的尺寸进行归一化处理，就是将上述图像的尺寸按照预定大小进行处理，使上述图像的尺寸一致；检测模块52对上述图像的颜色分布进行归一化处理，是为了使上述图像的颜色处于同一分布，避免上述图像的对比度和/或直方图分布不均衡。
进一步地，上述交通违章案件的鉴别装置还可以包括：训练模块55；
训练模块55，用于利用训练图像和上述训练图像对应的标注文件，对训练模型进行训练，获得训练好的深度神经网络模型。
本实施例中，训练模块55，具体用于将上述训练图像和上述训练图像对应的标注文件输入上述训练模型，利用深度神经网络算法对上述训练模型进行训练；上述训练图像对应的标注文件包括上述训练图像中关键要素所在的区域和上述关键要素的类别；当上述训练模型输出的结果与上述训练图像对应的标注文件之间的误差小于预定阈值时，获得训练好的深度神经网络模型。上述预定阈值的大小可以在具体实现时，根据系统性能和/或实现需求等自行设定，本实施例对上述预定阈值的大小不作限定。
其中，上述深度神经网络算法可以采用基于深度学习的多目标检测识别算法，例如Fast R-CNN、SSD或YOLO等方法，本实施例对此不作限定。
识别模块53，具体用于通过车牌识别技术对上述车牌的车牌号码进行识别，通过OCR对上述交通标志的指示和上述罚单中的处罚信息进行识别，上述关键要素中的信息包括以下之一或组合：上述车牌的车牌号码、上述交通标志的指示和上述罚单中的处罚信息。
其中，车牌识别技术(Vehicle License Plate Recognition；以下简称：VLPR)是计算机视频图像识别技术在车辆牌照识别中的一种应用。车牌识别技术可以将车辆的牌照从复杂背景中提取并识别出来，通过车牌提取、图像预处理、特征提取、车牌字符识别等技术，识别车辆牌号、颜色等信息。
具体地，首先，识别模块53可以对检测获得的车牌所在区域从上述图像中分离出来；然后，识别模块53再将上述车牌所在区域分割成单个字符，字符分割一般采用垂直投影法。由于字符在垂直方向上的投影必然在字符间或字符内的间隙处取得局部最小值的附近，并且这个位置应满足车牌的字符书写格式、字符、尺寸限制和一些 其他条件，因此利用垂直投影法进行字符分割有较好的效果。
最后，识别模块53可以基于模板匹配算法和基于人工神经网络算法对分割后的字符进行识别。其中，基于模板匹配算法首先将分割后的字符二值化并将其尺寸大小缩放为字符数据库中模板的大小，然后与所有的模板进行匹配，选择最佳匹配作为结果。基于人工神经网络的算法有两种：一种是先对字符进行特征提取，然后用所获得特征来训练神经网络分配器；另一种方法是直接把图像输入网络，由网络自动实现特征提取直至识别出结果。
下面以通过OCR对上述禁止停车标志物的类型和上述罚单中的处罚信息进行识别的过程进行介绍，下面的介绍以对上述禁止停车标志物的类型进行识别为例，其中，识别模块53通过OCR对上述禁止停车标志物的类型进行识别可以为：识别模块53通过OCR对禁止停车线、人行道和/或禁停路段进行识别。
具体地，首先，识别模块53对上述禁止停车标志物的图像进行预处理，预处理主要包括：二值化、噪声去除和倾斜较正等。
其中，二值化是对上述禁止停车标志物的图像的内容进行划分，分为前景信息与背景信息，可以简单的定义前景信息为黑色，背景信息为白色，这就是二值化图了；
噪声去除是根据噪声的特征对上述禁止停车标志物的图像进行去噪；
倾斜较正是对上述禁止停车标志物的图像的方向进行校正，避免上述禁止停车标志物的图像倾斜。
然后，识别模块53对上述禁止停车标志物的图像进行版面分析，也就是将上述禁止停车标志物的图像进行分段落和/或分行的过程。
接下来，识别模块53对上述禁止停车标志物的图像进行字符切割，然后对切割后的字符进行识别，最后识别模块53对识别获得的文字进行版面恢复，根据特定的语言上下文的关系，对识别结果进行校正。
本实施例中，确定模块54可以包括：完整性检测子模块541、位置关系确定子模块542、违规类型确定子模块543和合规性确定子模块544；
其中，完整性检测子模块541，用于对检测模块52检测获得的车辆的完整性进行检测；具体地，对于检测模块52检测获得的违法停车关键要素中的车辆，完整性检测子模块541可以检查上述车辆的相关图像中是否既包括车头，又包括车尾，如果是，则可以确定上述车辆完整；而如果上述车辆的相关图像中只有车头或车尾，那就可以确定上述车辆不完整。
位置关系确定子模块，用于在完整性检测子模块541确定上述车辆完整之后，确定检测模块52检测获得的禁止停车标志物与上述车辆的位置关系；
违规类型确定子模块543，用于根据上述禁止停车标志物与上述车辆的位置关系，以及识别模块53识别获得的禁止停车标志物的类型，确定上述违法停车案件的违规类型；其中，上述禁止停车标志物的类型可以包括：禁止停车线、禁止停车标识、禁停路段和/或人 行道等。
合规性确定子模块544，用于将违规类型确定子模块543确定的违法停车案件的违规类型和识别模块53识别获得的罚单中的处罚信息进行对比，根据上述违法停车案件的违规类型与上述罚单中的处罚信息是否匹配，确定上述违法停车案件处理的合规性。
具体地，合规性确定子模块544将上述违法停车案件的违规类型和检测获得的罚单中的处罚信息进行对比之后，需要结合交通法中的相关规定，确定上述违法停车案件的违规类型与上述罚单中的处罚信息是否匹配，如果匹配，则确定上述违法停车案件的处理符合规定；如果不匹配，则确定上述违法停车案件的处理不符合规定。
举例来说，假设检测模块52对违法停车案件的图像进行检测，获得的违法停车关键要素为车辆、上述车辆的车牌、禁止停车标志物和罚单，识别模块53进一步对检测获得的违法停车关键要素中的信息进行识别，获得上述车牌的车牌号码为“京N*****”，上述禁止停车标志物的类型为禁止停车标识，上述罚单中的处罚信息为罚款200元。
首先，完整性检测子模块541可以对检测获得的车辆的完整性进行检测，检查上述车辆的相关图像中是否既包括车头，又包括车尾，如果是，则可以确定上述车辆完整；
在完整性检测子模块541确定上述车辆完整之后，位置关系确定子模块542确定检测获得的禁止停车标志物与上述车辆的位置关系，例如：上述车辆与禁止停车标志物之间的距离小于1米；
进一步地，根据上述禁止停车标志物与上述车辆的位置关系，以及识别获得的上述禁止停车标志物的类型为禁止停车标识，违规类型确定子模块543可以确定上述违法停车案件的违规类型为车牌号码为“京N*****”的车辆在禁止停车的区域违法停车；
最后，合规性确定子模块544将上述违法停车案件的违规类型和识别获得的罚单中的处罚信息进行对比，这里就是将“车牌号码为‘京N*****’的车辆在禁止停车的区域违法停车”与“罚款200元”进行对比，结合交通法中的规定，确定上述违法停车案件的违规类型与上述罚单中的处罚信息匹配，进而可以确定上述违法停车案件的处理符合规定。
图7为本申请计算机设备一个实施例的结构示意图，上述计算机设备可以包括存储器、处理器及存储在上述存储器上并可在上述处理器上运行的计算机程序，上述处理器执行上述计算机程序时，可以实现本申请实施例提供的违法停车案件的鉴别方法。
其中，上述计算机设备可以为服务器，例如：云服务器，也可以为电子设备，例如：智能手机、智能手表或平板电脑等智能电子设备，本实施例对上述计算机设备的具体形态不作限定。
图7示出了适于用来实现本申请实施方式的示例性计算机设备12的框图。图7显示的计算机设备12仅仅是一个示例，不应对本申请实施例的功能和使用范围带来任何限制。
如图7所示，计算机设备12以通用计算设备的形式表现。计算 机设备12的组件可以包括但不限于：一个或者多个处理器或者处理单元16，系统存储器28，连接不同系统组件(包括系统存储器28和处理单元16)的总线18。
总线18表示几类总线结构中的一种或多种，包括存储器总线或者存储器控制器，外围总线，图形加速端口，处理器或者使用多种总线结构中的任意总线结构的局域总线。举例来说，这些体系结构包括但不限于工业标准体系结构(Industry Standard Architecture；以下简称：ISA)总线，微通道体系结构(Micro Channel Architecture；以下简称：MAC)总线，增强型ISA总线、视频电子标准协会(Video Electronics Standards Association；以下简称：VESA)局域总线以及外围组件互连(Peripheral Component Interconnection；以下简称：PCI)总线。
计算机设备12典型地包括多种计算机系统可读介质。这些介质可以是任何能够被计算机设备12访问的可用介质，包括易失性和非易失性介质，可移动的和不可移动的介质。
系统存储器28可以包括易失性存储器形式的计算机系统可读介质，例如随机存取存储器(Random Access Memory；以下简称：RAM)30和/或高速缓存存储器32。计算机设备12可以进一步包括其它可移动/不可移动的、易失性/非易失性计算机系统存储介质。仅作为举例，存储系统34可以用于读写不可移动的、非易失性磁介质(图7未显示，通常称为“硬盘驱动器”)。尽管图7中未示出，可以提供用于对可移动非易失性磁盘(例如“软盘”)读写的磁盘驱动器，以及对可移动非易失性光盘(例如：光盘只读存储器(Compact Disc Read Only Memory；以下简称：CD-ROM)、数字多功能只读光盘(Digital Video Disc Read Only Memory；以下简称：DVD-ROM)或者其它光介质)读写的光盘驱动器。在这些情况下，每个驱动器可以通过一个或者多个数据介质接口与总线18相连。存储器28可以包括至少一个程序产品，该程序产品具有一组(例如至少一个)程序模块，这些程序模块被配置以执行本申请各实施例的功能。
具有一组(至少一个)程序模块42的程序/实用工具40，可以存储在例如存储器28中，这样的程序模块42包括——但不限于——操作系统、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。程序模块42通常执行本申请所描述的实施例中的功能和/或方法。
计算机设备12也可以与一个或多个外部设备14(例如键盘、指向设备、显示器24等)通信，还可与一个或者多个使得用户能与该计算机设备12交互的设备通信，和/或与使得该计算机设备12能与一个或多个其它计算设备进行通信的任何设备(例如网卡，调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口22进行。并且，计算机设备12还可以通过网络适配器20与一个或者多个网络(例如局域网(Local Area Network；以下简称：LAN)，广域网(Wide Area Network；以下简称：WAN)和/或公共网络，例如因特网)通信。如图7所示，网络适配器20通过总线18与计算机设备12的其它模块通信。应当明白，尽管图7中未示出，可以结合计算机设备 12使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID系统、磁带驱动器以及数据备份存储系统等。
处理单元16通过运行存储在系统存储器28中的程序，从而执行各种功能应用以及数据处理，例如实现本申请实施例提供的违法停车案件的鉴别方法。
本申请实施例还提供一种计算机非易失性可读存储介质，其上存储有计算机程序，上述计算机程序被处理器执行时可以实现本申请实施例提供的违法停车案件的鉴别方法。
上述计算机非易失性可读存储介质可以采用一个或多个计算机可读的介质的任意组合。计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机存取存储器(RAM)、只读存储器(Read Only Memory；以下简称：ROM)、可擦式可编程只读存储器(Erasable Programmable Read Only Memory；以下简称：EPROM)或闪存、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本文件中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括——但不限于——电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括——但不限于——无线、电线、光缆、RF等等，或者上述的任意合适的组合。
可以以一种或多种程序设计语言或其组合来编写用于执行本申请操作的计算机程序代码，所述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络——包括局域网(Local Area Network；以下简称：LAN)或广域网(Wide Area Network；以下简称：WAN)连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本申请的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本申请的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现定制逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分，并且本申请的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能，这应被本申请的实施例所属技术领域的技术人员所理解。
取决于语境，如在此所使用的词语“如果”可以被解释成为“在……时”或“当……时”或“响应于确定”或“响应于检测”。类似地，取决于语境，短语“如果确定”或“如果检测(陈述的条件或事件)”可以被解释成为“当确定时”或“响应于确定”或“当检测(陈述的条件或事件)时”或“响应于检测(陈述的条件或事件)”。
需要说明的是，本申请实施例中所涉及的终端可以包括但不限于个人计算机(Personal Computer；以下简称：PC)、个人数字助理(Personal Digital Assistant；以下简称：PDA)、无线手持设备、平板电脑(Tablet Computer)、手机、MP3播放器、MP4播放器等。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如，多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元，可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介 质中，包括若干指令用以使得一台计算机装置(可以是个人计算机，服务器，或者网络装置等)或处理器(Processor)执行本申请各个实施例所述方法的部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory；以下简称：ROM)、随机存取存储器(Random Access Memory；以下简称：RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述仅为本申请的较佳实施例而已，并不用以限制本申请，凡在本申请的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本申请保护的范围之内。

Claims (20)

  1. 一种违法停车案件的鉴别方法，其特征在于，包括：
    获取违法停车案件的图像；
    对所述图像中的违法停车关键要素进行检测；
    对检测获得的所述违法停车关键要素中的信息进行识别；
    根据检测获得的所述违法停车关键要素、所述违法停车关键要素之间的位置关系和识别获得的所述违法停车关键要素中的信息，确定所述违法停车案件的违规类型和所述违法停车案件处理的合规性。
  2. 根据权利要求1所述的方法，其特征在于，所述违法停车案件的图像包括：交通执法人员拍摄的违法停车现场的图像、布控抓拍的违法停车现场的图像和/或违法停车现场视频的帧图像。
  3. 根据权利要求1或2所述的方法，其特征在于，所述对所述图像中的违法停车关键要素进行检测包括：
    对所述图像的尺寸和颜色分布进行归一化处理；
    利用预先训练的深度神经网络模型，对归一化处理后的图像进行图像识别，获得所述归一化处理后的图像中关键要素所在的区域和所述关键要素的类别，所述关键要素的类别包括以下之一或组合：车辆、所述车辆的车牌、交通标志和罚单。
  4. 根据权利要求3所述的方法，其特征在于，所述利用预先训练的深度神经网络模型，对归一化处理后的图像进行图像识别之前，还包括：
    利用训练图像和所述训练图像对应的标注文件，对训练模型进行训练，获得训练好的所述深度神经网络模型。
  5. 根据权利要求4所述的方法，其特征在于，所述利用训练图像和所述训练图像对应的标注文件，对训练模型进行训练包括：
    将所述训练图像和所述训练图像对应的标注文件输入所述训练模型，利用深度神经网络算法对所述训练模型进行训练；所述训练图像对应的标注文件包括所述训练图像中关键要素所在的区域和所述关键要素的类别；
    当所述训练模型输出的结果与所述训练图像对应的标注文件之间的误差小于预定阈值时，获得训练好的所述深度神经网络模型。
  6. 根据权利要求3所述的方法，其特征在于，所述违法停车关键要素中的信息包括以下之一或组合：所述车牌的车牌号码、所述禁止停车标志物的类型和所述罚单中的处罚信息；
    所述对检测获得的所述违法停车关键要素中的信息进行识别包括：
    通过车牌识别技术对所述车牌的车牌号码进行识别，通过光学字符识别对所述禁止停车标志物的类型和所述罚单中的处罚信息进行识别。
  7. 根据权利要求6所述的方法，其特征在于，所述根据检测获得的所述违法停车关键要素、所述违法停车关键要素之间的位置关系和识别获得的所述违法停车关键要素中的信息，确定所述违法停车案件的违规类型和所述违法停车案件处理的合规性包括：
    对检测获得的车辆的完整性进行检测；
    在确定所述车辆完整之后，确定检测获得的禁止停车标志物与所述车辆的位置关系；
    根据所述禁止停车标志物与所述车辆的位置关系，以及识别获得的禁 止停车标志物的类型，确定所述违法停车案件的违规类型；
    将所述违法停车案件的违规类型和识别获得的罚单中的处罚信息进行对比，根据所述违法停车案件的违规类型与所述罚单中的处罚信息是否匹配，确定所述违法停车案件处理的合规性。
  8. 一种违法停车案件的鉴别装置，其特征在于，包括：
    获取模块，用于获取违法停车案件的图像；
    检测模块，用于对所述获取模块获取的图像中的违法停车关键要素进行检测；
    识别模块，用于对所述检测模块检测获得的违法停车关键要素中的信息进行识别；
    确定模块，用于根据所述检测模块检测获得的违法停车关键要素、所述违法停车关键要素之间的位置关系和所述识别模块识别获得的违法停车关键要素中的信息，确定上述违法停车案件的违规类型和上述违法停车案件处理的合规性。
  9. 根据权利要求8所述的装置，其特征在于，所述违法停车案件的图像包括：交通执法人员拍摄的违法停车现场的图像、布控抓拍的违法停车现场的图像和/或违法停车现场视频的帧图像。
  10. 根据权利要求8或9所述的装置，其特征在于，
    所述检测模块，具体用于对所述图像的尺寸和颜色分布进行归一化处理；利用预先训练的深度神经网络模型，对归一化处理后的图像进行图像识别，获得所述归一化处理后的图像中关键要素所在的区域和所述关键要素的类别，所述关键要素的类别包括以下之一或组合：车辆、所述车辆的车牌、交通标志和罚单。
  11. 根据权利要求10所述的装置，其特征在于，还包括：
    训练模块，用于利用训练图像和所述训练图像对应的标注文件，对训练模型进行训练，获得训练好的深度神经网络模型。
  12. 根据权利要求11所述的装置，其特征在于，
    所述训练模块，具体用于将所述训练图像和所述训练图像对应的标注文件输入所述训练模型，利用深度神经网络算法对所述训练模型进行训练；所述训练图像对应的标注文件包括所述训练图像中关键要素所在的区域和所述关键要素的类别；当所述训练模型输出的结果与所述训练图像对应的标注文件之间的误差小于预定阈值时，获得训练好的所述深度神经网络模型。
  13. 根据权利要求10所述的装置，其特征在于，所述违法停车关键要素中的信息包括以下之一或组合：所述车牌的车牌号码、所述禁止停车标志物的类型和所述罚单中的处罚信息；
    所述识别模块，具体用于通过车牌识别技术对所述车牌的车牌号码进行识别，通过光学字符识别对所述禁止停车标志物的类型和所述罚单中的处罚信息进行识别。
  14. 根据权利要求13所述的装置，其特征在于，所述确定模块包括：
    完整性检测子模块，用于对所述检测模块检测获得的车辆的完整性进行检测；
    位置关系确定子模块，用于在所述完整性检测子模块确定所述 车辆完整之后，确定所述检测模块检测获得的禁止停车标志物与所述车辆的位置关系；
    违规类型确定子模块，用于根据所述禁止停车标志物与所述车辆的位置关系，以及所述识别模块识别获得的禁止停车标志物的类型，确定所述违法停车案件的违规类型；
    合规性确定子模块，用于将所述违规类型确定子模块确定的违法停车案件的违规类型和所述识别模块识别获得的罚单中的处罚信息进行对比，根据所述违法停车案件的违规类型与所述罚单中的处罚信息是否匹配，确定所述违法停车案件处理的合规性。
  15. 一种计算机设备，其特征在于，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序；
    所述处理器执行所述计算机程序时，用于获取违法停车案件的图像，对所述图像中的违法停车关键要素进行检测，对检测获得的所述违法停车关键要素中的信息进行识别，根据检测获得的所述违法停车关键要素、所述违法停车关键要素之间的位置关系和识别获得的所述违法停车关键要素中的信息，确定所述违法停车案件的违规类型和所述违法停车案件处理的合规性。
  16. 根据权利要求15所述的计算机设备，其特征在于，所述违法停车案件的图像包括：交通执法人员拍摄的违法停车现场的图像、布控抓拍的违法停车现场的图像和/或违法停车现场视频的帧图像。
  17. 根据权利要求15或16所述的计算机设备，其特征在于，所述处理器用于对所述图像中的违法停车关键要素进行检测包括：
    所述处理器，具体用于对所述图像的尺寸和颜色分布进行归一化处理；利用预先训练的深度神经网络模型，对归一化处理后的图像进行图像识别，获得所述归一化处理后的图像中关键要素所在的区域和所述关键要素的类别，所述关键要素的类别包括以下之一或组合：车辆、所述车辆的车牌、交通标志和罚单。
  18. 根据权利要求17所述的计算机设备，其特征在于，
    所述处理器，还用于在利用预先训练的深度神经网络模型，对归一化处理后的图像进行图像识别之前，利用训练图像和所述训练图像对应的标注文件，对训练模型进行训练，获得训练好的所述深度神经网络模型。
  19. 根据权利要求18所述的计算机设备，其特征在于，所述处理器用于利用训练图像和所述训练图像对应的标注文件，对训练模型进行训练包括：
    所述处理器，具体用于将所述训练图像和所述训练图像对应的标注文件输入所述训练模型，利用深度神经网络算法对所述训练模型进行训练；所述训练图像对应的标注文件包括所述训练图像中关键要素所在的区域和所述关键要素的类别；当所述训练模型输出的结果与所述训练图像对应的标注文件之间的误差小于预定阈值时，获得训练好的所述深度神经网络模型。
  20. 一种计算机非易失性可读存储介质，所述可读存储介质包括计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1-7中任一所述的方法。
PCT/CN2018/123548 2018-08-30 2018-12-25 违法停车案件的鉴别方法、装置和计算机设备 WO2020042489A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811004600.X 2018-08-30
CN201811004600.XA CN109147340A (zh) 2018-08-30 2018-08-30 违法停车案件的鉴别方法、装置和计算机设备

Publications (1)

Publication Number Publication Date
WO2020042489A1 true WO2020042489A1 (zh) 2020-03-05

Family

ID=64829544

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/123548 WO2020042489A1 (zh) 2018-08-30 2018-12-25 违法停车案件的鉴别方法、装置和计算机设备

Country Status (2)

Country Link
CN (1) CN109147340A (zh)
WO (1) WO2020042489A1 (zh)

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111694882A (zh) * 2020-06-02 2020-09-22 北京北大软件工程股份有限公司 一种处罚案件撤案的影响因素分析方法
CN111881713A (zh) * 2020-05-21 2020-11-03 北京嘀嘀无限科技发展有限公司 一种识别违停地点的方法、系统、装置及存储介质
CN112203054A (zh) * 2020-10-09 2021-01-08 平安国际智慧城市科技股份有限公司 监控视频点位标注方法、装置、存储介质及电子设备
CN112560711A (zh) * 2020-12-18 2021-03-26 深圳赛安特技术服务有限公司 非机动车交通违法判定方法、系统、装置及存储介质
CN112650876A (zh) * 2020-12-30 2021-04-13 北京嘀嘀无限科技发展有限公司 图像处理方法、装置、电子设备、存储介质和程序产品
CN112699776A (zh) * 2020-12-28 2021-04-23 南京星环智能科技有限公司 训练样本优化方法、目标检测模型生成方法、设备及介质
CN112861716A (zh) * 2021-02-05 2021-05-28 深圳市安软科技股份有限公司 违规物品摆放的监测方法、系统、设备及存储介质
CN113157775A (zh) * 2021-05-06 2021-07-23 湖北经济学院 流域案件特征信息挖掘系统及设备
CN113301499A (zh) * 2021-05-25 2021-08-24 宁波小遛共享信息科技有限公司 还车系统和还车方法
CN113378602A (zh) * 2020-03-10 2021-09-10 顺丰科技有限公司 物品违规码放检测方法、装置、服务器及存储介质
CN113470009A (zh) * 2021-07-26 2021-10-01 浙江大华技术股份有限公司 违规撑伞检测和识别方法、装置、电子设备及存储介质
WO2021209981A1 (en) * 2020-04-16 2021-10-21 Ratti Jayant Artificial intelligence and computer vision powered driving-performance assessment
CN113628430A (zh) * 2021-08-10 2021-11-09 珠海格力电器股份有限公司 违停报警方法、装置、电子设备及存储介质
CN113779251A (zh) * 2021-09-08 2021-12-10 平安国际智慧城市科技股份有限公司 线索信息获取方法、装置、介质及电子设备
CN113887420A (zh) * 2021-09-30 2022-01-04 浙江浩腾电子科技股份有限公司 一种基于ai的城市公共车位智能化检测识别系统
CN115938026A (zh) * 2023-03-13 2023-04-07 知鱼智联科技股份有限公司 基于多维特征的园区门禁管理方法、系统及存储介质
CN117424982A (zh) * 2023-10-16 2024-01-19 广州建软科技股份有限公司 一种智能布控球及其布控方法
CN112434601B (zh) * 2020-11-23 2024-05-10 平安科技（深圳）有限公司 基于行车视频的车辆违法检测方法、装置、设备及介质
CN112560711B (zh) * 2020-12-18 2024-06-07 深圳赛安特技术服务有限公司 非机动车交通违法判定方法、系统、装置及存储介质

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109993056A (zh) * 2019-02-25 2019-07-09 平安科技（深圳）有限公司 一种识别车辆违章行为的方法、服务器及存储介质
CN110009913B (zh) * 2019-03-27 2021-06-01 江苏智通交通科技有限公司 一种闯红灯车辆的非现场执法图片智能审核方法和系统
CN110766937B (zh) * 2019-05-22 2020-10-20 滴图(北京)科技有限公司 停车点识别方法、装置、电子设备以及可读存储介质
EP3751480A1 (en) * 2019-06-11 2020-12-16 Tata Consultancy Services Limited System and method for detecting on-street parking violations
CN110689028A (zh) * 2019-10-09 2020-01-14 中国人民公安大学 现场图测评方法、现场勘查记录测评方法及其装置
CN110718071B (zh) * 2019-10-25 2021-09-03 上海眼控科技股份有限公司 图像采集设备的校验方法、装置、计算机设备和存储介质
CN113538958A (zh) * 2020-04-14 2021-10-22 深圳富泰宏精密工业有限公司 停车警示方法、车载装置及可读存储介质
CN112270244A (zh) * 2020-10-23 2021-01-26 平安科技（深圳）有限公司 目标物违规监控方法、装置、电子设备及存储介质
CN112418061B (zh) * 2020-11-19 2024-01-23 城云科技（中国）有限公司 一种车辆禁停区域确定方法及系统
CN113284350B (zh) * 2021-05-21 2022-08-16 深圳市大道至简信息技术有限公司 一种基于激光的违停检测方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103065473A (zh) * 2012-11-29 2013-04-24 太原理工大学 一种违法停车车辆管理系统及方法
US8682036B2 (en) * 2012-04-06 2014-03-25 Xerox Corporation System and method for street-parking-vehicle identification through license plate capturing
CN106611500A (zh) * 2017-02-20 2017-05-03 郑州云海信息技术有限公司 一种违规取证与处罚装置、方法及违规管理系统
CN107679531A (zh) * 2017-06-23 2018-02-09 平安科技（深圳）有限公司 基于深度学习的车牌识别方法、装置、设备及存储介质
CN108389398A (zh) * 2018-04-27 2018-08-10 希社（上海）智能交通科技有限公司 违法停车取证系统及方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07114688A (ja) * 1993-10-14 1995-05-02 Nippon Signal Co Ltd:The 車両登録番号認識方法
US20090309974A1 (en) * 2008-05-22 2009-12-17 Shreekant Agrawal Electronic Surveillance Network System
CN102915641B (zh) * 2012-10-16 2016-04-13 浪潮集团有限公司 一种基于密码技术的交通违章信息处理方法
CN104504896B (zh) * 2014-08-15 2017-06-27 上海市政工程设计研究总院（集团）有限公司 一种交通违法自动记录系统及方法
CN105761500B (zh) * 2016-05-10 2019-02-22 腾讯科技（深圳）有限公司 交通事故处理方法及交通事故处理装置
KR101698026B1 (ko) * 2016-06-17 2017-01-19 주식회사 파킹패스 이동차량 추적에 의한 cctv 불법주정차 단속시스템
CN107688764B (zh) * 2016-08-03 2020-04-10 浙江宇视科技有限公司 检测车辆违章的方法及装置
CN106205144B (zh) * 2016-09-07 2018-06-19 东南大学 高速公路应急车道占用监督处罚方法和系统
CN106781525A (zh) * 2017-02-28 2017-05-31 广州交通信息化建设投资营运有限公司 一种车辆智能识别监控方法
CN108335489A (zh) * 2018-03-11 2018-07-27 西安电子科技大学 高速公路车辆行为语义分析及异常行为监控系统及方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8682036B2 (en) * 2012-04-06 2014-03-25 Xerox Corporation System and method for street-parking-vehicle identification through license plate capturing
CN103065473A (zh) * 2012-11-29 2013-04-24 太原理工大学 一种违法停车车辆管理系统及方法
CN106611500A (zh) * 2017-02-20 2017-05-03 郑州云海信息技术有限公司 一种违规取证与处罚装置、方法及违规管理系统
CN107679531A (zh) * 2017-06-23 2018-02-09 平安科技（深圳）有限公司 基于深度学习的车牌识别方法、装置、设备及存储介质
CN108389398A (zh) * 2018-04-27 2018-08-10 希社（上海）智能交通科技有限公司 违法停车取证系统及方法

Cited By (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113378602A (zh) * 2020-03-10 2021-09-10 顺丰科技有限公司 物品违规码放检测方法、装置、服务器及存储介质
WO2021209981A1 (en) * 2020-04-16 2021-10-21 Ratti Jayant Artificial intelligence and computer vision powered driving-performance assessment
AU2021255130B2 (en) * 2020-04-16 2024-01-18 Jayant RATTI Artificial intelligence and computer vision powered driving-performance assessment
CN111881713A (zh) * 2020-05-21 2020-11-03 北京嘀嘀无限科技发展有限公司 一种识别违停地点的方法、系统、装置及存储介质
CN111694882A (zh) * 2020-06-02 2020-09-22 北京北大软件工程股份有限公司 一种处罚案件撤案的影响因素分析方法
CN112203054A (zh) * 2020-10-09 2021-01-08 平安国际智慧城市科技股份有限公司 监控视频点位标注方法、装置、存储介质及电子设备
CN112434601B (zh) * 2020-11-23 2024-05-10 平安科技（深圳）有限公司 基于行车视频的车辆违法检测方法、装置、设备及介质
CN112560711A (zh) * 2020-12-18 2021-03-26 深圳赛安特技术服务有限公司 非机动车交通违法判定方法、系统、装置及存储介质
CN112560711B (zh) * 2020-12-18 2024-06-07 深圳赛安特技术服务有限公司 非机动车交通违法判定方法、系统、装置及存储介质
CN112699776A (zh) * 2020-12-28 2021-04-23 南京星环智能科技有限公司 训练样本优化方法、目标检测模型生成方法、设备及介质
CN112699776B (zh) * 2020-12-28 2022-06-21 南京星环智能科技有限公司 训练样本优化方法、目标检测模型生成方法、设备及介质
CN112650876A (zh) * 2020-12-30 2021-04-13 北京嘀嘀无限科技发展有限公司 图像处理方法、装置、电子设备、存储介质和程序产品
CN112861716A (zh) * 2021-02-05 2021-05-28 深圳市安软科技股份有限公司 违规物品摆放的监测方法、系统、设备及存储介质
CN113157775A (zh) * 2021-05-06 2021-07-23 湖北经济学院 流域案件特征信息挖掘系统及设备
CN113157775B (zh) * 2021-05-06 2022-10-11 湖北经济学院 流域案件特征信息挖掘系统及设备
CN113301499A (zh) * 2021-05-25 2021-08-24 宁波小遛共享信息科技有限公司 还车系统和还车方法
CN113301499B (zh) * 2021-05-25 2022-08-19 宁波小遛共享信息科技有限公司 还车系统和还车方法
CN113470009A (zh) * 2021-07-26 2021-10-01 浙江大华技术股份有限公司 违规撑伞检测和识别方法、装置、电子设备及存储介质
CN113470009B (zh) * 2021-07-26 2024-05-14 浙江大华技术股份有限公司 违规撑伞检测和识别方法、装置、电子设备及存储介质
CN113628430A (zh) * 2021-08-10 2021-11-09 珠海格力电器股份有限公司 违停报警方法、装置、电子设备及存储介质
CN113779251B (zh) * 2021-09-08 2024-04-19 平安国际智慧城市科技股份有限公司 线索信息获取方法、装置、介质及电子设备
CN113779251A (zh) * 2021-09-08 2021-12-10 平安国际智慧城市科技股份有限公司 线索信息获取方法、装置、介质及电子设备
CN113887420A (zh) * 2021-09-30 2022-01-04 浙江浩腾电子科技股份有限公司 一种基于ai的城市公共车位智能化检测识别系统
CN115938026B (zh) * 2023-03-13 2023-05-30 知鱼智联科技股份有限公司 基于多维特征的园区门禁管理方法、系统及存储介质
CN115938026A (zh) * 2023-03-13 2023-04-07 知鱼智联科技股份有限公司 基于多维特征的园区门禁管理方法、系统及存储介质
CN117424982A (zh) * 2023-10-16 2024-01-19 广州建软科技股份有限公司 一种智能布控球及其布控方法
CN117424982B (zh) * 2023-10-16 2024-04-16 广州建软科技股份有限公司 一种智能布控球及其布控方法

Also Published As

Publication number Publication date
CN109147340A (zh) 2019-01-04

Similar Documents

Publication Publication Date Title
WO2020042489A1 (zh) 违法停车案件的鉴别方法、装置和计算机设备
WO2020233270A1 (zh) 票据分析方法、分析装置、计算机设备和介质
JP6527678B2 (ja) プライバシー保護のための延滞車両特定
CN110163176B (zh) 车道线变化位置识别方法、装置、设备和介质
US20210192227A1 (en) Method and apparatus for detecting parking space usage condition, electronic device, and storage medium
CN103279756B (zh) 基于集成分类器的车辆检测分析系统及其检测分析方法
CN112052850B (zh) 车牌识别方法、装置、电子设备及存储介质
CN108694399B (zh) 车牌识别方法、装置及系统
CN109325424A (zh) 交通违章案件的鉴别方法、装置和计算机设备
WO2020038138A1 (zh) 样本标注方法及装置、损伤类别的识别方法及装置
CN112016510B (zh) 基于深度学习的信号灯识别方法和装置、设备及存储介质
US20220036115A1 (en) Method and apparatus for recognizing imaged information-bearing medium, computer device and medium
CN111178357B (zh) 车牌识别方法、系统、设备和存储介质
WO2022052375A1 (zh) 车辆识别方法及装置、电子设备及存储介质
CN112257413B (zh) 地址参数处理方法及相关设备
CN113055823A (zh) 一种基于路侧停车管理共享单车的方法及装置
CN112001269A (zh) 车辆识别方法、装置、计算机设备和存储介质
CN112861567A (zh) 一种车型分类方法及装置
CN113221894A (zh) 车辆的车牌号识别方法、装置、电子设备和存储介质
CN116311327B (zh) 处方图像检测方法
CN115810132B (zh) 一种裂缝朝向识别方法、装置、设备及存储介质
CN111898540A (zh) 车道线检测方法、装置、计算机设备及计算机可读存储介质
CN114141022A (zh) 应急车道占用行为检测方法、装置、电子设备及存储介质
CN111985304A (zh) 巡防告警方法、系统、终端设备及存储介质
CN114693722B (zh) 一种车辆行驶行为检测方法、检测装置及检测设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18931893

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 09.06.2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18931893

Country of ref document: EP

Kind code of ref document: A1