WO2018090771A1 - 一种车牌识别方法及装置 - Google Patents

一种车牌识别方法及装置 Download PDF

Info

Publication number
WO2018090771A1
WO2018090771A1 PCT/CN2017/106022 CN2017106022W WO2018090771A1 WO 2018090771 A1 WO2018090771 A1 WO 2018090771A1 CN 2017106022 W CN2017106022 W CN 2017106022W WO 2018090771 A1 WO2018090771 A1 WO 2018090771A1
Authority
WO
WIPO (PCT)
Prior art keywords
license plate
character
area
plate area
recognition result
Prior art date
Application number
PCT/CN2017/106022
Other languages
English (en)
French (fr)
Inventor
何海峰
Original Assignee
杭州海康威视数字技术股份有限公司
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 杭州海康威视数字技术股份有限公司 filed Critical 杭州海康威视数字技术股份有限公司
Publication of WO2018090771A1 publication Critical patent/WO2018090771A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/60Type of objects
    • G06V20/62Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/63Scene text, e.g. street names
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/24Aligning, centring, orientation detection or correction of the image
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/26Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/26Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/60Type of objects
    • G06V20/62Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/625License plates

Definitions

  • the present application relates to the field of intelligent transportation technologies, and in particular, to a license plate recognition method and device.
  • the license plate is the “identity card” of the vehicle and is an important information that distinguishes it from other motor vehicles.
  • the license plate recognition technology has been widely used in scenes such as bayonet, parking lot and electronic police to obtain the license plate information of vehicles in the scene, and exerts the power of “intelligent traffic algorithm” in many aspects such as public security management.
  • the license plate can be divided into double-layer license plates and single-layer license plates.
  • a double-layer license plate is a form of presence of a license plate.
  • the characters in the license plate are distributed in the upper and lower layers.
  • Figure 1 shows an example of a partial double deck license plate.
  • the license plate image to be identified it is necessary to match a plurality of pre-stored plurality of double-layer license plate templates to identify the license plate number.
  • the specific process is: for the license plate image to be identified, the license plate area is located according to the selected double deck license plate template, the upper layer character and the lower layer character of the license plate are separated from the license plate area, and the character segmentation result is subjected to character recognition, if If a reliable character is recognized, it is considered that the above-mentioned double-layer license plate template is successfully matched, and finally the recognition result of the license plate number is output; if no reliable character is recognized, another double-layer license plate template is selected, and the above process is repeated.
  • the license plate number of the license plate in the license plate image to be identified can be identified.
  • the license plate area must be completely executed every time the matching process is performed. The process of segmentation and recognition of the characters on the upper and lower layers of the license plate has a low efficiency in the license plate recognition process.
  • the purpose of the embodiment of the present application is to provide a license plate recognition method and device, which can improve the efficiency of the license plate recognition process.
  • the present application discloses a license plate recognition method, and the method includes:
  • the license plate area comprises: an upper license plate area and a lower license plate area
  • the double-layer license plate character distribution features include: an upper layer character feature corresponding to the upper license plate area, a lower layer character feature corresponding to the lower license plate area, and an upper license plate area and a lower layer The relative positional relationship of the license plate area;
  • the step of determining the second license plate area of the license plate number in the to-be-identified license plate image including:
  • the step of determining, according to the first license plate area and the pre-stored double-layer license plate character distribution feature, the second license plate area of the to-be-identified license plate image that does not identify the license plate number includes:
  • the step of identifying the characters in the second license plate area to obtain the second character recognition result includes:
  • the step of dividing the second license plate area to obtain a target character area includes:
  • the size correspondence is: a size of an upper license plate area character and a lower license plate area character Correspondence between dimensions;
  • the step of determining whether all the characters in the first character recognition result are located in the same row includes:
  • a license plate recognition device which includes:
  • a first area determining module configured to obtain a license plate image to be identified, and determine a first license plate area in the to-be-identified license plate image
  • a first result identification module configured to identify a character in the first license plate area, and obtain a first character recognition result
  • a recognition result judging module configured to determine whether all characters in the first character recognition result are located in the same row
  • a second area determining module configured to determine the to-be-identified license plate image according to the first license plate area and a pre-stored double-layer license plate character distribution feature when all characters in the first character recognition result are located in the same row
  • a second result identification module configured to identify a character in the second license plate area, and obtain a second character recognition result
  • a recognition result synthesizing module configured to synthesize the first character recognition result and the second character recognition result, and obtain a license plate number of the to-be-identified license plate image.
  • the license plate area comprises: an upper license plate area and a lower license plate area
  • the double-layer license plate character distribution features include: an upper layer character feature corresponding to the upper license plate area, a lower layer character feature corresponding to the lower license plate area, and an upper license plate area and a lower layer The relative positional relationship of the license plate area;
  • the second area determining module includes:
  • a matching submodule configured to match a license plate region to which the first license plate region belongs according to the character feature, the upper layer character feature, and the lower layer character feature in the first character recognition result;
  • a first determining submodule configured to determine, according to the matching result, the first license plate area, and the relative positional relationship, a second license plate area in which the license plate number is not recognized in the to-be-identified license plate image.
  • the second area determining module includes:
  • a correction submodule configured to correct the first license plate area according to the first character recognition result
  • the second determining sub-module is configured to determine, according to the modified license plate area and the pre-stored double-layer license plate character distribution feature, the second license plate area in the license plate image to be recognized that the license plate number is not recognized.
  • the second result identification module includes:
  • a dividing sub-module configured to divide the second license plate area to obtain a target character area
  • the identification submodule is configured to identify characters in the target character region to obtain a second character recognition result.
  • the segmentation submodule includes:
  • a determining unit configured to determine a second character size according to the first character size and a preset size correspondence, wherein the size correspondence is between: a size of an upper license plate area character and a size of a lower license plate area character Correspondence relationship;
  • a dividing unit configured to divide the second license plate area according to the second character size to obtain a target character area.
  • the identification result judging module is specifically configured to:
  • an electronic device suitable for license plate recognition comprising:
  • the circuit board is disposed inside the space enclosed by the housing, the processor and the memory are disposed on the circuit board; and the power supply circuit is used for each circuit of the electronic device or The device is powered;
  • the memory is for storing executable program code;
  • the processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory for performing the following steps:
  • the present application discloses an application program for executing a license plate recognition method provided by an embodiment of the present application at runtime.
  • the license plate recognition method includes:
  • the present application discloses a storage medium for storing executable code, which is used to execute the license plate recognition method provided by the embodiment of the present application at runtime.
  • the license plate recognition method includes:
  • the first character recognition result, and the pre-stored double-layer license plate character distribution feature determine the second license plate area of the license plate image that is not recognized in the license plate image, and need to match a large number of double-layer license plate templates, thereby improving the license plate recognition process. effectiveness.
  • Figure 1 is an example of a partial double-layer license plate image
  • FIG. 2 is a schematic flow chart of a license plate recognition method according to an embodiment of the present application.
  • FIG. 3 is another schematic flowchart of a license plate recognition method according to an embodiment of the present application.
  • FIG. 4 is a schematic structural diagram of a license plate recognition device according to an embodiment of the present application.
  • FIG. 5 is another schematic structural diagram of a license plate recognition device according to an embodiment of the present application.
  • the embodiment of the present application provides a license plate recognition method and device, which are applied to an electronic device, such as a computer, a tablet computer, a smart phone, and the like, which can improve the efficiency of the license plate recognition process.
  • FIG. 2 is a schematic flowchart of a license plate recognition method according to an embodiment of the present disclosure, which is applied to an electronic device, and the method includes the following steps:
  • Step S201 Obtain an image of the license plate to be identified, and determine a first license plate area in the license plate image to be identified.
  • the license plate image to be identified may be a vehicle image captured on the road.
  • the image of the license plate to be identified may be any image including the license plate number, which is not specifically limited in the embodiment of the present application.
  • the license plate in the license plate image to be identified may be a double deck license plate or a single deck license plate. Double decker The card refers to the license plate number distributed in the upper and lower two rows of license plates, that is, the upper license plate area and the lower license plate area.
  • the license plates shown in Figure 1 are all double-layer license plates.
  • a single-layer license plate refers to a license plate with a license plate number distributed in one line.
  • the first license plate region in the license plate image to be identified may be determined according to the texture feature of the preset license plate region. The specific process will not be repeated.
  • the first license plate area may include the upper license plate area and the lower license plate area, and may only include the upper license plate area or the lower license plate area.
  • Step S202 Identify characters in the first license plate area to obtain a first character recognition result.
  • the first character recognition result includes a character and a corresponding character type, a character area, and the like, and the character area is an area corresponding to the character in the image.
  • the first license plate area may be first divided by a vertical projection method and/or a connected domain method to obtain a character segmentation result, and then the preset is adopted.
  • the character recognizer recognizes the result of the character segmentation and obtains the first character recognition result.
  • Step S203 determining whether all characters in the first character recognition result are located in the same row, and if yes, indicating that the first character recognition result is not a recognition result of the double-layer license plate, that is, the first license plate area is not a double-layer license plate area, Step S204 can be continued.
  • the first character recognition result is the recognition result for the double-layer license plate, that is, the first license plate area belongs to the double-layer license plate area, that is, the second license plate area of the license plate image to be identified does not exist in the license plate image to be identified.
  • the first license plate area is not a double-layer license plate area, but it cannot be determined whether the license plate in the license plate image to be identified belongs to the double-layer license plate or the single-layer license plate, and only the first character can be determined.
  • the recognition result is not the result of recognition for the double deck license plate.
  • step S203 is to remove the case where the first character recognition result is already a double-layer license plate recognition result, in which case it is not necessary to continue to recognize the second license plate area.
  • Step S204 Determine, according to the first license plate area and the pre-stored double-layer license plate character distribution feature, a second license plate area in which the license plate number is not recognized in the license plate image to be identified.
  • the double-layer license plate character distribution feature may include a character feature of the upper license plate area and a character feature of the lower license plate area, a color correspondence relationship between the upper license plate area and a color correspondence relationship of the lower license plate area, and an upper license plate area and a lower license plate area.
  • Relative positional relationship is the color correspondence between the foreground color and the background color.
  • the color correspondence includes: the foreground is black, the background is white (black and white) and the foreground is white, and the background is black (white and black); characters Features may include character types and corresponding numbers, and the like.
  • the double-layer license plate character distribution feature may include the following contents:
  • the character features of the upper license plate area include: the character type includes a letter type and a numeric type, the number of characters of the letter type is 2 to 3, and the number of characters of the numeric type is 0 to 2.
  • the character features of the lower license plate area include: the character type includes the letter type and the numeric type, the number of characters of the letter type is 0 to 1, and the number of characters of the numeric type is 2 to 4.
  • the color correspondence between the upper license plate area and the color correspondence of the lower license plate area are the same, both: white and black.
  • the relative positional relationship between the upper license plate area and the lower license plate area includes that the upper license plate area is located directly above the lower license plate area and is within a specified range of N pixels from the lower license plate area.
  • the double-layer license plate character distribution feature may be pre-stored. Due to the large difference in license plate characteristics between different regions, it is possible to obtain double-layer license plate character distribution characteristics for double-layer license plates in the same area.
  • the double-layer license plate images numbered 1 to 11 belong to one region
  • the double-layer license plate images numbered 12 to 20 belong to another region
  • the license plates for the two regions can be targeted.
  • the double-layer license plate image sample in the designated area may be collected, the upper license plate character and the lower-level license plate character in the sample are marked, and the corresponding double-layer character distribution feature of the region is extracted and saved according to the mark in the sample and the sample.
  • Step S205 Identify characters in the second license plate area to obtain a second character recognition result.
  • step S203 Since the result of the determination in step S203 is YES, only all the characters in the first character recognition result are determined to be in the same line, and it is not determined whether the license plate in the license plate image to be identified belongs to the double-layer license plate or the single-layer license plate, and therefore, in the identification When the characters in the second license plate area are present, there may be cases where the recognition result is unsuccessful.
  • the method may further include: determining, according to the second character recognition result, whether the character recognition for the second license plate area is successful, if If the recognition is successful, step S206 is performed. If it is not recognized successfully, it will not be processed.
  • step S202 when the characters in the second license plate area are identified, the same process as step S202 may be used, or a different process from step S202 may be used, and the detailed process is not described in detail in this embodiment.
  • Step S206 synthesize the first character recognition result and the second character recognition result, and obtain the license plate number of the license plate image to be identified.
  • the first character recognition result and the second character recognition result may be synthesized according to the relative positional relationship between the first license plate area and the second license plate area, thereby obtaining the to-be-identified The license plate number of the license plate image.
  • the first character recognition result is placed in the second character recognition result.
  • the first character recognition result and the first character recognition result may be combined to obtain the license plate number of the license plate image to be identified.
  • the efficiency of the license plate recognition process can be improved.
  • the first license plate area belongs to the upper license plate area or the lower license plate area
  • this implementation For example, the second license plate area of the license plate image in which the license plate number is not recognized can be located, and finally the complete double deck license plate is recognized.
  • step S203 the determining whether all the characters in the first character recognition result are located in the same row may include:
  • the first character recognition result includes each character region, and according to the character regions, it can be determined whether the characters are located in the same row.
  • step S204 determining the image of the license plate to be identified according to the first license plate area and the pre-stored double-layer license plate character distribution feature.
  • the second license plate area in which the license plate number is not recognized may include:
  • correcting the first license plate area according to the first character recognition result can remove the influence of the interference factor, make the range of the first license plate area more accurate, and improve the accuracy and reliability of the detection.
  • the method may include: determining, according to the first character recognition result, a character area in which the character recognition is successful in the first license plate area, and successfully identifying the character according to the character recognition Area, determine the corrected license plate area.
  • a rivet area exists on the leftmost side of the first license plate area, and according to the first character recognition result, it can be known that the character recognition result of the rivet area is unidentified successfully, and the rivet area can be removed from the first license plate area.
  • FIG. 3 is another schematic flowchart of a license plate recognition method according to an embodiment of the present application, which is a modification of the embodiment shown in FIG. 2.
  • the license plate area includes: an upper license plate area and a lower license plate area
  • the double-layer license plate character distribution features include: an upper-level character feature corresponding to the upper license plate area, a lower-level character feature corresponding to the lower license plate area, and an upper license plate area and a lower license plate area. Relative positional relationship.
  • the upper character feature may include: the character type is “letter + number”, the total number of characters is 1 to 4, the number of letters is 1 to 3, and the number of digits is 0 to 2; the character of the lower layer can include: the type of the character is "letter + number”, the total number of characters is 2 to 5, the number of letters is 0 to 1, and the number of digits is 2 to 4; the relative positional relationship includes: the upper license plate area It coincides with the center of symmetry of the lower license plate area.
  • step S204 in the embodiment shown in FIG. 2 determining, according to the first license plate area and the pre-stored double-layer license plate character distribution feature, a second license plate area in which the license plate number is not recognized in the to-be-identified license plate image.
  • the method may include:
  • Step S204A Matching the license plate area to which the first license plate area belongs according to the character feature, the upper character feature and the lower character feature in the first character recognition result.
  • the upper character features and the lower character features are listed above. If the first character recognition result is 641, since the number of digits in the result is 3, the feature of "the number of digits is 0 to 2" in the upper layer character feature is not met, so it can be determined that the result conforms to the lower layer character feature, and the first license plate is matched. The area belongs to the lower license plate area. If the first character recognition result is 1M4U, since the number of digits and the number of letters in the result are both 2, which does not meet the feature of "the number of letters is 0 to 1" in the lower character feature, it can be determined that the result conforms to the upper character feature and matches. The first license plate area is obtained as belonging to the upper license plate area.
  • Step S204B Determine, according to the matching result, the first license plate area, and the relative positional relationship, a second license plate area in which the license plate number is not recognized in the license plate image to be identified.
  • the matching first license plate area belongs to the lower license plate area
  • the matching obtains the first license plate area belonging to the upper license plate area, it may be determined that the second license plate area belongs to the lower license plate area, and when the second license plate area is determined, the preset range below the first license plate area may be determined as the second license plate area.
  • the preset range may be determined by determining a range of a single character in the first license plate area according to the first character recognition result, and determining a product of a range of the single character and the preset value as the preset range.
  • the product of the width range of the single character and the first preset value may be determined as the width of the preset range
  • the product of the height range of the single character and the second preset value may be determined as the height of the preset range.
  • the first preset value may be determined according to the number of single-layer characters of a large number of sample license plates, for example, 6-8
  • the second preset value may be determined according to the height of a plurality of sample license plate single-layer characters, for example, 1 to 1.5.
  • determining, according to the first character recognition result, the width of a single character in the first license plate area is 5
  • the number of pixels is 8 pixels.
  • the second preset value can be taken as 1.5.
  • an area of 30 pixel width and 12 pixel height directly below the first license plate area may be determined as the second license plate area.
  • the second license plate area is determined according to the first character recognition result and the double-layer license plate character distribution feature, and the second license plate area is not required to be searched from the entire unidentified license plate image.
  • the interference of other content in the image of the license plate to be identified can be reduced, and the second license plate area can be positioned more accurately and quickly.
  • step S204B according to the matching result, the first license plate area and the relative positional relationship, determining the second number of the license plate number not recognized in the license plate image to be identified.
  • the license plate area may specifically include:
  • the left and right boundaries of the first license plate area may be re-locked according to the character area corresponding to the character recognized by the first character recognition result.
  • correcting the first license plate area according to the first character recognition result can improve the accuracy of the first license plate area, thereby improving the accuracy of the entire license plate recognition process.
  • the step S205 in the embodiment shown in FIG. 2 the step of identifying the characters in the second license plate area, and obtaining the second character recognition result may specifically include:
  • Step S205A Dividing the second license plate area to obtain a target character area.
  • the second license plate area may be divided according to the vertical projection method and/or the connected domain method.
  • step S205B is performed, and if the segmentation is unsuccessful, the segmentation process is not processed. .
  • Step S205B Identify characters in the target character area to obtain a second character recognition result.
  • the characters in the target character area are recognized, the characters of the target character area can be recognized according to the preset character classifier.
  • step S205A segmenting the second license plate region to obtain the target character region may include:
  • Step 1 Obtain a first character size according to a size of each character region in the first character recognition result.
  • the size may be at least one of a width and a height.
  • the first character size is a size value capable of representing the size of each character region in the first character recognition result.
  • Step 2 Determine a second character size according to the first character size and a preset size correspondence, wherein the size correspondence is: a correspondence between a size of an upper license plate area character and a size of a lower license plate area character relationship.
  • the above size correspondence may be that the width of the character of the upper license plate area is equal to the width of the character of the lower license plate area.
  • Step 3 Dividing the second license plate area according to the second character size to obtain a target character area.
  • the second license plate area when the second license plate area is divided, the second license plate area may be first divided by the vertical projection method and/or the connected domain method, and then, according to the first division, based on the second character size, Correct the result of the first split.
  • the accuracy of the character segmentation result can be improved by dividing the character of the second license plate region according to the first character recognition result and the correspondence between the size of the upper license plate region character and the size of the lower license plate region character.
  • FIG. 4 is a schematic flowchart of a license plate recognition device according to an embodiment of the present disclosure.
  • the device is applied to an electronic device corresponding to the embodiment shown in FIG. 2, and the device includes:
  • a first area determining module 401 configured to obtain a license plate image to be identified, and determine a first license plate area in the to-be-identified license plate image;
  • a first result identification module 402 configured to identify a character in the first license plate area, and obtain a first character recognition result
  • the recognition result judging module 403 is configured to determine whether all the characters in the first character recognition result are located in the same row;
  • a second area determining module 404 configured to determine the to-be-identified license plate according to the first license plate area and a pre-stored double-layer license plate character distribution feature when all characters in the first character recognition result are located in the same row a second license plate area in which no license plate number is recognized in the image;
  • a second result recognition module 405, configured to identify a character in the second license plate area, and obtain a second character recognition result
  • the recognition result synthesizing module 406 is configured to synthesize the first character recognition result and the second character recognition result to obtain a license plate number of the to-be-identified license plate image.
  • the identification result determining module 403 may be specifically configured to:
  • the second area determining module 404 may specifically include:
  • a correction submodule (not shown) for correcting the first license plate area according to the first character recognition result
  • a second determining sub-module (not shown), configured to determine, according to the modified license plate area and the pre-stored double-layer license plate character distribution feature, the second license plate area of the to-be-identified license plate image that does not recognize the license plate number .
  • FIG. 5 is another schematic structural diagram of a license plate recognition method according to an embodiment of the present application.
  • the embodiment is based on the improvement of the embodiment shown in FIG. 4, and the unmodified portion is the same as the embodiment shown in FIG. 4.
  • This embodiment corresponds to the method embodiment shown in FIG.
  • the license plate area includes: an upper license plate area and a lower license plate area
  • the double-layer license plate character distribution features include: an upper layer character feature corresponding to the upper license plate area, a lower layer character feature corresponding to the lower license plate area, and an upper license plate area and The relative positional relationship of the lower license plate area.
  • the second area determining module 404 may specifically include:
  • the matching sub-module 501 is configured to match the license plate area to which the first license plate area belongs according to the character feature, the upper layer character feature and the lower layer character feature in the first character recognition result;
  • the first determining sub-module 502 is configured to determine, according to the matching result, the first license plate area and the relative positional relationship, a second license plate area in which the license plate number is not recognized in the to-be-identified license plate image.
  • the second result identification module 405 may specifically include:
  • a dividing sub-module 503 configured to divide the second license plate area to obtain a target character area
  • the identification sub-module 504 is configured to identify characters in the target character region to obtain a second character recognition result.
  • the segmentation sub-module 503 may specifically include:
  • a determining unit (not shown), configured to determine a second character size according to the first character size and a preset size correspondence, wherein the size correspondence is: a size of the upper license plate area character and a lower layer Correspondence between the sizes of the license plate area characters;
  • a dividing unit (not shown) for dividing the second license plate area according to the second character size to obtain a target character area.
  • the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.
  • the embodiment of the present application provides an electronic device, which is suitable for license plate recognition, and the electronic device includes:
  • the circuit board is disposed inside the space enclosed by the housing, the processor and the memory are disposed on the circuit board; and the power supply circuit is used for each circuit of the electronic device or The device is powered;
  • the memory is for storing executable program code;
  • the processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory for performing the following steps:
  • the electronic device can exist in various forms, including but not limited to:
  • Mobile communication devices These devices are characterized by mobile communication functions and are mainly aimed at providing voice and data communication.
  • Such terminals include: smart phones (such as iPhone), multimedia phones, functional phones, and low-end phones.
  • Ultra-mobile PC devices These devices belong to the category of personal computers and have calculations. And processing functions, generally also have mobile Internet features.
  • Such terminals include: PDAs, MIDs, and UMPC devices, such as the iPad.
  • Portable entertainment devices These devices can display and play multimedia content. Such devices include: audio, video players (such as iPod), handheld game consoles, e-books, and smart toys and portable car navigation devices.
  • the server consists of a processor, a hard disk, a memory, a system bus, etc.
  • the server is similar to a general-purpose computer architecture, but because of the need to provide highly reliable services, processing power and stability High reliability in terms of reliability, security, scalability, and manageability.
  • the first embodiment of the present application can determine the number of the license plate number that is not recognized in the license plate image to be identified.
  • the second license plate area there is no need to match a large number of double-layer license plate templates, so the efficiency of the license plate recognition process can be improved.
  • the embodiment of the present application provides an application program for executing the license plate recognition method provided by the embodiment of the present application at runtime.
  • the license plate recognition method includes:
  • the first embodiment of the present application can determine the number of the license plate number that is not recognized in the license plate image to be identified.
  • the second license plate area there is no need to match a large number of double-layer license plate templates, so the efficiency of the license plate recognition process can be improved.
  • the present application provides a storage medium for storing executable code, which is used at runtime to perform the license plate recognition method provided by the embodiments of the present application.
  • the license plate recognition method includes:
  • the first embodiment of the present application can determine the number of the license plate number that is not recognized in the license plate image to be identified.
  • the second license plate area there is no need to match a large number of double-layer license plate templates, so the efficiency of the license plate recognition process can be improved.
  • the storage medium referred to herein means a ROM/RAM, a magnetic disk, an optical disk, or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (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)
  • Character Discrimination (AREA)
  • Character Input (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种车牌识别方法及装置,所述车牌识别方法包括:获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域(S201);识别所述第一车牌区域中的字符,获得第一字符识别结果(S202);判断所述第一字符识别结果中的所有字符是否位于同一行内(S203);如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域(S204);识别所述第二车牌区域中的字符,获得第二字符识别结果(S205);合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码(S206)。该方法能够提高车牌识别过程的效率。

Description

一种车牌识别方法及装置
本申请要求于2016年11月16日提交中国专利局、申请号为201611032450.4、发明名称为“一种车牌识别方法及装置”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
本申请涉及智能交通技术领域,特别涉及一种车牌识别方法及装置。
背景技术
车牌是车辆的“身份证”,是区别于其他机动车辆的一项重要信息。车牌识别技术已被广泛应用在卡口、停车场和电子警察等场景中,以获取场景内车辆的号牌信息,在治安管理等众多方面发挥着“智能交通算法”的威力。
根据车牌中字符分布的位置,可以将车牌分成双层车牌和单层车牌。双层车牌是车牌的一种存在形式,这种车牌中的字符分布在上、下两层中。全球各个国家和地区的双层车牌种类繁多,并无统一的标准。图1所示为部分双层车牌的实例图。
在识别双层车牌时,现有技术中,通常针对待识别车牌图像,需要一一匹配预先保存的多个双层车牌模板,进而识别出车牌号码。具体的过程是,针对待识别车牌图像,根据选定的某一双层车牌模板定位车牌区域,从上述车牌区域中分割出车牌的上层字符和下层字符,再对字符分割结果进行字符识别,如果识别出可靠的字符,则认为上述双层车牌模板匹配成功了,最终输出车牌号码的识别结果;如果没有识别出可靠的字符,则选择另一个双层车牌模板,重复上述过程。
通常情况下,采用上述方法进行双层车牌识别时,能够识别出待识别车牌图像中车牌的车牌号码,但是,由于需要匹配大量双层车牌模板,且每次匹配过程都要完整执行一遍车牌区域定位、车牌上下层字符的分割和识别过程,其车牌识别过程效率较低。
发明内容
本申请实施例的目的在于提供了一种车牌识别方法及装置,能够提高车牌识别过程的效率。
为了达到上述目的,本申请公开了一种车牌识别方法,所述方法包括:
获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
识别所述第一车牌区域中的字符,获得第一字符识别结果;
判断所述第一字符识别结果中的所有字符是否位于同一行内;
如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
识别所述第二车牌区域中的字符,获得第二字符识别结果;
合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
可选的,车牌区域包括:上层车牌区域和下层车牌区域,所述双层车牌字符分布特征包括:上层车牌区域对应的上层字符特征,下层车牌区域对应的下层字符特征,以及上层车牌区域与下层车牌区域的相对位置关系;
所述根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域的步骤,包括:
根据所述第一字符识别结果中的字符特征、所述上层字符特征和所述下层字符特征,匹配所述第一车牌区域所属的车牌区域;
根据匹配结果、所述第一车牌区域以及所述相对位置关系,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
可选的,所述根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域的步骤,包括:
根据所述第一字符识别结果,修正所述第一车牌区域;
根据修正后的车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
可选的,所述识别所述第二车牌区域中的字符,获得第二字符识别结果的步骤,包括:
分割所述第二车牌区域,获得目标字符区域;
识别所述目标字符区域中的字符,获得第二字符识别结果。
可选的,所述分割所述第二车牌区域,获得目标字符区域的步骤,包括:
根据所述第一字符识别结果中各个字符区域的尺寸,获得第一字符尺寸;
根据所述第一字符尺寸以及预设的尺寸对应关系,确定第二字符尺寸,其中,所述尺寸对应关系为:上层车牌区域字符的尺寸和下层车牌区域字符 的尺寸之间的对应关系;
根据所述第二字符尺寸,分割所述第二车牌区域,获得目标字符区域。
可选的,所述判断所述第一字符识别结果中的所有字符是否位于同一行内的步骤,包括:
判断所述第一字符识别结果中的所有字符在车牌区域中的位置分布是否为单层分布,如果是,则确定所述第一字符识别结果中的所有字符位于同一行内。
为了达到上述目的,本申请公开了一种车牌识别装置,所述装置包括:
第一区域确定模块,用于获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
第一结果识别模块,用于识别所述第一车牌区域中的字符,获得第一字符识别结果;
识别结果判断模块,用于判断所述第一字符识别结果中的所有字符是否位于同一行内;
第二区域确定模块,用于当所述第一字符识别结果中的所有字符位于同一行内时,根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
第二结果识别模块,用于识别所述第二车牌区域中的字符,获得第二字符识别结果;
识别结果合成模块,用于合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
可选的,车牌区域包括:上层车牌区域和下层车牌区域,所述双层车牌字符分布特征包括:上层车牌区域对应的上层字符特征,下层车牌区域对应的下层字符特征,以及上层车牌区域与下层车牌区域的相对位置关系;
所述第二区域确定模块,包括:
匹配子模块,用于根据所述第一字符识别结果中的字符特征、所述上层字符特征和所述下层字符特征,匹配所述第一车牌区域所属的车牌区域;
第一确定子模块,用于根据匹配结果、所述第一车牌区域以及所述相对位置关系,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
可选的,所述第二区域确定模块,包括:
修正子模块,用于根据所述第一字符识别结果,修正所述第一车牌区域;
第二确定子模块,用于根据修正后的车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
可选的,所述第二结果识别模块,包括:
分割子模块,用于分割所述第二车牌区域,获得目标字符区域;
识别子模块,用于识别所述目标字符区域中的字符,获得第二字符识别结果。
可选的,所述分割子模块,包括:
获得单元,根据所述第一字符识别结果中各个字符区域的尺寸,获得第一字符尺寸;
确定单元,用于根据所述第一字符尺寸以及预设的尺寸对应关系,确定第二字符尺寸,其中,所述尺寸对应关系为:上层车牌区域字符的尺寸和下层车牌区域字符的尺寸之间的对应关系;
分割单元,用于根据所述第二字符尺寸,分割所述第二车牌区域,获得目标字符区域。
可选的,所述识别结果判断模块,具体用于:
判断所述第一字符识别结果中的所有字符在车牌区域中的位置分布是否为单层分布,如果是,则确定所述第一字符识别结果中的所有字符位于同一行内。
为了达到上述目的,本申请公开了一种电子设备,适用于车牌识别,所述电子设备包括:
壳体、处理器、存储器、电路板和电源电路,其中,电路板安置在壳体围成的空间内部,处理器和存储器设置在电路板上;电源电路,用于为电子设备的各个电路或器件供电;存储器用于存储可执行程序代码;处理器通过读取存储器中存储的可执行程序代码来运行与可执行程序代码对应的程序,以用于执行以下步骤:
获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
识别所述第一车牌区域中的字符,获得第一字符识别结果;
判断所述第一字符识别结果中的所有字符是否位于同一行内;
如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
识别所述第二车牌区域中的字符,获得第二字符识别结果;
合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
为了达到上述目的,本申请公开了一种应用程序,所述应用程序用于在运行时执行本申请实施例提供的车牌识别方法。其中,该车牌识别方法包括:
获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
识别所述第一车牌区域中的字符,获得第一字符识别结果;
判断所述第一字符识别结果中的所有字符是否位于同一行内;
如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
识别所述第二车牌区域中的字符,获得第二字符识别结果;
合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
为了达到上述目的,本申请公开了一种存储介质,其特征在于,用于存储可执行代码,所述可执行代码在运行时用于执行本申请实施例提供的车牌识别方法。其中,该车牌识别方法包括:
获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
识别所述第一车牌区域中的字符,获得第一字符识别结果;
判断所述第一字符识别结果中的所有字符是否位于同一行内;
如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
识别所述第二车牌区域中的字符,获得第二字符识别结果;
合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
由上述技术方案可见,本申请实施例中,首先确定所获得的待识别车牌图像中的第一车牌区域,识别所述第一车牌区域中的字符,获得第一字符识别结果,然后判断第一字符识别结果是否为针对双层车牌的识别结果;如果否,则根据第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域,并识别所述第二车牌区域中的字符,获得第二字符识别结果。最后,合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
也就是说,本申请实施例,根据待识别车牌图像中第一车牌区域和对应 的第一字符识别结果,以及预先保存的双层车牌字符分布特征,确定待识别车牌图像中未识别出车牌号码的第二车牌区域,无需匹配大量双层车牌模板,因此能够提高车牌识别过程的效率。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单的介绍。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他的附图。
图1为部分双层车牌图像实例图;
图2为本申请实施例提供的车牌识别方法的一种流程示意图;
图3为本申请实施例提供的车牌识别方法的另一种流程示意图;
图4为本申请实施例提供的车牌识别装置的一种结构示意图;
图5为本申请实施例提供的车牌识别装置的另一种结构示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整的描述。显然,所描述的实施例仅仅是本申请的一部分实施例,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做出创造性劳动的前提下所获得的所有其他实施例,都属于本申请保护的范围。
本申请实施例提供了一种车牌识别方法及装置,应用于电子设备,该电子设备包括计算机、平板电脑、智能手机等设备,能够提高车牌识别过程的效率。
下面通过具体实施例,对本申请进行详细说明。
图2为本申请实施例提供的车牌识别方法的一种流程示意图,应用于电子设备,该方法包括如下步骤:
步骤S201:获得待识别车牌图像,并确定待识别车牌图像中的第一车牌区域。
其中,待识别车牌图像可以是道路上抓拍的车辆图像。当然,待识别车牌图像可以是任意一种包含车牌号码的图像,本申请实施例对此不做具体限定。待识别车牌图像中的车牌可能是双层车牌,也可能是单层车牌。双层车 牌,是指车牌号码分布于上下两行的车牌,即包括上层车牌区域和下层车牌区域。图1所示的车牌均属于双层车牌。单层车牌,是指车牌号码分布于一行的车牌。
作为执行主体的电子设备,在获得待识别车牌图像之后,确定待识别车牌图像中的第一车牌区域时,可以根据预设的车牌区域的纹理特征,确定待识别车牌图像中的第一车牌区域,具体过程不再赘述。
可以理解的是,第一车牌区域可能同时包含上层车牌区域和下层车牌区域,也可能只包含上层车牌区域或下层车牌区域。
步骤S202:识别第一车牌区域中的字符,获得第一字符识别结果。
其中,第一字符识别结果包括字符及对应的字符类型、字符区域等,字符区域即为该字符在图像中对应的区域。
具体的,识别第一车牌区域中的字符,获得第一字符识别结果时,可以首先采用垂直投影法和/或连通域法对第一车牌区域进行分割,获得字符分割结果,然后采用预设的字符识别器对字符分割结果进行识别,获得第一字符识别结果。
步骤S203:判断第一字符识别结果中的所有字符是否位于同一行内,如果是,则说明第一字符识别结果不是一种双层车牌的识别结果,即说明第一车牌区域不是双层车牌区域,可以继续执行步骤S204。
如果否,则说明第一字符识别结果是针对双层车牌的识别结果,即第一车牌区域属于双层车牌区域,也就是说待识别车牌图像中不存在待识别车牌号码的第二车牌区域。
需要说明的是,当上述判断结果为是时,能说明第一车牌区域不是双层车牌区域,但是不能确定待识别车牌图像中的车牌属于双层车牌还是单层车牌,只能确定第一字符识别结果不为针对双层车牌的识别结果。
可以理解的是,步骤S203的作用在于,去除第一字符识别结果已经是双层车牌识别结果的情况,这种情况下不需要再继续识别第二车牌区域了。
步骤S204:根据第一车牌区域和预先保存的双层车牌字符分布特征,确定待识别车牌图像中未识别出车牌号码的第二车牌区域。
其中,双层车牌字符分布特征,可以包括上层车牌区域的字符特征和下层车牌区域的字符特征,上层车牌区域的颜色对应关系和下层车牌区域的颜色对应关系,以及上层车牌区域与下层车牌区域的相对位置关系。其中,颜 色对应关系为前景色与背景色的颜色对应关系,颜色对应关系具体包括:前景为黑色、背景为白色(黑字白底)和前景为白色、背景为黑色(白字黑底)两种;字符特征可以包括字符类型及对应的数量等。
例如,对于图1所示双层车牌图像中编号为1~11的车牌,其具有的双层车牌字符分布特征可以包括如下内容:
上层车牌区域的字符特征,包括:字符类型包括字母类型和数字类型,字母类型的字符数量为2~3,数字类型的字符数量为0~2。
下层车牌区域的字符特征,包括:字符类型包括字母类型和数字类型,字母类型的字符数量为0~1,数字类型的字符数量为2~4。
上层车牌区域的颜色对应关系和下层车牌区域的颜色对应关系一致,均为:白字黑底。
上层车牌区域与下层车牌区域的相对位置关系包括:上层车牌区域位于下层车牌区域正上方,且与下层车牌区域间隔N个像素的指定范围内。
需要说明的是,上述双层车牌字符分布特征可以是预先保存的。由于各个地区之间车牌特征差异较大,因此可以针对同一个地区的双层车牌获得双层车牌字符分布特征。
例如,在图1所示双层车牌图像中,编号为1~11的双层车牌图像属于一个地区,编号为12~20的双层车牌图像属于另一个地区,可以针对这两个地区的车牌分别获得双层车牌字符分布特征:
具体的,可以采集指定地区的双层车牌图像样本,标记样本中上层车牌字符和下层车牌字符,根据上述样本和样本中的标记,提取并保存该地区对应的双层字符分布特征。
步骤S205:识别第二车牌区域中的字符,获得第二字符识别结果。
由于步骤S203中的判断结果为是时,只能确定第一字符识别结果中的所有字符位于同一行内,并不能确定待识别车牌图像中的车牌属于双层车牌还是单层车牌,因此,在识别第二车牌区域中的字符时,有可能存在识别结果不成功的情况。
因此,在本实施例的一种实施方式中,为了提高车牌识别的准确性,在步骤S205之后,还可以包括:根据第二字符识别结果,判断针对第二车牌区域的字符识别是否成功,如果识别成功,则执行步骤S206。如果未识别成功,则不予处理。
具体的,当针对第二车牌区域进行字符识别时,可以判断是否能识别出置信度高于预设阈值的字符,如果是,则可以确定针对第二车牌区域的字符识别成功。
需要说明的是,识别第二车牌区域中的字符时,可以采用与步骤S202相同的过程,也可以采用与步骤S202不同的过程,其详细过程本实施例不再赘述。
步骤S206:合成第一字符识别结果和第二字符识别结果,获得待识别车牌图像的车牌号码。
具体的,合成第一字符识别结果和第二字符识别结果时,可以根据第一车牌区域和第二车牌区域的相对位置关系,合成第一字符识别结果和第二字符识别结果,从而获得待识别车牌图像的车牌号码。
例如,当确定第一车牌区域和第二车牌区域的相对位置关系为第一车牌区域位于上层车牌区域,第二车牌区域位于下层车牌区域时,将第一字符识别结果置于第二字符识别结果的左侧,即可获得待识别车牌图像的车牌号码。
需要说明的是,当第一字符识别结果和/或第二字符识别结果中存在未识别成功的字符识别结果(即置信度比较低的字符识别结果)时,在合成第一字符识别结果和第二字符识别结果时,可以将第一字符识别结果中已识别成功的字符识别结果和第二字符识别结果中已识别成功的字符识别结果进行合成,获得待识别车牌图像的车牌号码。
由上述内容可知,本实施例中,首先确定所获得的待识别车牌图像中的第一车牌区域,识别所述第一车牌区域中的字符,获得第一字符识别结果,然后判断第一字符识别结果是否为针对双层车牌的识别结果;如果否,则根据第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域,并识别所述第二车牌区域中的字符,获得第二字符识别结果。最后,合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
也就是说,本实施例,根据待识别车牌图像中第一车牌区域和对应的第一字符识别结果,以及预先保存的双层车牌字符分布特征,确定待识别车牌图像中未识别出车牌号码的第二车牌区域,最终识别出整个车牌号码,无需匹配大量双层车牌模板,因此能够提高车牌识别过程的效率。
同时,不管第一车牌区域属于上层车牌区域还是下层车牌区域,本实施 例都能定位出待识别车牌图像中未识别出车牌号码的第二车牌区域,并最终识别出完整的双层车牌。
基于图2所示实施例的另一实施方式中,步骤S203,所述判断所述第一字符识别结果中的所有字符是否位于同一行内,具体可以包括:
判断所述第一字符识别结果中的所有字符在车牌区域中的位置分布是否为单层分布,如果是,则确定所述第一字符识别结果中的所有字符位于同一行内。
其中,第一字符识别结果中包含每个字符区域,根据这些字符区域能够确定这些字符是否位于同一行内。
基于图2所示实施例的另一实施方式中,为了提高检测的准确性,步骤S204,根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域,可以包括:
根据所述第一字符识别结果,修正所述第一车牌区域;根据修正后的车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
可以理解的是,根据第一字符识别结果修正第一车牌区域,可以去除干扰因素的影响,使第一车牌区域的范围更加准确,提高检测的准确性和可靠性。
具体的,根据所述第一字符识别结果,修正所述第一车牌区域时,可以包括:根据第一字符识别结果,确定第一车牌区域中字符识别成功的字符区域,根据字符识别成功的字符区域,确定修正后的车牌区域。
例如,第一车牌区域中最左侧存在铆钉区域,而根据第一字符识别结果可以知道该铆钉区域的字符识别结果为未识别成功,则可以从第一车牌区域中去除该铆钉区域。
图3为本申请实施例提供的车牌识别方法的另一流程示意图,该实施例为图2所示实施例的改进。其中,车牌区域包括:上层车牌区域和下层车牌区域,所述双层车牌字符分布特征包括:上层车牌区域对应的上层字符特征,下层车牌区域对应的下层字符特征,以及上层车牌区域与下层车牌区域的相对位置关系。
例如,针对图1中编号为1~11的车牌,上层字符特征可以包括:字符类型为“字母+数字”类型,总字符数量为1~4,字母数量为1~3,数字数量为 0~2;下层字符特征可以包括:字符类型为“字母+数字”类型,总字符数量为2~5,字母数量为0~1,数字数量为2~4;相对位置关系包括:上层车牌区域与下层车牌区域的对称中心重合。
具体的,图2所示实施例中的步骤S204,根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域,具体可以包括:
步骤S204A:根据第一字符识别结果中的字符特征、上层字符特征和下层字符特征,匹配第一车牌区域所属的车牌区域。
仍旧以图1中编号为1~11的车牌为例,上层字符特征和下层字符特征已经在上述内容中列出。如果第一字符识别结果为641,由于该结果中数字数量为3,不符合上层字符特征中“数字数量为0~2”的特征,因此可以确定该结果符合下层字符特征,匹配得到第一车牌区域属于下层车牌区域。如果第一字符识别结果为1M4U,由于该结果中数字数量和字母数量均为2,不符合下层字符特征中“字母数量为0~1”的特征,因此可以确定该结果符合上层字符特征,匹配得到第一车牌区域属于上层车牌区域。
步骤S204B:根据匹配结果、第一车牌区域以及所述相对位置关系,确定待识别车牌图像中未识别出车牌号码的第二车牌区域。
当匹配得到第一车牌区域属于下层车牌区域时,可以确定第二车牌区域属于上层车牌区域,在确定第二车牌区域时,可以将第一车牌区域上方的预设范围确定为第二车牌区域。
当匹配得到第一车牌区域属于上层车牌区域时,可以确定第二车牌区域属于下层车牌区域,在确定第二车牌区域时,可以将第一车牌区域下方的预设范围确定为第二车牌区域。
其中,预设范围可以采用以下方式确定:根据第一字符识别结果确定第一车牌区域中单个字符的范围,将单个字符的范围与预设值的乘积确定为预设范围。
具体的,可以将单个字符的宽度范围与第一预设值的乘积确定为预设范围的宽度,将单个字符的高度范围与第二预设值的乘积确定为预设范围的高度。第一预设值可以根据大量样本车牌单层字符的数量确定,例如取为6~8,第二预设值可以根据大量样本车牌单层字符的高度确定,例如取为1~1.5。
例如,根据第一字符识别结果确定第一车牌区域中单个字符的宽度为5 个像素,高度为8个像素,第一预设值取为6,则可以确定预设范围的宽度为5个像素*6=30个像素;第二预设值可以取为1.5,则可以确定预设范围的高度范围为8个像素*1.5=12个像素。对应的,可以将第一车牌区域正下方30个像素宽度、12个像素高度的区域确定为第二车牌区域。
综上所述,在本实施例中,根据第一字符识别结果和双层车牌字符分布特征,确定第二车牌区域,无需从整个待识别车牌图像中的无方向性地寻找第二车牌区域,可以减少待识别车牌图像中其他内容的干扰,能够更准确、更快速地定位第二车牌区域。
为了进一步提高检测的准确性,在本实施例中,步骤S204B,根据匹配结果、所述第一车牌区域以及所述相对位置关系,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域,具体可以包括:
根据所述第一字符识别结果,修正所述第一车牌区域;根据修正后的车牌区域和匹配结果以及所述相对位置关系,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
具体的,在修正第一车牌区域时,可以根据第一字符识别结果中识别成功的字符对应的字符区域,重新卡位第一车牌区域的左右边界。
可以理解的是,根据第一字符识别结果修正第一车牌区域,可以提高第一车牌区域的准确性,进而提高整个车牌识别过程的准确性。
在图3所示实施例中,图2所示实施例中的步骤S205,识别所述第二车牌区域中的字符,获得第二字符识别结果的步骤,具体可以包括:
步骤S205A:分割所述第二车牌区域,获得目标字符区域。
在分割第二车牌区域时,可以根据垂直投影法和/或连通域法分割第二车牌区域。
作为一种具体实施方式,在分割第二车牌区域时,还可以根据分割结果判断针对第二车牌区域的分割过程是否成功,如果分割成功,则执行步骤S205B,如果分割不成功,则不予处理。
步骤S205B:识别所述目标字符区域中的字符,获得第二字符识别结果。
在识别目标字符区域中的字符时,可以根据预设的字符分类器识别目标字符区域的字符。
基于图3所示实施例的另一实施方式中,为了提高字符分割过程的准确性,步骤S205A,分割所述第二车牌区域,获得目标字符区域,可以包括:
步骤1:根据所述第一字符识别结果中各个字符区域的尺寸,获得第一字符尺寸。其中,该尺寸可以为宽度和高度中的至少一个。
可以理解的是,通常,车牌中同一层车牌区域中的各个字符的尺寸是基本相同的。第一字符尺寸是能够代表第一字符识别结果中各个字符区域大小的一个尺寸值。
步骤2:根据所述第一字符尺寸以及预设的尺寸对应关系,确定第二字符尺寸,其中,所述尺寸对应关系为:上层车牌区域字符的尺寸和下层车牌区域字符的尺寸之间的对应关系。
例如,上述尺寸对应关系可以为上层车牌区域字符的宽度等于下层车牌区域字符的宽度。
步骤3:根据所述第二字符尺寸,分割所述第二车牌区域,获得目标字符区域。
具体的,在分割第二车牌区域时,可以首先采用垂直投影法和/或连通域法对第二车牌区域进行第一次分割,然后在第一次分割的基础上,根据第二字符尺寸,修正第一次分割的结果。
综上所述,根据第一字符识别结果以及上层车牌区域字符的尺寸和下层车牌区域字符的尺寸之间的对应关系,分割第二车牌区域的字符,能够提高字符分割结果的准确性。
图4为本申请实施例提供的车牌识别装置的一种流程示意图,与图2所示实施例相对应,应用于电子设备,所述装置包括:
第一区域确定模块401,用于获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
第一结果识别模块402,用于识别所述第一车牌区域中的字符,获得第一字符识别结果;
识别结果判断模块403,用于判断所述第一字符识别结果中的所有字符是否位于同一行内;
第二区域确定模块404,用于当所述第一字符识别结果中的所有字符位于同一行内时,根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
第二结果识别模块405,用于识别所述第二车牌区域中的字符,获得第二字符识别结果;
识别结果合成模块406,用于合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
基于图4所示实施例的另一实施方式中,所述识别结果判断模块403,具体可以用于:
判断所述第一字符识别结果中的所有字符在车牌区域中的位置分布是否为单层分布,如果是,则确定所述第一字符识别结果中的所有字符位于同一行内。
基于图4所示实施例的另一实施方式中,第二区域确定模块404,具体可以包括:
修正子模块(图中未示出),用于根据所述第一字符识别结果,修正所述第一车牌区域;
第二确定子模块(图中未示出),用于根据修正后的车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
图5为本申请实施例提供的车牌识别方法的另一结构示意图,该实施例为基于图4所示实施例的改进,未改进之处与图4所示实施例相同。该实施例与图3所示方法实施例相对应。本实施例中,车牌区域包括:上层车牌区域和下层车牌区域,所述双层车牌字符分布特征包括:上层车牌区域对应的上层字符特征,下层车牌区域对应的下层字符特征,以及上层车牌区域与下层车牌区域的相对位置关系。
在图5所示实施例中,第二区域确定模块404,具体可以包括:
匹配子模块501,用于根据所述第一字符识别结果中的字符特征、所述上层字符特征和所述下层字符特征,匹配所述第一车牌区域所属的车牌区域;
第一确定子模块502,用于根据匹配结果、所述第一车牌区域以及所述相对位置关系,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
在图5所示实施例中,第二结果识别模块405,具体可以包括:
分割子模块503,用于分割所述第二车牌区域,获得目标字符区域;
识别子模块504,用于识别所述目标字符区域中的字符,获得第二字符识别结果。
基于图5所示实施例的另一实施方式中,所述分割子模块503,具体可以包括:
获得单元(图中未示出),根据所述第一字符识别结果中各个字符区域的尺寸,获得第一字符尺寸;
确定单元(图中未示出),用于根据所述第一字符尺寸以及预设的尺寸对应关系,确定第二字符尺寸,其中,所述尺寸对应关系为:上层车牌区域字符的尺寸和下层车牌区域字符的尺寸之间的对应关系;
分割单元(图中未示出),用于根据所述第二字符尺寸,分割所述第二车牌区域,获得目标字符区域。
由于上述装置实施例是基于方法实施例得到的,与该方法具有相同的技术效果,因此装置实施例的技术效果在此不再赘述。
对于装置实施例而言,由于其基本相似于方法实施例,所以描述得比较简单,相关之处参见方法实施例的部分说明即可。
本申请实施例提供了一种电子设备,适用于车牌识别,所述电子设备包括:
壳体、处理器、存储器、电路板和电源电路,其中,电路板安置在壳体围成的空间内部,处理器和存储器设置在电路板上;电源电路,用于为电子设备的各个电路或器件供电;存储器用于存储可执行程序代码;处理器通过读取存储器中存储的可执行程序代码来运行与可执行程序代码对应的程序,以用于执行以下步骤:
获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
识别所述第一车牌区域中的字符,获得第一字符识别结果;
判断所述第一字符识别结果中的所有字符是否位于同一行内;
如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
识别所述第二车牌区域中的字符,获得第二字符识别结果;
合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
其中,该电子设备可以以多种形式存在,包括但不限于:
(1)移动通信设备:这类设备的特点是具备移动通信功能,并且以提供话音、数据通信为主要目标。这类终端包括:智能手机(例如iPhone)、多媒体手机、功能性手机以及低端手机等。
(2)超移动个人计算机设备:这类设备属于个人计算机的范畴,有计算 和处理功能,一般也具备移动上网特性。这类终端包括:PDA、MID和UMPC设备等,例如iPad。
(3)便携式娱乐设备:这类设备可以显示和播放多媒体内容。该类设备包括:音频、视频播放器(例如iPod),掌上游戏机,电子书,以及智能玩具和便携式车载导航设备。
(4)服务器:提供计算服务的设备,服务器的构成包括处理器、硬盘、内存、系统总线等,服务器和通用的计算机架构类似,但是由于需要提供高可靠的服务,因此在处理能力、稳定性、可靠性、安全性、可扩展性、可管理性等方面要求较高。
(5)其他具有数据交互功能的电子装置。
可见,本申请实施例可以根据待识别车牌图像中第一车牌区域和对应的第一字符识别结果,以及预先保存的双层车牌字符分布特征,确定待识别车牌图像中未识别出车牌号码的第二车牌区域,无需匹配大量双层车牌模板,因此能够提高车牌识别过程的效率。
本申请实施例提供了一种应用程序,所述应用程序用于在运行时执行本申请实施例提供的车牌识别方法。其中,该车牌识别方法包括:
获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
识别所述第一车牌区域中的字符,获得第一字符识别结果;
判断所述第一字符识别结果中的所有字符是否位于同一行内;
如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
识别所述第二车牌区域中的字符,获得第二字符识别结果;
合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
可见,本申请实施例可以根据待识别车牌图像中第一车牌区域和对应的第一字符识别结果,以及预先保存的双层车牌字符分布特征,确定待识别车牌图像中未识别出车牌号码的第二车牌区域,无需匹配大量双层车牌模板,因此能够提高车牌识别过程的效率。
本申请提供了一种存储介质,用于存储可执行代码,所述可执行代码在运行时用于执行本申请实施例提供的车牌识别方法。其中,该车牌识别方法包括:
获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
识别所述第一车牌区域中的字符,获得第一字符识别结果;
判断所述第一字符识别结果中的所有字符是否位于同一行内;
如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
识别所述第二车牌区域中的字符,获得第二字符识别结果;
合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
可见,本申请实施例可以根据待识别车牌图像中第一车牌区域和对应的第一字符识别结果,以及预先保存的双层车牌字符分布特征,确定待识别车牌图像中未识别出车牌号码的第二车牌区域,无需匹配大量双层车牌模板,因此能够提高车牌识别过程的效率。
需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
本领域普通技术人员可以理解,上述实施方式中的全部或部分步骤是能够通过程序指令相关的硬件来完成的,所述的程序可以存储于计算机可读取存储介质中。这里所称存储介质,是指ROM/RAM、磁碟、光盘等。
以上所述仅为本申请的较佳实施例而已,并非用于限定本申请的保护范围。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,均包含在本申请的保护范围内。

Claims (15)

  1. 一种车牌识别方法,其特征在于,所述方法包括:
    获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
    识别所述第一车牌区域中的字符,获得第一字符识别结果;
    判断所述第一字符识别结果中的所有字符是否位于同一行内;
    如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
    识别所述第二车牌区域中的字符,获得第二字符识别结果;
    合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
  2. 根据权利要求1所述的方法,其特征在于,车牌区域包括:上层车牌区域和下层车牌区域,所述双层车牌字符分布特征包括:上层车牌区域对应的上层字符特征,下层车牌区域对应的下层字符特征,以及上层车牌区域与下层车牌区域的相对位置关系;
    所述根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域的步骤,包括:
    根据所述第一字符识别结果中的字符特征、所述上层字符特征和所述下层字符特征,匹配所述第一车牌区域所属的车牌区域;
    根据匹配结果、所述第一车牌区域以及所述相对位置关系,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
  3. 根据权利要求1所述的方法,其特征在于,所述根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域的步骤,包括:
    根据所述第一字符识别结果,修正所述第一车牌区域;
    根据修正后的车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
  4. 根据权利要求1所述的方法,其特征在于,所述识别所述第二车牌区域中的字符,获得第二字符识别结果的步骤,包括:
    分割所述第二车牌区域,获得目标字符区域;
    识别所述目标字符区域中的字符,获得第二字符识别结果。
  5. 根据权利要求4所述的方法,其特征在于,所述分割所述第二车牌区 域,获得目标字符区域的步骤,包括:
    根据所述第一字符识别结果中各个字符区域的尺寸,获得第一字符尺寸;
    根据所述第一字符尺寸以及预设的尺寸对应关系,确定第二字符尺寸,其中,所述尺寸对应关系为:上层车牌区域字符的尺寸和下层车牌区域字符的尺寸之间的对应关系;
    根据所述第二字符尺寸,分割所述第二车牌区域,获得目标字符区域。
  6. 根据权利要求1所述的方法,其特征在于,所述判断所述第一字符识别结果中的所有字符是否位于同一行内的步骤,包括:
    判断所述第一字符识别结果中的所有字符在车牌区域中的位置分布是否为单层分布,如果是,则确定所述第一字符识别结果中的所有字符位于同一行内。
  7. 一种车牌识别装置,其特征在于,所述装置包括:
    第一区域确定模块,用于获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
    第一结果识别模块,用于识别所述第一车牌区域中的字符,获得第一字符识别结果;
    识别结果判断模块,用于判断所述第一字符识别结果中的所有字符是否位于同一行内;
    第二区域确定模块,用于当所述第一字符识别结果中的所有字符位于同一行内时,根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
    第二结果识别模块,用于识别所述第二车牌区域中的字符,获得第二字符识别结果;
    识别结果合成模块,用于合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
  8. 根据权利要求7所述的装置,其特征在于,车牌区域包括:上层车牌区域和下层车牌区域,所述双层车牌字符分布特征包括:上层车牌区域对应的上层字符特征,下层车牌区域对应的下层字符特征,以及上层车牌区域与下层车牌区域的相对位置关系;
    所述第二区域确定模块,包括:
    匹配子模块,用于根据所述第一字符识别结果中的字符特征、所述上层 字符特征和所述下层字符特征,匹配所述第一车牌区域所属的车牌区域;
    第一确定子模块,用于根据匹配结果、所述第一车牌区域以及所述相对位置关系,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
  9. 根据权利要求7所述的装置,其特征在于,所述第二区域确定模块,包括:
    修正子模块,用于根据所述第一字符识别结果,修正所述第一车牌区域;
    第二确定子模块,用于根据修正后的车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域。
  10. 根据权利要求7所述的装置,其特征在于,所述第二结果识别模块,包括:
    分割子模块,用于分割所述第二车牌区域,获得目标字符区域;
    识别子模块,用于识别所述目标字符区域中的字符,获得第二字符识别结果。
  11. 根据权利要求10所述的装置,其特征在于,所述分割子模块,包括:
    获得单元,根据所述第一字符识别结果中各个字符区域的尺寸,获得第一字符尺寸;
    确定单元,用于根据所述第一字符尺寸以及预设的尺寸对应关系,确定第二字符尺寸,其中,所述尺寸对应关系为:上层车牌区域字符的尺寸和下层车牌区域字符的尺寸之间的对应关系;
    分割单元,用于根据所述第二字符尺寸,分割所述第二车牌区域,获得目标字符区域。
  12. 根据权利要求7所述的装置,其特征在于,所述识别结果判断模块,具体用于:
    判断所述第一字符识别结果中的所有字符在车牌区域中的位置分布是否为单层分布,如果是,则确定所述第一字符识别结果中的所有字符位于同一行内。
  13. 一种电子设备,其特征在于,适用于车牌识别,所述电子设备包括:
    壳体、处理器、存储器、电路板和电源电路,其中,电路板安置在壳体围成的空间内部,处理器和存储器设置在电路板上;电源电路,用于为电子设备的各个电路或器件供电;存储器用于存储可执行程序代码;处理器通过 读取存储器中存储的可执行程序代码来运行与可执行程序代码对应的程序,以用于执行以下步骤:
    获得待识别车牌图像,并确定所述待识别车牌图像中的第一车牌区域;
    识别所述第一车牌区域中的字符,获得第一字符识别结果;
    判断所述第一字符识别结果中的所有字符是否位于同一行内;
    如果是,则根据所述第一车牌区域和预先保存的双层车牌字符分布特征,确定所述待识别车牌图像中未识别出车牌号码的第二车牌区域;
    识别所述第二车牌区域中的字符,获得第二字符识别结果;
    合成所述第一字符识别结果和第二字符识别结果,获得所述待识别车牌图像的车牌号码。
  14. 一种应用程序,其特征在于,所述应用程序用于在运行时执行权利要求1-6任一项所述的车牌识别方法。
  15. 一种存储介质,其特征在于,用于存储可执行代码,所述可执行代码在运行时用于执行权利要求1-6任一项所述的车牌识别方法。
PCT/CN2017/106022 2016-11-16 2017-10-13 一种车牌识别方法及装置 WO2018090771A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611032450.4A CN108073928B (zh) 2016-11-16 2016-11-16 一种车牌识别方法及装置
CN201611032450.4 2016-11-16

Publications (1)

Publication Number Publication Date
WO2018090771A1 true WO2018090771A1 (zh) 2018-05-24

Family

ID=62145992

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/106022 WO2018090771A1 (zh) 2016-11-16 2017-10-13 一种车牌识别方法及装置

Country Status (2)

Country Link
CN (1) CN108073928B (zh)
WO (1) WO2018090771A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109711407A (zh) * 2018-12-28 2019-05-03 深圳市捷顺科技实业股份有限公司 一种车牌识别的方法及相关装置
CN111027534A (zh) * 2018-10-09 2020-04-17 杭州海康威视数字技术股份有限公司 一种紧凑双车牌检测方法及装置
CN111046891A (zh) * 2018-10-11 2020-04-21 杭州海康威视数字技术股份有限公司 车牌识别模型的训练方法、车牌识别方法和装置
CN111275796A (zh) * 2020-01-17 2020-06-12 北京迈格威科技有限公司 车牌合成方法、装置、计算机设备和存储介质
CN111340045A (zh) * 2020-02-12 2020-06-26 浙江大华技术股份有限公司 车牌号码的识别方法、装置及存储介质
CN111639636A (zh) * 2020-05-29 2020-09-08 北京奇艺世纪科技有限公司 一种字符识别方法及装置
CN111695563A (zh) * 2020-06-10 2020-09-22 北京筑梦园科技有限公司 一种单双层车牌识别方法、服务器及停车收费系统
CN111950659A (zh) * 2020-09-01 2020-11-17 湖南国科微电子股份有限公司 双层车牌图像处理方法、装置、电子设备及存储介质
CN115131776A (zh) * 2022-04-12 2022-09-30 公安部交通管理科学研究所 一种融合新能源汽车特征的车牌图像字符分类识别方法
WO2023169273A1 (zh) * 2022-03-11 2023-09-14 阿里巴巴(中国)有限公司 图像识别方法、装置、设备和存储介质

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110543882A (zh) * 2018-05-29 2019-12-06 北京深鉴智能科技有限公司 一种字符串识别方法与装置
CN110728278A (zh) * 2018-07-16 2020-01-24 杭州海康威视数字技术股份有限公司 车牌识别方法及装置
CN110728276B (zh) * 2018-07-16 2022-12-06 杭州海康威视数字技术股份有限公司 一种车牌识别方法及装置
CN109447117B (zh) * 2018-09-25 2023-06-30 平安科技(深圳)有限公司 双层车牌识别方法、装置、计算机设备及存储介质
CN109726678B (zh) * 2018-12-28 2023-02-28 深圳市捷顺科技实业股份有限公司 一种车牌识别的方法及相关装置
CN110070085B (zh) * 2019-04-30 2021-11-02 北京百度网讯科技有限公司 车牌识别方法和装置
CN111353500B (zh) * 2020-02-25 2023-07-28 上海其高电子科技有限公司 双行车牌的自动识别方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7182492B1 (en) * 2003-12-22 2007-02-27 Robert Louis Walter License plate system having enhanced illumination
CN103226696A (zh) * 2013-04-07 2013-07-31 布法罗机器人科技(苏州)有限公司 车牌的识别系统及方法
CN103413147A (zh) * 2013-08-28 2013-11-27 庄浩洋 一种车牌识别方法和系统
CN105528605A (zh) * 2014-09-28 2016-04-27 江苏省兴泽实业发展有限公司 一种基于投影与识别的双层车牌字符分割方法
CN105631470A (zh) * 2015-12-21 2016-06-01 深圳市捷顺科技实业股份有限公司 一种车牌类型验证的方法及系统

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103400133A (zh) * 2013-08-06 2013-11-20 河海大学 基于投影和连通域法的双行结构车牌字符分割方法
CN105894004A (zh) * 2014-12-16 2016-08-24 中防通用电信技术有限公司 一种针对车牌识别系统中定位不全车牌的补全技术
US20160203379A1 (en) * 2015-01-12 2016-07-14 TigerIT Americas, LLC Systems, methods and devices for the automated verification and quality control and assurance of vehicle identification plates

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7182492B1 (en) * 2003-12-22 2007-02-27 Robert Louis Walter License plate system having enhanced illumination
CN103226696A (zh) * 2013-04-07 2013-07-31 布法罗机器人科技(苏州)有限公司 车牌的识别系统及方法
CN103413147A (zh) * 2013-08-28 2013-11-27 庄浩洋 一种车牌识别方法和系统
CN105528605A (zh) * 2014-09-28 2016-04-27 江苏省兴泽实业发展有限公司 一种基于投影与识别的双层车牌字符分割方法
CN105631470A (zh) * 2015-12-21 2016-06-01 深圳市捷顺科技实业股份有限公司 一种车牌类型验证的方法及系统

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111027534A (zh) * 2018-10-09 2020-04-17 杭州海康威视数字技术股份有限公司 一种紧凑双车牌检测方法及装置
CN111027534B (zh) * 2018-10-09 2023-05-26 杭州海康威视数字技术股份有限公司 一种紧凑双车牌检测方法及装置
CN111046891A (zh) * 2018-10-11 2020-04-21 杭州海康威视数字技术股份有限公司 车牌识别模型的训练方法、车牌识别方法和装置
CN109711407B (zh) * 2018-12-28 2023-02-28 深圳市捷顺科技实业股份有限公司 一种车牌识别的方法及相关装置
CN109711407A (zh) * 2018-12-28 2019-05-03 深圳市捷顺科技实业股份有限公司 一种车牌识别的方法及相关装置
CN111275796A (zh) * 2020-01-17 2020-06-12 北京迈格威科技有限公司 车牌合成方法、装置、计算机设备和存储介质
CN111275796B (zh) * 2020-01-17 2023-09-19 北京迈格威科技有限公司 车牌合成方法、装置、计算机设备和存储介质
CN111340045A (zh) * 2020-02-12 2020-06-26 浙江大华技术股份有限公司 车牌号码的识别方法、装置及存储介质
CN111340045B (zh) * 2020-02-12 2023-09-01 浙江大华技术股份有限公司 车牌号码的识别方法、装置及存储介质
CN111639636A (zh) * 2020-05-29 2020-09-08 北京奇艺世纪科技有限公司 一种字符识别方法及装置
CN111695563A (zh) * 2020-06-10 2020-09-22 北京筑梦园科技有限公司 一种单双层车牌识别方法、服务器及停车收费系统
CN111695563B (zh) * 2020-06-10 2022-07-05 北京筑梦园科技有限公司 一种单双层车牌识别方法、服务器及停车收费系统
CN111950659A (zh) * 2020-09-01 2020-11-17 湖南国科微电子股份有限公司 双层车牌图像处理方法、装置、电子设备及存储介质
WO2023169273A1 (zh) * 2022-03-11 2023-09-14 阿里巴巴(中国)有限公司 图像识别方法、装置、设备和存储介质
CN115131776A (zh) * 2022-04-12 2022-09-30 公安部交通管理科学研究所 一种融合新能源汽车特征的车牌图像字符分类识别方法

Also Published As

Publication number Publication date
CN108073928B (zh) 2021-04-02
CN108073928A (zh) 2018-05-25

Similar Documents

Publication Publication Date Title
WO2018090771A1 (zh) 一种车牌识别方法及装置
WO2018121006A1 (zh) 一种车牌定位方法及装置
WO2018028306A1 (zh) 一种车牌号码识别方法及装置
US9171204B2 (en) Method of perspective correction for devanagari text
US9082038B2 (en) Dram c adjustment of automatic license plate recognition processing based on vehicle class information
US10303968B2 (en) Method and apparatus for image recognition
US9053361B2 (en) Identifying regions of text to merge in a natural image or video frame
US9773186B2 (en) Methods for mobile image capture of vehicle identification numbers in a non-document
CN108073926B (zh) 一种车牌识别方法及装置
CN108108734B (zh) 一种车牌识别方法及装置
CN108229466B (zh) 一种车牌识别方法及装置
CN112997190B (zh) 车牌识别方法、装置及电子设备
CN110008997B (zh) 图像纹理相似度识别方法、装置及计算机可读存储介质
CN112380978B (zh) 基于关键点定位的多人脸检测方法、系统及存储介质
CN111507324B (zh) 卡片边框识别方法、装置、设备和计算机存储介质
CN111681256A (zh) 图像边缘检测方法、装置、计算机设备和可读存储介质
CN111160395A (zh) 图像识别方法、装置、电子设备和存储介质
CN111382625A (zh) 道路标识识别方法、装置及电子设备
CN108090484B (zh) 一种车牌识别方法及装置
CN110232381B (zh) 车牌分割方法、装置、计算机设备及计算机可读存储介质
CN113228105A (zh) 一种图像处理方法、装置和电子设备
WO2018058573A1 (zh) 对象检测方法、对象检测装置以及电子设备
CN114155363A (zh) 换流站车辆识别方法、装置、计算机设备和存储介质
CN111709377A (zh) 特征提取方法、目标重识别方法、装置及电子设备
CN112257628A (zh) 一种户外比赛运动员的身份识别方法、装置及设备

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: 17872262

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17872262

Country of ref document: EP

Kind code of ref document: A1