CN104636748B - A kind of method and device of number plate identification - Google Patents
A kind of method and device of number plate identification Download PDFInfo
- Publication number
- CN104636748B CN104636748B CN201310564073.9A CN201310564073A CN104636748B CN 104636748 B CN104636748 B CN 104636748B CN 201310564073 A CN201310564073 A CN 201310564073A CN 104636748 B CN104636748 B CN 104636748B
- Authority
- CN
- China
- Prior art keywords
- number plate
- preset
- character
- characters
- plate area
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 230000011218 segmentation Effects 0.000 claims description 41
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000012706 support-vector machine Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Landscapes
- Character Discrimination (AREA)
Abstract
The embodiment of the invention discloses a kind of method and devices of number plate identification, including：Obtain the number plate region in image；According to the identification region parameter of preset characters and each intercharacter Relation Parameters, the identification region of each character is determined；The identification region of each character is identified with preset Character recognizer, obtains recognition result；Calculate the sum of corresponding matching degree of each character；Within a preset range, according to default ergodic condition, the identification region parameter of preset characters and each intercharacter Relation Parameters are adjusted, back to character recognition region step is determined, until having traversed the value in preset range according to default ergodic algorithm；From the sum of corresponding matching degree of all calculated each characters, the highest recognition result of the sum of the corresponding matching degree of each character is obtained as preferred recognition result；Export preferred recognition result.The present invention has evaded the problem of image processing step is by noise jamming in character recognition, improves system stability and anti-interference.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a method and a device for identifying a number plate.
Background
The number plate identification technology has wide application in practical application, and can be summarized into the following aspects: number plate identification is a technical basis of many important applications, such as in the field of intelligent transportation, automatic parking management, high-speed automatic charging, vehicle violation monitoring and the like; number plate recognition is the basis for positioning and managing vehicles, such as in the field of mass video retrieval, license plate recognition, house number recognition, character and road sign recognition and the like; number plate identification provides an important clue for accurate retrieval.
In the prior art, the number plate identification method generally comprises the following three steps: number plate positioning, character segmentation and character recognition. The character segmentation is a key step of the number plate identification method. In the prior art, a method of image processing is generally adopted to perform character segmentation, for example: image enhancement, binarization, connected region analysis methods and the like. These methods are greatly affected by noise, and the character segmentation and the character recognition may fail due to the noise.
In addition, in the prior art, character segmentation in number plate recognition usually processes characters individually, rather than processing the number plate as a whole, and the relationship between the characters is not fully considered, so that the individual character segmentation is easily influenced by noise of an input image, the system stability and the anti-interference performance are low, and the method cannot be widely applied to complex environments.
Disclosure of Invention
The embodiment of the invention provides a method and a device for identifying a number plate, which are used for avoiding the noise interference in the image processing step and improving the system stability and anti-interference performance.
In order to achieve the purpose, the embodiment of the invention discloses a method for identifying a number plate, which comprises the following steps:
step A, acquiring a number plate area in an image;
b, determining the recognition area of each character in the number plate area according to the recognition area parameters of the preset characters in the number plate area and the relation parameters among the characters;
step C, recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: identifying each character and the matching degree corresponding to each character;
d, calculating the sum of the matching degrees corresponding to all the characters in the number plate area;
step E, in a preset range, adjusting the identification area parameters and the relationship parameters among the characters of the preset characters in the number plate area according to a preset traversal condition, returning to the step B until values in the preset range are traversed according to a preset traversal algorithm, and obtaining the identification result after the identification area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time and the sum of the calculated matching degrees corresponding to the characters in the number plate area;
step F, obtaining the recognition result with the highest matching degree sum corresponding to each character in the number plate area from the matching degree sum corresponding to each character in all the calculated number plate areas as the preferred recognition result;
and G, outputting the preferred identification result as a license plate identification result.
Preferably, the parameters of the recognition area of the preset character include: coordinate values (x, y) of the central point of a segmentation frame of a preset character in the number plate area;
the dividing frame is as follows: the width of the rectangle is a first preset value, and the ratio of the width to the height of the rectangle is 1: 2;
the inter-character relation parameter comprises: the width value w between two adjacent characters in the number plate area; a width coefficient value k between two adjacent characters preset in the number plate area;
the preset range comprises: the variation range of x and y in the coordinate values (x, y) of the central point of the segmentation frame of the preset character; the variation range of the width value w between the two adjacent characters; and the preset variation range of the width coefficient value k between two adjacent characters.
Preferably, the determining the recognition area of each character in the number plate area according to the recognition area parameter and the inter-character relation parameter of the preset character in the number plate area includes:
obtaining the coordinate value (x) of the central point of the segmentation frame of the preset character in the number plate areapre1，ypre1）；
Obtaining the width value w between two adjacent characters in the number plate areapre1；
Obtaining the value k of the width coefficient between two adjacent characters preset in the number plate areapre1；
According to x = xpre1+(n-npre1)*wpre1Or x = xpre1+(n-npre1)*wpre1±(wpre1*(kpre1-1)) to determine the coordinates (x, y) of the centre point of the segmentation box for the other characters in the number platepre1) (ii) a N ispre1Presetting the serial numbers of characters in the number plate area, wherein n is the serial numbers of other characters in the number plate area;
and determining the recognition area of each character in the number plate area by taking the coordinate of the central position of the segmentation frame of each character in the number plate area as a reference.
Preferably, the traversing of the values in the preset range according to the preset traversal algorithm is as follows:
traversing values in the variation range of x and y, the variation range of the width value w between the two adjacent characters and the variation range of the width coefficient value k between the two adjacent characters in the coordinate values (x, y) of the central point of the segmentation frame of the preset character according to a preset traversal algorithm;
the preset traversal algorithm comprises the following steps: gradient descent or newton's method or exhaustive method.
Preferably, the obtaining a recognition result with the highest sum of matching degrees corresponding to the characters in the number plate area as a preferred recognition result includes:
according toAcquiring the highest value of the sum of the matching degrees corresponding to all characters in the number plate area; wherein,
Smaxthe highest value of the sum of the matching degrees of all the characters in the number plate area in the plurality of recognition results; ciA division frame for the ith character in the number plate area; (x)i,yi) Is divided into a division frame CiThe center position coordinates of (a); n is the number of characters contained in a preset number plate area; fscore{Ci(xi,yi) Is to use a preset character recognizer pair CiThe characters contained in the data are identified to obtain a matching degree value;the sum of the matching degrees of all characters in the number plate area corresponding to each recognition result is obtained;
will SmaxThe corresponding recognition result is used as the number plate recognition result.
Preferably, the preset character recognizer is: a support vector machine based recognizer or a neural network based recognizer or a deep learning based recognizer.
Preferably, before outputting the preferred recognition result as the number plate recognition result, the method further comprises:
step H, judging whether the sum of the matching degrees corresponding to the characters in the number plate area corresponding to the optimal recognition result is larger than a preset first threshold value or not;
step I, if the value is larger than the value, executing step G, otherwise, executing step J;
step J, judging whether the times of zooming the number plate area is larger than a preset second threshold value or not;
step K, if the number plate area is not larger than the preset number plate area, zooming the number plate area by using a first preset proportion, returning to the step B, and otherwise, executing the step L;
and L, acquiring a preferred recognition result with the highest sum of the matching degrees of all characters in the number plate from a plurality of preferred recognition results.
Preferably, the scaling the number plate area by the first preset proportion includes:
scaling the number plate area by a first preset proportion; and
the preset range is scaled by a first preset ratio.
In order to achieve the above object, the embodiment of the present invention discloses a device for identifying a number plate, the device comprising:
a number plate area acquisition unit for acquiring a number plate area in the image;
the identification area determining unit is used for determining the identification area of each character in the number plate area according to the identification area parameter of the preset character in the number plate area and the relation parameter among the characters;
the character recognition unit is used for recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: identifying each character and the matching degree corresponding to each character;
the matching degree sum calculating unit is used for calculating the sum of the matching degrees corresponding to all the characters in the number plate area;
the parameter adjusting unit is used for adjusting the recognition area parameters and the relationship parameters among the characters of the preset characters in the number plate area within a preset range, and repeatedly calling the recognition area determining unit, the character recognition unit and the matching degree sum calculating unit until values within the preset range are traversed according to a preset traversal algorithm, so that a recognition result obtained after the recognition area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time and the calculated matching degree sum corresponding to each character in the number plate area is obtained;
the preferred recognition result obtaining unit is used for obtaining a recognition result with the highest matching degree sum corresponding to each character in the number plate area from the matching degree sum corresponding to each character in all the calculated number plate areas as a preferred recognition result;
and the output unit is used for outputting the preferred identification result as the number plate identification result.
Preferably, the identification area determining unit is configured to determine the identification area of each character in the number plate area according to a coordinate value (x, y) of a center point of a segmentation frame of a preset character in the number plate area, a width value w between two adjacent characters, and a width coefficient value k between two adjacent preset characters;
the parameter adjusting unit comprises a preset range acquiring subunit;
the preset range acquiring subunit is configured to acquire a variation range of x and y in the coordinate values (x, y) of the center point of the segmentation frame of the preset character, a variation range of a width value w between two adjacent characters, and a variation range of a width coefficient value k between two adjacent characters.
Preferably, the identification area determination unit includes: the system comprises a coordinate acquisition subunit, a width coefficient acquisition subunit, a coordinate calculation subunit and an identification area determination subunit;
the coordinate acquiring subunit is used for acquiring the coordinate value (x) of the central point of the segmentation frame of the preset character in the number plate areapre1，ypre1）；
The width acquisition subunit is used for acquiring the width value w between two adjacent characters in the number plate areapre1；
The width coefficient acquisition subunit is used for acquiring a width coefficient value k between two adjacent characters preset in the number plate areapre1；
The coordinate calculation subunit is used for calculating the coordinate according to x = xpre1+(n-npre1)*wpre1Or
x=xpre1+(n-npre1)*wpre1±(wpre1*(kpre1-1)) to determine the coordinates (x, y) of the centre point of the segmentation box for the other characters in the number platepre1) (ii) a N ispre1Presetting the serial numbers of characters in the number plate area, wherein n is the serial numbers of other characters in the number plate area;
and the identification area determining subunit is used for determining the identification area of each character in the number plate area by taking the coordinate of the central position of the segmentation frame of each character in the number plate as the standard.
Preferably, the preferred recognition result obtaining unit includes: a highest matching degree sum calculation subunit and an optimal recognition result acquisition subunit;
the highest matching degree sum calculating subunit is used for obtaining the highest value of the matching degree sum corresponding to each character in the number plate area, wherein,
Smaxthe highest value of the sum of the matching degrees of all the characters in the number plate area in the plurality of recognition results; ciA division frame for the ith character in the number plate area; (x)i,yi) Is divided into a division frame CiThe center position coordinates of (a); n is the number of characters contained in a preset number plate area; fscore{Ci(xi,yi) Is to use a preset character recognizer pair CiThe characters contained in the data are identified to obtain a matching degree value;the sum of the matching degrees of all characters in the number plate area corresponding to each recognition result is obtained;
the preferred identification result acquisition subunit is used for acquiring SmaxThe corresponding recognition result is used as the number plate recognition result.
Preferably, the number plate recognition device further comprises:
the preferred recognition result judging unit is used for judging whether the sum of the matching degrees corresponding to the characters in the number plate area corresponding to the preferred recognition result is greater than a preset first threshold value or not before outputting the preferred recognition result as the number plate recognition result;
a scaling frequency judging unit, configured to judge whether the frequency of scaling the number plate area is greater than a preset second threshold when a sum of matching degrees corresponding to characters in the number plate area corresponding to the preferred recognition result is not greater than a preset first threshold;
the number plate area zooming unit is used for zooming the number plate area by using a first preset proportion when the number plate area zooming times are not more than a preset second threshold value, calling the identification area determining unit, the character identification unit, the matching degree sum calculating unit, the parameter adjusting unit and the preferred identification result obtaining unit on the zoomed number plate area, and obtaining a preferred identification result in the zoomed area;
the output unit is further configured to output the preferred recognition result as a number plate recognition result when the sum of the matching degrees corresponding to the characters in the number plate area corresponding to the preferred recognition result is greater than a preset first threshold;
the output unit is further configured to obtain a preferred recognition result with a highest sum of matching degrees of each character in the number plate from the multiple preferred recognition results when the number of times of scaling the number plate area reaches a preset second threshold.
Preferably, the number plate area scaling unit includes: the number plate area scaling subunit and the preset condition scaling subunit are connected;
the number plate area zooming subunit is used for zooming the number plate area by using a first preset proportion;
the preset condition scaling subunit is configured to scale the preset condition by using a first preset ratio.
According to the technical scheme, the embodiment of the invention provides a method and a device for identifying a number plate, which comprises the steps of firstly obtaining a number plate area in an image; determining the recognition area of each character in the number plate area according to the recognition area parameters of the preset characters in the number plate area and the relation parameters among the characters; recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: identifying each character and the matching degree corresponding to each character; calculating the sum of the matching degrees corresponding to all characters in the number plate area; in a preset range, adjusting recognition area parameters and relationship parameters among characters of preset characters in the number plate area according to preset traversal conditions, returning to the step of determining the recognition areas of the characters until values in the preset range are traversed according to a preset traversal algorithm, and obtaining recognition results after the recognition area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time and the sum of the calculated matching degrees corresponding to the characters in the number plate area; obtaining a recognition result with the highest matching degree sum corresponding to each character in the number plate area from the sum of the matching degrees corresponding to each character in all the calculated number plate areas as a preferred recognition result; and outputting the preferred identification result as a number plate identification result. Therefore, in the embodiment, when the number plate is identified, the character identifier is adopted, and the identification area of each character in the number plate is determined according to the identification area parameters of the preset characters and the relation parameters among the characters, so that the problem that the image processing step is interfered by noise in the character identification process in the prior art is solved, and the stability and the anti-interference performance of the system are improved. Of course, it is not necessary for any product or method of practicing the invention to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for identifying a number plate according to an embodiment of the present invention;
FIG. 2a is a schematic flow chart of another method for identifying license plates according to an embodiment of the present invention;
FIG. 2b is a number plate area image;
fig. 3 is a schematic structural diagram of a license plate identification device according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of another license plate identification device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a method and a device for identifying a number plate, wherein in the scheme, a number plate area in an image is obtained; determining the recognition area of each character in the number plate area according to the recognition area parameters of the preset characters in the number plate area and the relation parameters among the characters; recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: identifying each character and the matching degree corresponding to each character; calculating the sum of the matching degrees corresponding to all characters in the number plate area; in a preset range, adjusting recognition area parameters and relationship parameters among characters of preset characters in the number plate area according to preset traversal conditions, returning to the step of determining the recognition areas of the characters until values in the preset range are traversed according to a preset traversal algorithm, and obtaining recognition results after the recognition area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time and the sum of the calculated matching degrees corresponding to the characters in the number plate area; obtaining a recognition result with the highest matching degree sum corresponding to each character in the number plate area from the sum of the matching degrees corresponding to each character in all the calculated number plate areas as a preferred recognition result; and outputting the preferred identification result as a number plate identification result.
The present invention will be described in detail below with reference to specific examples.
Fig. 1 is a schematic flow chart of a method for identifying a number plate according to an embodiment of the present invention, where the method includes the following steps:
step 101: a number plate area in the image is acquired.
In this embodiment, the method for acquiring the number plate area in the image may be: the method comprises a detection method based on character texture features, a detection method based on Adaboost or a detection method based on SVM. Of course, the present application is only described by way of example, and the method for acquiring the number plate area in the image in practical application is not limited to this.
Step 102: and determining the recognition area of each character in the number plate area according to the recognition area parameters of the preset characters in the number plate area and the relation parameters among the characters.
It should be noted that the preset character may be any one of the characters in the number plate area.
In this embodiment, the presetting of the recognition area parameters of the characters may include:
and coordinate values (x, y) of the central point of the segmentation frame of the preset character in the number plate area.
It should be noted that, in practical application, the identification region parameter of the preset character may also be a coordinate value of an upper left corner and a coordinate value of an upper right corner of a segmentation frame of the preset character in the number plate region.
The dividing frame may be: the width is a first preset value, and the ratio of the width to the height is a rectangle of 1: 2.
The inter-character relation parameter may include:
the width value w between two adjacent characters in the number plate area;
and the value k of the width coefficient between two adjacent characters preset in the number plate area.
It should be noted that, the present application is only described by way of example, and in practical applications, the recognition area parameter, the segmentation frame, and the relationship parameter between the characters of the preset character are not limited thereto.
In this embodiment, determining the recognition area of each character in the number plate area according to the recognition area parameter of the preset character in the number plate area and the relationship parameter between each character may include:
obtaining the coordinate value (x) of the central point of the segmentation frame of the preset character in the number plate areapre1，ypre1）；
Obtaining the width value w between two adjacent characters in the number plate areapre1；
Obtaining the value k of the width coefficient between two adjacent characters preset in the number plate areapre1；
According to x = xpre1+(n-npre1)*wpre1Or x = xpre1+(n-npre1)*wpre1±(wpre1*(kpre1-1)) to determine the coordinates (x, y) of the centre point of the segmentation box for the other characters in the number platepre1) (ii) a N ispre1Presetting the serial numbers of characters in the number plate area, wherein n is the serial numbers of other characters in the number plate area;
and determining the recognition area of each character in the number plate area by taking the coordinate of the central position of the segmentation frame of each character in the number plate area as a reference.
Step 103: recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: and identifying each character and the matching degree corresponding to each character.
In this embodiment, the preset character recognizer may be one of the following, but is not limited to the following: a support vector machine based recognizer, a neural network based recognizer, or a deep learning based recognizer.
Step 104: and calculating the sum of the matching degrees corresponding to the characters in the number plate area.
Step 105: and adjusting the identification area parameters of the preset characters and the relationship parameters among the characters in the number plate area within a preset range, returning to the step 102 until values in the preset range are traversed according to a preset traversal algorithm, and obtaining the identification result after the identification area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time, and the sum of the matching degrees corresponding to the characters in the number plate area is calculated.
In this embodiment, the preset range may include: the variation range of x and y in the coordinate value (x, y) of the center point of the divided frame of the preset character, the variation range of the width value w between two adjacent characters and the variation range of the width coefficient value k between two adjacent characters are preset.
In this embodiment, traversing values within a preset range according to a preset traversal algorithm includes:
traversing values in a variation range of x and y, a variation range of a width value w between two adjacent characters and a variation range of a width coefficient value k between two adjacent characters in the coordinate values (x, y) of the central point of the segmentation frame of the preset character according to a preset traversal algorithm.
The preset traversal algorithm is as follows: gradient descent or newton's method or exhaustive method.
Step 106: and acquiring a recognition result with the highest matching degree sum corresponding to each character in the number plate area from the calculated matching degree sum corresponding to each character in the number plate area as a preferred recognition result.
In this embodiment, obtaining the recognition result with the highest sum of the matching degrees corresponding to the characters in the number plate area as the preferred recognition result may include:
according toAcquiring the highest value of the sum of the matching degrees corresponding to all characters in the number plate area; wherein,
Smaxthe highest value of the sum of the matching degrees of all the characters in the number plate area in the plurality of recognition results; ciA division frame for the ith character in the number plate area; (x)i,yi) Is divided into a division frame CiThe center position coordinates of (a); n is the number of characters contained in a preset number plate area; fscore{Ci(xi,yi) Is to use a preset character recognizer pair CiThe characters contained in the data are identified to obtain a matching degree value;the sum of the matching degrees of all characters in the number plate area corresponding to each recognition result is obtained;
will SmaxThe corresponding recognition result is used as the number plate recognition result.
Step 107: and outputting the preferred identification result as a number plate identification result.
As can be seen from the above, in the embodiment, when the number plate is identified, the character identifier is adopted, and the identification area of each character in the number plate is determined through the identification area determination of the preset character and the relation parameter between the characters, so that the problem that the image processing step is interfered by noise in the character identification process is avoided, and the system stability and the anti-interference performance are improved.
Fig. 2a is a schematic flow chart of another method for identifying a number plate according to an embodiment of the present invention, the method including the following steps:
step 201: a number plate area in the image is acquired.
Step 202: and determining the recognition area of each character in the number plate area according to the recognition area parameters of the preset characters in the number plate area and the relation parameters among the characters.
Step 203: recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: and identifying each character and the matching degree corresponding to each character.
Step 204: and calculating the sum of the matching degrees corresponding to the characters in the number plate area.
Step 205: and adjusting the identification area parameters of the preset characters and the relationship parameters among the characters in the number plate area within a preset range, returning to the step 202 until values within the preset range are traversed according to a preset traversal algorithm, obtaining the identification result after the identification area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time, and calculating the sum of the matching degrees corresponding to the characters in the number plate area.
Step 206: and acquiring a recognition result with the highest matching degree sum corresponding to each character in the number plate area from the calculated matching degree sum corresponding to each character in the number plate area as a preferred recognition result.
It should be noted that, in this embodiment, steps 201 to 206 may be identical to steps 101 to 106 in the embodiment shown in fig. 1, and are not repeated here.
Step 207: and judging whether the sum of the matching degrees corresponding to the characters in the number plate area corresponding to the preferred identification result is greater than a preset first threshold, if so, performing step 211, and otherwise, performing step 208.
Step 208: and judging whether the number of times of zooming the number plate area is greater than a preset second threshold value, if not, performing step 209, otherwise, performing step 210.
Step 209: and (3) scaling the number plate area by using a first preset proportion, and returning to the step (202) to obtain a preferred identification result on the scaled number plate area.
In this embodiment, scaling the number plate area by a first preset ratio may include: scaling the number plate area by a first preset proportion; and scaling the preset range by a first preset proportion.
Step 210: and acquiring a preferred recognition result with the highest sum of the matching degrees of all characters in the number plate from a plurality of preferred recognition results.
Step 211: and outputting the preferred identification result as a number plate identification result.
As can be seen from the above, in the embodiment, when the number plate is identified, the character identifier is adopted, and the identification area of each character in the number plate is determined according to the identification area determination parameter of the preset character and the relation parameter between the characters, so that the problem that the image processing step is interfered by noise in the character identification process is avoided, and the system stability and the anti-interference performance are improved.
The present application will be described in detail with reference to specific examples.
Referring to fig. 2b, it is assumed that the width coefficient value between the second character and the third character preset in the number plate region is k, the width value is k × w, and the width between two adjacent characters is w.
Step 201: the number plate area shown in figure 2b is obtained.
Step 202: suppose that the fourth character in the preset number plate region, namely the division box C corresponding to the character "24The center point coordinates (x, y) of (1) are (219, 69), the width value w between two adjacent characters in the number plate area is 9, the width coefficient value k between two adjacent characters is 1.1, and then the center point coordinates of the division boxes of other characters in the number plate area are respectively: a first character (219 + (1-4) × 9-9 × (1.1-1), 69), a second character (219 + (2-4) × 9-9 × (1.1-1), 69), a third character (219 + (3-4) × 9, 69), a fifth character (219 + (5-4) × 9, 69), a sixth character (219 + (6-4) × 9, 69), a seventh character (219 + (7-4) × 9, 69), and assuming that the frame is a rectangle with a width of A and an aspect ratio of 1:2, the recognition region of each character, i.e., the frame C, is determined by the coordinate value of the center point of the frame of each character and the width and the height of the frame1、C2……C7The corresponding area.
Step 203: assuming that the preset character recognizer is a recognizer based on a support vector machine, recognizing the recognition area of each character by using the recognizer to obtain each recognized character and the matching degree S corresponding to each character1、S2……S7。
Step 204: calculating the sum of the matching degrees corresponding to each character S = S1+S2+……+S7。
Step 205: assuming that the variation range of x and y in the coordinate values (x, y) of the center point of the division frame of the fourth character in the number plate region is: x e [219,221], y e [69,71], the variation range of the width value w between two adjacent characters is: w ∈ [9,11], the range of variation of the value of width coefficient k between two adjacent characters is: k belongs to [1.1,1.3], and assuming that the preset traversal algorithm is an exhaustion method, traversing all combinations of { x, y, w, k } in the preset range according to the exhaustion method, repeating the steps 202 to 204 for corresponding values of each group of { x, y, w, k }, and obtaining the recognition result of each character in the number plate region corresponding to each group of { x, y, w, k } and the sum S ', S ' ' … … of the matching degree corresponding to each character. In the traversal process, it is assumed that the change step of x, y and w is 1, and the change step of k is 0.1.
Step 206: obtaining the highest value S of S and S ', S' … …maxAnd taking the character recognition result corresponding to the highest value as the preferred recognition result.
Step 207: judging the sum S of the matching degrees corresponding to each character in the number plate area corresponding to the preferred identification resultmaxWhether it is greater than a preset first threshold value X. Suppose, SmaxIf less than X, proceed to step 208.
Step 208: assuming that the number plate area has been scaled 3 times, it is determined whether the number plate area has been scaled more than a preset second threshold Y, assuming that Y is equal to 5. Since 3 is less than 5, step 209 is performed.
Step 209: zooming the number plate area by using a preset first proportion Z, zooming the preset range by using the first proportion Z, returning to the step 202, acquiring a preferable identification result on the zoomed area, and repeating the steps 202 to 209 until S is acquiredmaxThe recognition result or the number of times of scaling larger than the first threshold value X is larger than the second threshold value.
And when the scaling times are larger than a second threshold value, acquiring a preferred recognition result with the highest sum of the matching degrees of all the characters in the number plate from a plurality of preferred recognition results.
When obtaining SmaxAnd when the identification result is larger than the first threshold value X, outputting the preferred identification result as a number plate identification result.
As can be seen from the above, in the embodiment, when identifying the number plate, the identifier based on the support vector machine is adopted, and the identification area of each character in the number plate is determined through the coordinate value of the center point of the segmentation frame of the preset character, the width between adjacent characters and the width coefficient between adjacent characters, so that the problem that the image processing step is interfered by noise in the character identification process is avoided, and the stability and the anti-interference performance of the system are improved.
Fig. 3 is a schematic structural diagram of a number plate identification device according to an embodiment of the present invention, where the number plate identification device includes: a number plate area acquisition unit 301, a recognition area determination unit 302, a character recognition unit 303, a matching degree sum calculation unit 304, a parameter adjustment unit 305, a preferred recognition result acquisition unit 306, and an output unit 307.
The number plate area acquiring unit 301 is configured to acquire a number plate area in an image;
a recognition area determining unit 302, configured to determine a recognition area of each character in the number plate area according to a recognition area parameter of a preset character in the number plate area and a relationship parameter between each character;
a character recognition unit 303, configured to recognize a recognition area of each character by using a preset character recognizer, and obtain a recognition result; the recognition result comprises: identifying each character and the matching degree corresponding to each character;
a matching degree sum calculating unit 304, configured to calculate a sum of matching degrees corresponding to each character in the number plate area;
a parameter adjusting unit 305, configured to adjust a recognition area parameter and a relationship parameter between characters of a preset character in the number plate area within a preset range, and repeatedly call the recognition area determining unit 302, the character recognition unit 303, and the matching degree sum calculating unit 304 until values in the preset range are traversed according to a preset traversal algorithm, so as to obtain a recognition result after the recognition area parameter and the relationship parameter between characters of the preset character in the number plate area are adjusted each time, and a calculated matching degree sum corresponding to each character in the number plate area;
a preferred recognition result obtaining unit 306, configured to obtain, as a preferred recognition result, a recognition result with a highest sum of matching degrees corresponding to each character in the number plate area from the sum of matching degrees corresponding to each character in all the calculated number plate areas;
an output unit 307 for outputting the preferred recognition result as a number plate recognition result.
In this embodiment, the identification area determining unit 302 is configured to determine an identification area of each character in the number plate area according to a central point coordinate value (x, y) of a segmentation frame of a preset character in the number plate area, a width value w between two adjacent characters, and a width coefficient value k between two adjacent preset characters;
the parameter adjusting unit 305 may include a preset range acquiring subunit (not shown in the figure). The preset range acquiring subunit is configured to acquire a variation range of x and y in the coordinate values (x, y) of the center point of the segmentation frame of the preset character, a variation range of a width value w between two adjacent characters, and a variation range of a width coefficient value k between two adjacent characters.
In this embodiment, the identifying area determining unit 302 may include: a coordinate acquiring subunit, a width coefficient acquiring subunit, a coordinate calculating subunit, and an identification area determining subunit (not shown in the figure). The coordinate acquiring subunit is used for acquiring the coordinate value (x) of the central point of the segmentation frame of the preset character in the number plate areapre1，ypre1) (ii) a The width acquisition subunit is used for acquiring the width value w between two adjacent characters in the number plate areapre1(ii) a The width coefficient acquisition subunit is used for acquiring a width coefficient value k between two adjacent characters preset in the number plate areapre1(ii) a The coordinate calculation subunit is used for calculating the coordinate according to x = xpre1+(n-npre1)*wpre1Or
x=xpre1+(n-npre1)*wpre1±(wpre1*(kpre1-1)) to determine the coordinates (x, y) of the centre point of the segmentation box for the other characters in the number platepre1) (ii) a N ispre1Presetting the serial numbers of characters in the number plate area, wherein n is the serial numbers of other characters in the number plate area; and the identification area determining subunit is used for determining the identification area of each character in the number plate area by taking the coordinate of the central position of the segmentation frame of each character in the number plate as the standard.
In this embodiment, the preferred recognition result obtaining unit 306 may include: a highest matching degree sum calculating subunit and a preferred recognition result obtaining subunit (not shown in the figure). The highest matching degree sum calculating subunit is used for obtaining the highest value of the matching degree sum corresponding to each character in the number plate area, wherein,Smaxthe highest value of the sum of the matching degrees of all the characters in the number plate area in the plurality of recognition results; ciA division frame for the ith character in the number plate area; (x)i,yi) Is divided into a division frame CiThe center position coordinates of (a); n is the number of characters contained in a preset number plate area; fscore{Ci(xi,yi) Is to use a preset character recognizer pair CiThe characters contained in the data are identified to obtain a matching degree value;the sum of the matching degrees of all characters in the number plate area corresponding to each recognition result is obtained; the preferred identification result acquisition subunit is used for acquiring SmaxThe corresponding recognition result is used as the number plate recognition result.
As can be seen from the above, in the embodiment, when the number plate is identified, the character identifier is adopted, and the identification area of each character in the number plate is determined according to the identification area determination parameter of the preset character and the relation parameter between the characters, so that the problem that the image processing step is interfered by noise in the character identification process is avoided, and the system stability and the anti-interference performance are improved.
Fig. 4 is a schematic structural diagram of another license plate identification apparatus provided in an embodiment of the present invention, where the apparatus includes: a number plate region acquisition unit 401, a recognition region determination unit 402, a character recognition unit 403, a matching degree sum calculation unit 404, a parameter adjustment unit 405, a preferred recognition result acquisition unit 406, a preferred recognition result determination unit 407, a scaling number determination unit 408, a number plate region scaling unit 409, and an output unit 410.
It should be noted that, in the present embodiment, the number plate region acquisition unit 401, the recognition region determination unit 402, the character recognition unit 403, the matching degree sum calculation unit 404, the parameter adjustment unit 405, the preferred recognition result acquisition unit 406, and the output unit 410 may be completely the same as the number plate region acquisition unit 301, the recognition region determination unit 302, the character recognition unit 303, the matching degree sum calculation unit 304, the parameter adjustment unit 305, the preferred recognition result acquisition unit 306, and the output unit 307 in the embodiment shown in fig. 3, respectively, and are not repeated here.
A preferred recognition result judging unit 407, configured to judge whether a sum of matching degrees corresponding to characters in the number plate area corresponding to a preferred recognition result is greater than a preset first threshold value before outputting the preferred recognition result as a number plate recognition result;
a scaling frequency judging unit 408, configured to judge whether the frequency of scaling the number plate region is greater than a preset second threshold when the sum of the matching degrees corresponding to the characters in the number plate region corresponding to the preferred recognition result is not greater than a preset first threshold;
a number plate area scaling unit 409, configured to scale the number plate area by using a first preset proportion when the number plate area scaling frequency is not greater than a preset second threshold, and call the identification area determining unit, the character identification unit, the matching degree sum calculating unit, the parameter adjusting unit, and the preferred identification result obtaining unit on the scaled number plate area, so as to obtain a preferred identification result in the scaled area;
the output unit 410 is further configured to output the preferred recognition result as a license plate recognition result when a sum of matching degrees corresponding to each character in the license plate area corresponding to the preferred recognition result is greater than a preset first threshold.
The output unit 410 is further configured to, when the number of times of scaling the number plate area reaches a preset second threshold, obtain, from the multiple preferred recognition results, a preferred recognition result with a highest sum of matching degrees of the characters in the number plate.
In this embodiment, the number plate area scaling unit 409 may include: a number plate area scaling sub-unit and a preset condition scaling sub-unit (not shown in the figure). The number plate area zooming subunit is used for zooming the number plate area by using a first preset proportion; the preset condition scaling subunit is configured to scale the preset condition by using a first preset ratio.
As can be seen from the above, in the embodiment, when the number plate is identified, the character identifier is adopted, the problem that the image processing step is interfered by noise in the character identification process is avoided, the identification area of each character in the number plate is determined according to the identification area determination parameter of the preset character and the relation parameter between the characters, and the system stability and the anti-interference performance are improved.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Those skilled in the art will appreciate that all or part of the steps in the above method embodiments may be implemented by a program to instruct relevant hardware to perform the steps, and the program may be stored in a computer-readable storage medium, which is referred to herein as a storage medium, such as: ROM/RAM, magnetic disk, optical disk, etc.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A method of license plate identification, the method comprising:
step A, acquiring a number plate area in an image;
b, determining the recognition area of each character in the number plate area according to the recognition area parameters of the preset characters in the number plate area and the relation parameters among the characters;
step C, recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: identifying each character and the matching degree corresponding to each character;
d, calculating the sum of the matching degrees corresponding to all the characters in the number plate area;
step E, in a preset range, adjusting the identification area parameters and the relationship parameters among the characters of the preset characters in the number plate area according to a preset traversal algorithm, returning to the step B until values in the preset range are traversed according to the preset traversal algorithm, and obtaining the identification result after the identification area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time and the sum of the calculated matching degrees corresponding to the characters in the number plate area;
step F, obtaining the recognition result with the highest matching degree sum corresponding to each character in the number plate area from the matching degree sum corresponding to each character in all the calculated number plate areas as the preferred recognition result;
g, outputting the preferred identification result as a license plate identification result;
the identification region parameters of the preset characters comprise: coordinate values (x, y) of the central point of a segmentation frame of a preset character in the number plate area;
the dividing frame is as follows: the width of the rectangle is a first preset value, and the ratio of the width to the height of the rectangle is 1: 2;
the inter-character relation parameter comprises: the width value w between two adjacent characters in the number plate area; a width coefficient value k between two adjacent characters preset in the number plate area;
the preset range comprises: the variation range of x and y in the coordinate values (x, y) of the central point of the segmentation frame of the preset character; the variation range of the width value w between the two adjacent characters; the preset variation range of the width coefficient value k between two adjacent characters;
the determining the recognition area of each character in the number plate area according to the recognition area parameter and the relation parameter between the characters of the preset character in the number plate area comprises the following steps:
obtaining the coordinate value (x) of the central point of the segmentation frame of the preset character in the number plate areapre1，ypre1)；
Obtaining between two adjacent characters in the number plate areaWidth value wpre1；
Obtaining the value k of the width coefficient between two adjacent characters preset in the number plate areapre1；
According to x ═ xpre1+(n-npre1)*wpre1Or x ═ xpre1+(n-npre1)*wpre1±(wpre1*(kpre1-1)) to determine the coordinates (x, y) of the centre point of the segmentation box for the other characters in the number platepre1) (ii) a N ispre1Presetting the serial numbers of characters in the number plate area, wherein n is the serial numbers of other characters in the number plate area;
and determining the recognition area of each character in the number plate area by taking the coordinate of the central position of the segmentation frame of each character in the number plate area as a reference.
2. The method according to claim 1, wherein traversing values within a preset range according to a preset traversal algorithm is performed by:
traversing values in the variation range of x and y, the variation range of the width value w between the two adjacent characters and the variation range of the width coefficient value k between the two adjacent characters in the coordinate values (x, y) of the central point of the segmentation frame of the preset character according to a preset traversal algorithm;
the preset traversal algorithm comprises the following steps: gradient descent or newton's method or exhaustive method.
3. The method according to any one of claims 1 or 2, wherein the obtaining of the recognition result with the highest sum of the matching degrees corresponding to the characters in the number plate area as the preferred recognition result comprises:
according toAcquiring the highest value of the sum of the matching degrees corresponding to all characters in the number plate area; wherein,
Smaxthe highest value of the sum of the matching degrees of all the characters in the number plate area in the plurality of recognition results; ciIs a number plate areaA dividing frame of the ith character; (x)i,yi) Is divided into a division frame CiThe center position coordinates of (a); n is the number of characters contained in a preset number plate area; fscore{Ci(xi,yi) Is to use a preset character recognizer pair CiThe characters contained in the data are identified to obtain a matching degree value;the sum of the matching degrees of all characters in the number plate area corresponding to each recognition result is obtained;
will SmaxThe corresponding recognition result is used as the number plate recognition result.
4. The method according to any one of claims 1 or 2, wherein the preset character recognizer is: a support vector machine based recognizer or a neural network based recognizer or a deep learning based recognizer.
5. The method of claim 1, wherein before outputting the preferred recognition result as a number plate recognition result, the method further comprises:
step H, judging whether the sum of the matching degrees corresponding to the characters in the number plate area corresponding to the optimal recognition result is larger than a preset first threshold value or not;
step I, if the value is larger than the value, executing step G, otherwise, executing step J;
step J, judging whether the times of zooming the number plate area is larger than a preset second threshold value or not;
step K, if the number plate area is not larger than the preset number plate area, zooming the number plate area by using a first preset proportion, returning to the step B, and otherwise, executing the step L;
and L, acquiring a preferred recognition result with the highest sum of the matching degrees of all characters in the number plate from a plurality of preferred recognition results.
6. The method of claim 5, wherein the scaling the number plate area by a first preset scale comprises:
scaling the number plate area by a first preset proportion; and
the preset range is scaled by a first preset ratio.
7. An apparatus for license plate identification, the apparatus comprising:
a number plate area acquisition unit for acquiring a number plate area in the image;
the identification area determining unit is used for determining the identification area of each character in the number plate area according to the identification area parameter of the preset character in the number plate area and the relation parameter among the characters;
the character recognition unit is used for recognizing the recognition area of each character by using a preset character recognizer to obtain a recognition result; the recognition result comprises: identifying each character and the matching degree corresponding to each character;
the matching degree sum calculating unit is used for calculating the sum of the matching degrees corresponding to all the characters in the number plate area;
the parameter adjusting unit is used for adjusting the recognition area parameters and the relationship parameters among the characters of the preset characters in the number plate area within a preset range, and repeatedly calling the recognition area determining unit, the character recognition unit and the matching degree sum calculating unit until values within the preset range are traversed according to a preset traversal algorithm, so that a recognition result obtained after the recognition area parameters and the relationship parameters among the characters of the preset characters in the number plate area are adjusted each time and the calculated matching degree sum corresponding to each character in the number plate area is obtained;
the preferred recognition result obtaining unit is used for obtaining a recognition result with the highest matching degree sum corresponding to each character in the number plate area from the matching degree sum corresponding to each character in all the calculated number plate areas as a preferred recognition result;
an output unit for outputting the preferred recognition result as a number plate recognition result;
the identification area determining unit is used for determining the identification area of each character in the number plate area according to the coordinate value (x, y) of the central point of the segmentation frame of the preset character in the number plate area, the width value w between two adjacent characters and the width coefficient value k between two adjacent preset characters;
the parameter adjusting unit comprises a preset range acquiring subunit;
the preset range acquiring subunit is configured to acquire a variation range of x and y in coordinate values (x, y) of a center point of a segmentation frame of a preset character, a variation range of a width value w between two adjacent characters, and a variation range of a width coefficient value k between two adjacent characters;
the identification area determination unit includes: the system comprises a coordinate acquisition subunit, a width coefficient acquisition subunit, a coordinate calculation subunit and an identification area determination subunit;
the coordinate acquiring subunit is used for acquiring the coordinate value (x) of the central point of the segmentation frame of the preset character in the number plate areapre1，ypre1)；
The width acquisition subunit is used for acquiring the width value w between two adjacent characters in the number plate areapre1；
The width coefficient acquisition subunit is used for acquiring a width coefficient value k between two adjacent characters preset in the number plate areapre1；
The coordinate calculation subunit is used for calculating the coordinate according to x ═ xpre1+(n-npre1)*wpre1Or
x＝xpre1+(n-npre1)*wpre1±(wpre1*(kpre1-1)) to determine the coordinates (x, y) of the centre point of the segmentation box for the other characters in the number platepre1) (ii) a N ispre1Presetting the serial numbers of characters in the number plate area, wherein n is the serial numbers of other characters in the number plate area;
and the identification area determining subunit is used for determining the identification area of each character in the number plate area by taking the coordinate of the central position of the segmentation frame of each character in the number plate as the standard.
8. The apparatus of claim 7, wherein the preferred recognition result obtaining unit comprises: a highest matching degree sum calculation subunit and an optimal recognition result acquisition subunit;
the highest matching degree sum calculating subunit is used for obtaining the highest value of the matching degree sum corresponding to each character in the number plate area, wherein,
Smaxthe highest value of the sum of the matching degrees of all the characters in the number plate area in the plurality of recognition results; ciA division frame for the ith character in the number plate area; (x)i,yi) Is divided into a division frame CiThe center position coordinates of (a); n is the number of characters contained in a preset number plate area; fscore{Ci(xi,yi) Is to use a preset character recognizer pair CiThe characters contained in the data are identified to obtain a matching degree value;the sum of the matching degrees of all characters in the number plate area corresponding to each recognition result is obtained;
the preferred identification result acquisition subunit is used for acquiring SmaxThe corresponding recognition result is used as the number plate recognition result.
9. The apparatus of claim 7, further comprising:
the preferred recognition result judging unit is used for judging whether the sum of the matching degrees corresponding to the characters in the number plate area corresponding to the preferred recognition result is greater than a preset first threshold value or not before outputting the preferred recognition result as the number plate recognition result;
a scaling frequency judging unit, configured to judge whether the frequency of scaling the number plate area is greater than a preset second threshold when a sum of matching degrees corresponding to characters in the number plate area corresponding to the preferred recognition result is not greater than a preset first threshold;
the number plate area zooming unit is used for zooming the number plate area by using a first preset proportion when the number plate area zooming times are not more than a preset second threshold value, calling the identification area determining unit, the character identification unit, the matching degree sum calculating unit, the parameter adjusting unit and the preferred identification result obtaining unit on the zoomed number plate area, and obtaining a preferred identification result in the zoomed area;
the output unit is further configured to output the preferred recognition result as a number plate recognition result when the sum of the matching degrees corresponding to the characters in the number plate area corresponding to the preferred recognition result is greater than a preset first threshold;
the output unit is further configured to obtain a preferred recognition result with a highest sum of matching degrees of each character in the number plate from the multiple preferred recognition results when the number of times of scaling the number plate area reaches a preset second threshold.
10. The apparatus of claim 9, wherein the number plate area scaling unit comprises: the number plate area scaling subunit and the preset condition scaling subunit are connected;
the number plate area zooming subunit is used for zooming the number plate area by using a first preset proportion;
the preset condition scaling subunit is configured to scale the preset condition by using a first preset ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310564073.9A CN104636748B (en) | 2013-11-14 | 2013-11-14 | A kind of method and device of number plate identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310564073.9A CN104636748B (en) | 2013-11-14 | 2013-11-14 | A kind of method and device of number plate identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104636748A CN104636748A (en) | 2015-05-20 |
| CN104636748B true CN104636748B (en) | 2018-08-17 |
Family
ID=53215476
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310564073.9A Active CN104636748B (en) | 2013-11-14 | 2013-11-14 | A kind of method and device of number plate identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104636748B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105975968B (en) * | 2016-05-06 | 2019-03-26 | 西安理工大学 | A kind of deep learning license plate character recognition method based on Caffe frame |
| CN106295645B (en) * | 2016-08-17 | 2019-11-29 | 东方网力科技股份有限公司 | A kind of license plate character recognition method and device |
| CN108319951B (en) * | 2017-01-16 | 2020-10-02 | 杭州海康威视数字技术股份有限公司 | Method and device for recognizing characters in license plate |
| CN110135416B (en) * | 2018-02-09 | 2021-06-04 | 杭州海康威视数字技术股份有限公司 | License plate recognition method and device |
| CN109448142A (en) * | 2018-11-09 | 2019-03-08 | 郑州云海信息技术有限公司 | A kind of intelligent high-speed charging system and method based on cloud computing |
| CN110502990B (en) * | 2019-07-17 | 2022-06-03 | 上海展湾信息科技有限公司 | Method and system for data acquisition by image processing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096821A (en) * | 2010-04-15 | 2011-06-15 | 西安理工大学 | Number plate identification method under strong interference environment on basis of complex network theory |
| CN102663377A (en) * | 2012-03-15 | 2012-09-12 | 华中科技大学 | Character recognition method based on template matching |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6026177A (en) * | 1995-08-29 | 2000-02-15 | The Hong Kong University Of Science & Technology | Method for identifying a sequence of alphanumeric characters |
| CN102722707A (en) * | 2012-06-11 | 2012-10-10 | 复旦大学 | License plate character segmentation method based on connected region and gap model |
| CN103065137B (en) * | 2012-12-30 | 2016-05-25 | 四川九洲电器集团有限责任公司 | A kind of license plate character recognition method |
-
2013
- 2013-11-14 CN CN201310564073.9A patent/CN104636748B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096821A (en) * | 2010-04-15 | 2011-06-15 | 西安理工大学 | Number plate identification method under strong interference environment on basis of complex network theory |
| CN102663377A (en) * | 2012-03-15 | 2012-09-12 | 华中科技大学 | Character recognition method based on template matching |
Non-Patent Citations (1)
| Title |
|---|
| 最大类间方差车牌字符分割的模板匹配算法;王兴玲;《计算机工程》;20061031;第32卷(第19期);正文前言，第1.3节-第2.2节，附图1，3，4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104636748A (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104636748B (en) | A kind of method and device of number plate identification | |
| CN104834922B (en) | Gesture identification method based on hybrid neural networks | |
| CN104063712B (en) | A kind of information of vehicles extracting method and system | |
| CN107392139B (en) | Lane line detection method based on Hough transform and terminal equipment | |
| WO2016138838A1 (en) | Method and device for recognizing lip-reading based on projection extreme learning machine | |
| CN105354866A (en) | Polygon contour similarity detection method | |
| CN113537316B (en) | Vehicle detection method based on 4D millimeter wave radar point cloud | |
| WO2016065701A1 (en) | Image text recognition method and device | |
| CN108305260B (en) | Method, device and equipment for detecting angular points in image | |
| CN107103326A (en) | The collaboration conspicuousness detection method clustered based on super-pixel | |
| CN103473537B (en) | A kind of target image contour feature method for expressing and device | |
| CN104049760B (en) | The acquisition methods and system of a kind of man-machine interaction order | |
| CN103996195A (en) | Image saliency detection method | |
| CN113569968B (en) | Model training method, target detection method, device, equipment and storage medium | |
| CN106650660A (en) | Vehicle type recognition method and terminal | |
| CN109934077B (en) | Image identification method and electronic equipment | |
| CN106157283A (en) | The detection method of lane segmentation thing and device | |
| CN112085701A (en) | Face ambiguity detection method and device, terminal equipment and storage medium | |
| WO2021051939A1 (en) | Document area positioning method and device | |
| CN103971112A (en) | Image feature extracting method and device | |
| CN109949324B (en) | Contour detection method based on non-classical receptive field nonlinear subunit response | |
| CN112052782A (en) | Around-looking-based parking space identification method, device, equipment and storage medium | |
| CN103246877B (en) | Based on the recognition of face novel method of image outline | |
| Cai et al. | Vehicle Detection Based on Deep Dual‐Vehicle Deformable Part Models | |
| CN111027534A (en) | Compact double-license-plate detection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230330 Address after: 518000 Room 1304, University Town Entrepreneurship Park, Lishan Road, Taoyuan Street, Nanshan District, Shenzhen, Guangdong Province Patentee after: SHENZHEN SINO SMART AUTOMOBILE TECHNOLOGY CO.,LTD. Address before: 100089 Wanliu Xingbiao Jiayuan 12-1709, Haidian District, Beijing Patentee before: Zhang Weiwei |