CN116702809A - Simple pattern license plate recognition and detection method - Google Patents
Simple pattern license plate recognition and detection method Download PDFInfo
- Publication number
- CN116702809A CN116702809A CN202310574177.1A CN202310574177A CN116702809A CN 116702809 A CN116702809 A CN 116702809A CN 202310574177 A CN202310574177 A CN 202310574177A CN 116702809 A CN116702809 A CN 116702809A
- Authority
- CN
- China
- Prior art keywords
- license plate
- simple pattern
- color
- plate
- code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 11
- 239000003086 colorant Substances 0.000 claims abstract description 6
- 238000013461 design Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 12
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000005065 mining Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
- G06K7/1482—Methods for optical code recognition the method including quality enhancement steps using fuzzy logic or natural solvers, such as neural networks, genetic algorithms and simulated annealing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
- G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
- G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
- G06K19/06018—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking one-dimensional coding
- G06K19/06028—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking one-dimensional coding using bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
- G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
- G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
- G06K19/06037—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking multi-dimensional coding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
- G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
- G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
- G06K19/06046—Constructional details
- G06K19/06075—Constructional details the marking containing means for error correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
- G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
- G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
- G06K19/06046—Constructional details
- G06K19/06131—Constructional details the marking comprising a target pattern, e.g. for indicating the center of the bar code or for helping a bar code reader to properly orient the scanner or to retrieve the bar code inside of an image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/1413—1D bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/1417—2D bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1443—Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1452—Methods for optical code recognition including a method step for retrieval of the optical code detecting bar code edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1456—Methods for optical code recognition including a method step for retrieval of the optical code determining the orientation of the optical code with respect to the reader and correcting therefore
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
- G06K7/1473—Methods for optical code recognition the method including quality enhancement steps error correction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a simple pattern license plate recognition and detection method, which comprises the following steps: designing simple pattern license plates formed by areas with different colors, and setting coding rules of the simple pattern license plates; the simple pattern license plate is arranged on a carriage back plate, and a differential color pattern is formed through a boundary color area of the simple pattern license plate and a background color area of the carriage back plate; photographing a carriage rear plate at the rear part of a passing vehicle through a camera arranged in the identification area, and transmitting the vehicle photograph to a vehicle identification management system; and the vehicle identification management system performs deep neural network learning comparison on the simple pattern license plates in the vehicle photos, and calculates according to the coding rules to obtain license plate codes. The method effectively improves the accuracy of license plate recognition and detection in the mine production environment.
Description
Technical Field
The invention relates to a license plate recognition method, in particular to a simple pattern license plate recognition detection method.
Background
In mine production work, the vehicle is one of important tools in mine production flow, plays an important role in aspects of ore transportation, cargo loading and unloading and the like, and realizes intelligent and efficient management of the vehicle as a basis for realizing automatic management of the whole operation flow.
The vehicle is an entity, and different entities exist independently, so that unique identification information is required to be provided for the vehicle, and the unique identification information is used for distinguishing different vehicles. Because the vehicles belong to mass standard production, namely vehicles of the same brand and the same type have very little difference in appearance, color and the like, even if different manufacturers only belong to the same type, the appearance of the vehicles is not greatly different, and therefore, unique identification information is required to be manufactured for the vehicles manually.
The traditional method generally adopts the combination of characters and numbers as license plates, so that mine production management is also carried out, and the license plates are sprayed on a baffle plate at the rear of the vehicle in a spraying mode. Although the method can have good effects in general scenes, such as scenes with simple ordinary roads and environments, the method cannot have the same effects in mining production environments. Because the mining production environment is more complicated, the topography is abominable, seals dirt too much, light is very inhomogeneous etc. leads to the license plate to receive wearing and tearing easily, shelter from and pollute, finally leads to using the license plate to discern the effect of vehicle not good. Vehicle identification is critical to the management of the whole mining workflow, so how to improve the accuracy of license plate identification is a problem to be solved urgently.
Disclosure of Invention
The invention provides a simple pattern license plate recognition and detection method, which aims at: the simple pattern license plate suitable for the mine production environment is designed, and the accuracy of mine license plate recognition is improved by combining a specific license plate recognition detection method.
The technical scheme of the invention is as follows:
a simple pattern license plate recognition and detection method comprises the following steps:
s1: designing simple pattern license plates formed by areas with different colors, and setting coding rules of the simple pattern license plates;
s2: the simple pattern license plate is arranged on a carriage back plate, and a differential color pattern is formed through a boundary color area of the simple pattern license plate and a background color area of the carriage back plate;
s3: photographing a carriage rear plate at the rear part of a passing vehicle through a camera arranged in the identification area, and transmitting the vehicle photograph to a vehicle identification management system;
s4: and the vehicle identification management system performs deep neural network learning comparison on the simple pattern license plates in the vehicle photos, and calculates according to the coding rules to obtain license plate codes.
Further, the simple pattern license plate is a bar code plate, the bar code plate adopts a background color bar and a white color bar which are vertically arranged at intervals, the starting position and the ending position of the color bar are both the background color bar, and the color of the background color bar is consistent with that of a carriage back plate.
Further, the positions of the two sides of the bar code board are a white left side boundary standard position and a white right side boundary standard position, and a check bit is arranged between the right side boundary standard position and a background color bar of the termination position.
Further, the color bar coding rule of the bar code card is as follows: the background color bars and the white color bars are provided with two widths, wherein the fine background color bar code is 0, the fine white color bar code is 1, the coarse background color bar code is 2, and the coarse white color bar code is 3.
Further, the calculating step of the value of the check bit is as follows:
a. the position code of the color bars of the check bit is 1, and the position code of the bar code card is obtained by increasing the position code of each color bar from right to left from the start of the check bit;
b. the color bars with even number are coded by the position, the corresponding color bar codes are multiplied by 3 respectively and summed to obtain a result x;
c. the position codes are color bars with odd numbers, and the corresponding color bar codes are summed to obtain a result y;
d. and c, summing the results obtained in the step b and the step c, wherein z=x+y, and subtracting z from a number which is greater than or equal to z and is the smallest integer multiple of 10, wherein the difference is the value of the check bit.
Further, the simple pattern license plate is a two-dimensional code plate, the two-dimensional code plate is in a nine-grid shape, 9 blocks in the nine-grid are respectively background color blocks or white blocks, and the background color blocks are consistent with the colors of the carriage rear plate.
Further, a circle of frame boundary standard positions are arranged on the periphery of the nine-grid, the upper left corner area of the frame boundary standard positions is a positioning area, and the positioning area is used for determining the image direction of the two-dimensional code nine-grid shape.
Further, the color block coding rule of the two-dimensional code plate is as follows: the color block code of the background color block is 0, the color block code of the white block is 1, and the blocks contacted by the positioning area start to be coded from top to bottom and from left to right.
Further, the neural network learning step is as follows:
setting up a feature representation learning layer based on a deep neural network, obtaining a feature map of an image, detecting the feature map, taking a vehicle tailgate area as an interested area ROI, and obtaining a corresponding candidate area set from the feature map;
performing binarization processing on the image by utilizing a local self-adaptive threshold value, and further screening the contour conforming to the license plate according to the obtained contour information to obtain a license plate region;
traversing the pixel values of four edges of the license plate, respectively fitting straight lines, calculating intersection points of the straight lines, and calculating a projection matrix according to the coordinates of the intersection points and the real size of the license plate;
IV, performing projection transformation on the obtained original image of the region of interest by using a projection matrix to obtain a license plate with corrected direction;
and V, obtaining simple pattern license plate information by using local self-adaptive threshold and contour information, and obtaining corresponding pattern license plate codes by calculating and comparing with the design size.
Further, in step S4, a simple pattern license plate library is preset in the vehicle recognition management system, and the license plate code obtained by the neural network learning is checked with the license plate code in the simple pattern license plate library in a table look-up mode, so as to obtain a correct and consistent license plate code.
Compared with the prior art, the invention has the following beneficial effects:
(1) The simple pattern license plate is designed based on the bar code technology and the two-dimensional code technology in the object recognition, and the license plate recognition detection is carried out in a deep neural network learning mode, so that the problem of low license plate recognition accuracy caused by mine production environment factors is greatly avoided;
(2) The specific positions of the bar code plate and the two-dimensional code plate are provided with check bits or positioning areas for identifying the direction of the simple pattern license plate, so that the accuracy of pattern license plate identification is further improved;
(3) In license plate recognition, a deep neural network learning and statistical method are combined, and patterns and a coding library are compared twice, so that the accuracy of license plate recognition is further improved.
Drawings
FIG. 1 is a schematic diagram of a bar code pattern license plate without check bits;
FIG. 2 is a schematic diagram of a bar code pattern license plate with check bits;
FIG. 3 is a schematic view of a bar code pattern license plate mounted to a rear plate of a vehicle cabin;
FIG. 4 is a schematic diagram of a two-dimensional code pattern license plate without a positioning area;
FIG. 5 is a schematic diagram of a two-dimensional code pattern license plate with a positioning area;
fig. 6 is a schematic structural diagram of a two-dimensional code pattern license plate mounted on a rear plate of a carriage;
fig. 7 is a schematic process diagram of the method.
Description of the embodiments
The technical scheme of the invention is described in detail below with reference to the accompanying drawings:
a simple pattern license plate recognition and detection method comprises the following steps:
s1: and designing simple pattern license plates formed by areas with different colors, and setting coding rules of the simple pattern license plates.
Preferably, as shown in fig. 1 to 3, the simple pattern license plate is a bar code plate 1, the bar code plate 1 adopts a background color bar 5 and a white bar 6 which are vertically arranged at intervals, the starting position and the ending position of the color bars are both the background color bar 5, that is, the total number of the color bars is an odd number, and the color of the background color bar 5 is consistent with that of a carriage back plate 12. In this embodiment, 9 strips are arranged at intervals between the background color strip 5 and the white strip 6, so that the standard specification of more than 100 vehicles can be satisfied.
The two sides of the bar code board 1 are provided with a white left side standard bit 2 and a white right side standard bit 3, and the width of the standard bit is fixed (5-8 cm). The left side margin specification bit 2 and the right side margin specification bit 3 are used to determine the license plate range, i.e., the space bounded by the left and right margin specification bits shown in fig. 1-3 is the license plate range, where no encoded information is represented. The start and end of the bar code are composed of background color bars 5, and then arranged in a one-to-one alternate manner according to the background color bars 5 and white bars 6 to form the bar code. Each color bar is assigned a fixed code, so when the bar code is determined, the corresponding code is also determined.
The color bar coding rule of the bar code plate is as follows: the background color bars 5 and the white color bars 6 are respectively provided with two widths, wherein the fine background color bar code is 0, the fine white color bar code is 1, the coarse background color bar code is 2, and the coarse white color bar code is 3. The corresponding license plate code of fig. 2 is 030121030.
A check bit 4 is arranged between the right side edge specification bit 3 and the background color bar 5 at the end position, and the purpose of the check bit is to check whether the obtained code is accurate. The calculation step of the value of the check bit 4 is as follows:
a. the position code of the color bar of the check bit 4 is 1, and from the check bit 4, each color bar of the bar code card is progressively increased from right to left, so that the position code of the bar code card from left to right is {10, 9, 8, 7, 6, 5, 4, 3, 2, 1};
b. the color bars with even number are coded by the position, the corresponding color bar codes are multiplied by 3 respectively and summed to obtain a result x;
x=0*3+0*3+2*3+0*3+0*3=6
c. the position codes are color bars with odd numbers, and the corresponding color bar codes are summed to obtain a result y;
y=3+1+1+3=8
d. and c, summing the results obtained in the step b and the step c, wherein z=x+y, and subtracting z from a number which is greater than or equal to z and is the smallest integer multiple of 10, wherein the difference is the value of the check bit 4.
In this embodiment, z=x+y=14, 20-14=6, and the value of parity bit 4 is 6.
Further preferably, as shown in fig. 4 to 6, the simple pattern license plate is a two-dimensional code plate 7, the two-dimensional code plate 7 adopts a nine-grid shape, 9 blocks in the nine-grid are respectively a background color block 10 or a white block 11, and the color of the background color block 10 is consistent with that of a carriage rear plate 12.
The periphery of the nine-grid is provided with a circle of frame boundary standard positions 8, the upper left corner area of the frame boundary standard positions 8 is a positioning area 9, and the positioning area 9 is used for determining the image direction of the two-dimensional code nine-grid shape.
The color block coding rule of the two-dimensional code plate is as follows: the color block code of the background color block is 0, the color block code of the white block is 1, and the blocks contacted by the positioning area 9 are sequentially coded from top to bottom and from left to right, so that the sequence code corresponding to the two-dimensional code is obtained. The license plate code corresponding to fig. 4 is 101010100. Since the above design provides for the picture orientation, i.e. the default image is the positive orientation. If the image is reversed due to external factors, the license plate is likely to be encoded as: 001010101.
in the actual detection process, the direction of the image can be judged according to the positioning area 9, so that adjustment is further made, and the accuracy of two-dimensional code identification is improved.
S2: the simple pattern license plate is mounted on the car rear plate 12, and a differential color pattern is formed by the boundary color region of the simple pattern license plate and the background color region of the car rear plate 12.
S3: the vehicle photograph is transmitted to the vehicle recognition management system by photographing the cabin back plate 12 of the rear part of the passing vehicle through the camera provided in the recognition area.
S4: and the vehicle identification management system performs deep neural network learning comparison on the simple pattern license plates in the vehicle photos, and calculates according to the coding rules to obtain license plate codes.
Preferably, as shown in fig. 7, the learning step of the neural network is:
setting up a feature representation learning layer based on a deep neural network, obtaining a feature map of an image, detecting the feature map, taking a vehicle tailgate area as an interested area ROI, and obtaining a corresponding candidate area set from the feature map;
performing binarization processing on the image by utilizing a local self-adaptive threshold value, and further screening the contour conforming to the license plate according to the obtained contour information to obtain a license plate region;
traversing the pixel values of four edges of the license plate, respectively fitting straight lines, calculating intersection points of the straight lines, and calculating a projection matrix according to the coordinates of the intersection points and the real size of the license plate;
IV, performing projection transformation on the obtained original image of the region of interest by using a projection matrix to obtain a license plate with corrected direction;
and V, obtaining simple pattern license plate information (bar code or two-dimensional code) by using local self-adaptive threshold and contour information, and obtaining corresponding pattern license plate codes by calculating and comparing with the design size.
Specifically, if the check bit 4 is identified, judging that the simple pattern license plate is a bar code plate 1, calculating the width of the bar code and the gap, comparing the width with the design size, and acquiring a corresponding bar code license plate code by inquiring a license plate code rule preset in a vehicle identification management system; if the positioning area 9 is identified, judging that the license plate with the simple pattern is the two-dimensional code plate 7, extracting the position of a background color block in the pattern, and acquiring the corresponding two-dimensional code license plate code by inquiring a license plate code rule preset in a vehicle identification management system.
Further preferably, a simple pattern license plate library of the mining operation vehicle is preset in the vehicle identification management system, and the license plate codes obtained by the neural network learning are checked with the license plate codes in the simple pattern license plate library in a table look-up mode to obtain the license plate codes which are accurate and consistent.
The method aims to solve the problem that an original license plate is easy to damage and pollute in a mining operation environment, and designs and realizes a new-generation bar code license plate based on an object identification bar code technology, wherein the new-generation bar code license plate consists of color bars with different widths, the color bars are divided into a white bar and a background color bar, and a left boundary standard position and a right boundary standard position consist of white bars with fixed widths. The corresponding codes of color bars with different widths are inconsistent. Therefore, the width of the color bar is important for accurately expressing license plate codes, and in order to verify the width identification effect, experiments are carried out in the actual production environment, and the bar code plate is basically not affected by local pollution as only the width of lines in the pattern is required to be identified, so that the identification accuracy of the bar code plate is hundred percent.
Although bar code cards have achieved better performance, they have limited coding range due to space constraints, and while meeting current production scale needs, they cannot be extended to larger scales. The two-dimensional code is also a commonly used object identification code, and compared with the bar code, the two-dimensional code has stronger expression capability, can cover a larger representation space and can accommodate larger representation contents. Therefore, in order to solve the problem of weak expression capacity of the bar code cards, the method designs a simple two-dimensional code card based on a two-dimensional space. The design rule is as follows: the foreground color is white block, the background color is background color block of the vehicle, the periphery is white boundary box (frame boundary standard position) with uniform width, and the middle is divided into nine gridsForm of the invention. The design scheme can generate 2 9 The license plate code greatly expands the representation space. In addition, as the two-dimensional code is square in shape, each small cell is square, compared with the bar code, the two-dimensional code is easier to identify, and therefore performance of license plate identification is improved.
In order to verify the identification effect of the two-dimensional code square, experiments are carried out in the actual production environment, and as only the positions of background color blocks in the patterns are needed to be identified, the two-dimensional code cards are not affected by local pollution basically, and the identification accuracy of the two-dimensional code cards is hundred percent.
The present invention is not limited to the above embodiments, and any person who can learn the structural changes made under the teaching of the present invention can fall within the scope of the present invention if the present invention has the same or similar technical solutions.
The technology, shape, and construction parts of the present invention, which are not described in detail, are known in the art.
Claims (10)
1. The simple pattern license plate recognition and detection method is characterized by comprising the following steps of:
s1: designing simple pattern license plates formed by areas with different colors, and setting coding rules of the simple pattern license plates;
s2: the simple pattern license plate is arranged on a carriage back plate, and a differential color pattern is formed through a boundary color area of the simple pattern license plate and a background color area of the carriage back plate;
s3: photographing a carriage rear plate at the rear part of a passing vehicle through a camera arranged in the identification area, and transmitting the vehicle photograph to a vehicle identification management system;
s4: and the vehicle identification management system performs deep neural network learning comparison on the simple pattern license plates in the vehicle photos, and calculates according to the coding rules to obtain license plate codes.
2. The simple pattern license plate recognition detection method as claimed in claim 1, wherein: the simple pattern license plate is a bar code plate, the bar code plate adopts a background color bar and a white color bar which are vertically arranged at intervals, the starting position and the ending position of the color bar are both the background color bar, and the color of the background color bar is consistent with that of a carriage back plate.
3. The simple pattern license plate recognition detection method as claimed in claim 2, wherein: the two sides of the bar code board are a white left side standard bit and a white right side boundary standard bit, and a check bit is arranged between the right side standard bit and a background color bar at the termination position.
4. The simple pattern license plate recognition detection method as claimed in claim 3, wherein the color bar coding rule of the bar code plate is: the background color bars and the white color bars are provided with two widths, wherein the fine background color bar code is 0, the fine white color bar code is 1, the coarse background color bar code is 2, and the coarse white color bar code is 3.
5. The method for detecting simple pattern license plate recognition according to claim 4, wherein the step of calculating the value of the check bit is:
a. the position code of the color bars of the check bit is 1, and the position code of the bar code card is obtained by increasing the position code of each color bar from right to left from the start of the check bit;
b. the color bars with even number are coded by the position, the corresponding color bar codes are multiplied by 3 respectively and summed to obtain a result x;
c. the position codes are color bars with odd numbers, and the corresponding color bar codes are summed to obtain a result y;
d. and c, summing the results obtained in the step b and the step c, wherein z=x+y, and subtracting z from a number which is greater than or equal to z and is the smallest integer multiple of 10, wherein the difference is the value of the check bit.
6. The simple pattern license plate recognition detection method as claimed in claim 1, wherein: the simple pattern license plate is a two-dimensional code plate, the two-dimensional code plate is in a nine-grid shape, 9 blocks in the nine-grid are respectively background color blocks or white blocks, and the background color blocks are consistent with the colors of the carriage rear plate.
7. The simple pattern license plate recognition detection method as claimed in claim 6, wherein: the periphery of the nine-grid is provided with a circle of frame boundary standard positions, the upper left corner area of the frame boundary standard positions is a positioning area, and the positioning area is used for determining the image direction of the two-dimensional code nine-grid shape.
8. The simple pattern license plate recognition detection method of claim 7, wherein the color block coding rule of the two-dimensional code plate is: the color block code of the background color block is 0, the color block code of the white block is 1, and the blocks contacted by the positioning area start to be coded from top to bottom and from left to right.
9. The simple pattern license plate recognition detection method as claimed in claim 1, wherein the neural network learning step comprises:
setting up a feature representation learning layer based on a deep neural network, obtaining a feature map of an image, detecting the feature map, taking a vehicle tailgate area as an interested area ROI, and obtaining a corresponding candidate area set from the feature map;
performing binarization processing on the image by utilizing a local self-adaptive threshold value, and further screening the contour conforming to the license plate according to the obtained contour information to obtain a license plate region;
traversing the pixel values of four edges of the license plate, respectively fitting straight lines, calculating intersection points of the straight lines, and calculating a projection matrix according to the coordinates of the intersection points and the real size of the license plate;
IV, performing projection transformation on the obtained original image of the region of interest by using a projection matrix to obtain a license plate with corrected direction;
and V, obtaining simple pattern license plate information by using local self-adaptive threshold and contour information, and obtaining corresponding pattern license plate codes by calculating and comparing with the design size.
10. The simple pattern license plate recognition detection method as claimed in claim 1, wherein: in step S4, a simple pattern license plate library is preset in the vehicle recognition management system, and the license plate codes obtained by the neural network learning are checked with the license plate codes in the simple pattern license plate library in a table look-up mode, so as to obtain the license plate codes which are accurate and consistent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310574177.1A CN116702809A (en) | 2023-05-22 | 2023-05-22 | Simple pattern license plate recognition and detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310574177.1A CN116702809A (en) | 2023-05-22 | 2023-05-22 | Simple pattern license plate recognition and detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116702809A true CN116702809A (en) | 2023-09-05 |
Family
ID=87834912
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310574177.1A Pending CN116702809A (en) | 2023-05-22 | 2023-05-22 | Simple pattern license plate recognition and detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116702809A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012128669A (en) * | 2010-12-15 | 2012-07-05 | Sumitomo Electric Ind Ltd | Vehicle color determining device, computer program and vehicle color determining method |
| CN107730905A (en) * | 2017-06-13 | 2018-02-23 | 银江股份有限公司 | Multitask fake license plate vehicle vision detection system and method based on depth convolutional neural networks |
| CN110110676A (en) * | 2019-05-13 | 2019-08-09 | 深圳市万物云科技有限公司 | Automatic error correction method and device based on Car license recognition |
| CN113095320A (en) * | 2021-04-01 | 2021-07-09 | 湖南大学 | License plate recognition method and system and computing device |
-
2023
- 2023-05-22 CN CN202310574177.1A patent/CN116702809A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012128669A (en) * | 2010-12-15 | 2012-07-05 | Sumitomo Electric Ind Ltd | Vehicle color determining device, computer program and vehicle color determining method |
| CN107730905A (en) * | 2017-06-13 | 2018-02-23 | 银江股份有限公司 | Multitask fake license plate vehicle vision detection system and method based on depth convolutional neural networks |
| CN110110676A (en) * | 2019-05-13 | 2019-08-09 | 深圳市万物云科技有限公司 | Automatic error correction method and device based on Car license recognition |
| CN113095320A (en) * | 2021-04-01 | 2021-07-09 | 湖南大学 | License plate recognition method and system and computing device |
Non-Patent Citations (2)
| Title |
|---|
| 《职业教育物流类专业产教融合创新教材智能物流技术》: "《职业教育物流类专业产教融合创新教材智能物流技术》", 31 March 2021, 机械工业出版社, pages: 10 * |
| 林丽芝: "条形码技术在车牌识别中的应用研究", 《辽宁科技学院学报》, vol. 18, no. 6, 31 December 2016 (2016-12-31), pages 1 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109886896B (en) | Blue license plate segmentation and correction method | |
| CN109949578B (en) | Vehicle line pressing violation automatic auditing method based on deep learning | |
| CN102375982B (en) | Multi-character characteristic fused license plate positioning method | |
| CN106295646B (en) | A kind of registration number character dividing method and device based on deep learning | |
| CN111179152B (en) | Road identification recognition method and device, medium and terminal | |
| US20210342606A1 (en) | Parking Detection Method, System, Processing Device and Storage Medium | |
| CN108596166A (en) | A kind of container number identification method based on convolutional neural networks classification | |
| CN102629322B (en) | Character feature extraction method based on stroke shape of boundary point and application thereof | |
| CN106815583B (en) | Method for positioning license plate of vehicle at night based on combination of MSER and SWT | |
| CN115170669B (en) | Identification and positioning method and system based on edge feature point set registration and storage medium | |
| CN105069456A (en) | License plate character segmentation method and apparatus | |
| CN106156752B (en) | A kind of model recognizing method based on inverse projection three-view diagram | |
| CN110110608B (en) | Forklift speed monitoring method and system based on vision under panoramic monitoring | |
| CN112560538B (en) | Method for quickly positioning damaged QR (quick response) code according to image redundant information | |
| CN102915544A (en) | Video image motion target extracting method based on pattern detection and color segmentation | |
| CN111507411B (en) | Image comparison method and system | |
| CN107031033A (en) | It is a kind of can 3D printing hollow out Quick Response Code model generating method and system | |
| CN112183301B (en) | Intelligent building floor identification method and device | |
| CN109919144A (en) | Drivable region detection method, device, computer storage medium and drive test visual apparatus | |
| CN103544489A (en) | Device and method for locating automobile logo | |
| CN104751466A (en) | Deform able object tracking algorithm based on visual salience and system thereof | |
| CN111881832A (en) | Lane target detection method, device, equipment and computer readable storage medium | |
| CN110567383A (en) | pantograph abrasion early warning system and detection method based on structural forest and sub-pixels | |
| CN110516666B (en) | License plate positioning method based on combination of MSER and ISODATA | |
| CN111462119A (en) | Wide and thick plate shearing stock layout method based on machine vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |