CN108664978B - Character segmentation method and device for fuzzy license plate - Google Patents
Character segmentation method and device for fuzzy license plate Download PDFInfo
- Publication number
- CN108664978B CN108664978B CN201710203959.9A CN201710203959A CN108664978B CN 108664978 B CN108664978 B CN 108664978B CN 201710203959 A CN201710203959 A CN 201710203959A CN 108664978 B CN108664978 B CN 108664978B
- Authority
- CN
- China
- Prior art keywords
- license plate
- character
- point
- character segmentation
- interval
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Character Input (AREA)
Abstract
The invention provides a character segmentation method and device for a fuzzy license plate. The method comprises the following steps: preprocessing a fuzzy license plate to obtain a binary image of the fuzzy license plate; calculating a histogram of the binary image; converting the histogram into a corresponding numerical axis graph; determining character segmentation points of the fuzzy license plate according to the numerical axis graph; and segmenting the fuzzy license plate according to the character segmentation points. The method and the device can improve the accuracy of fuzzy license plate character segmentation.
Description
Technical Field
The invention relates to the technical field of license plate recognition, in particular to a character segmentation method and device for a fuzzy license plate.
Background
In a license plate recognition system in the traffic field, most license plate character segmentation algorithms adopt a projection analysis method at present, license plate characters are binarized firstly by the character segmentation method, and then binary images are projected in the horizontal direction, the character segmentation method has higher requirements on the definition of characters in shot images, and the segmentation effect is better for clear license plates with clear character intervals. However, for the fuzzy license plate, because many characters in the fuzzy license plate are adhered together, and the interval between two characters in the horizontal direction after the binary image is projected is not 0, the characters cannot be correctly segmented, so that the existing license plate character segmentation method is frequently wrong when the characters of the fuzzy license plate are segmented, and the accuracy is low.
Disclosure of Invention
In view of this, the present invention provides a method and an apparatus for segmenting characters of a fuzzy license plate, which can improve the accuracy of segmenting characters of the fuzzy license plate.
In a first aspect, the present invention provides a method for segmenting characters of a fuzzy license plate, including:
preprocessing a fuzzy license plate to obtain a binary image of the fuzzy license plate;
calculating a histogram of the binary image;
converting the histogram into a corresponding numerical axis graph;
determining character segmentation points of the fuzzy license plate according to the numerical axis graph;
and segmenting the fuzzy license plate according to the character segmentation points.
Optionally, the determining the character segmentation points of the fuzzy license plate according to the numerical axis graph comprises:
zeroing the ordinate of a point of which the ordinate is smaller than a first threshold value in the numerical axis curve graph to obtain a plurality of non-zero curves;
judging whether the interval length of each non-zero curve is greater than a second threshold value, and searching candidate segmentation points of the non-zero curves when the interval length of the non-zero curves is greater than the second threshold value;
estimating the character width in the interval according to the interval length of the non-zero curve;
and determining the character segmentation points of the interval by using the abscissa of the candidate segmentation points and the estimated character width.
Optionally, the candidate segmentation points of the non-zero curve include: the interval demarcation point and the effective valley point of the non-zero curve.
Optionally, the determining the character segmentation point of the fuzzy license plate according to the numerical axis graph further comprises:
and when the interval length of the non-zero curve is less than or equal to a second threshold value, determining the interval boundary point as the character segmentation point of the interval.
In a second aspect, the present invention provides a device for segmenting characters of a fuzzy license plate, comprising:
the system comprises a preprocessing module, a storage module and a display module, wherein the preprocessing module is used for preprocessing a fuzzy license plate to obtain a binary image of the fuzzy license plate;
the calculation module is used for calculating a histogram of the binary image;
the graph conversion module is used for converting the histogram into a corresponding numerical axis curve graph;
the character segmentation point determination module is used for determining character segmentation points of the fuzzy license plate according to the numerical axis curve graph;
and the character segmentation module is used for segmenting the fuzzy license plate according to the character segmentation points.
Optionally, the character segmentation point determination module includes:
the curve optimization unit is used for setting the vertical coordinate of a point with the vertical coordinate smaller than a first threshold value in the numerical axis curve graph to zero to obtain a plurality of non-zero curves;
the judging unit is used for judging whether the interval length of each non-zero curve is greater than a second threshold value;
the searching unit is used for searching candidate segmentation points of the non-zero curve when the interval length of the non-zero curve is larger than a second threshold value;
the estimation unit is used for estimating the character width in the interval according to the interval length of the non-zero curve;
and the first determining unit is used for determining the character segmentation points of the interval by using the abscissa of the candidate segmentation point and the estimated character width.
Optionally, the searching unit is configured to search an interval boundary point and a valid valley point of the non-zero curve.
Optionally, the character segmentation point determination module further includes:
and the second determining unit is used for determining the section boundary point as the character segmentation point of the section when the section length of the non-zero curve is less than or equal to a second threshold value.
The invention provides a character segmentation method and a character segmentation device for a fuzzy license plate, which are characterized in that the fuzzy license plate is preprocessed to obtain a binary image of the fuzzy license plate; then calculating a histogram of the binary image; then converting the histogram into a corresponding numerical axis curve graph and determining character segmentation points of the fuzzy license plate according to the numerical axis curve graph; and finally, segmenting the fuzzy license plate according to the character segmentation points. Compared with the prior art, the character segmentation method for the fuzzy license plate can accurately determine character segmentation points and improve the accuracy of character segmentation of the fuzzy license plate.
Drawings
FIG. 1 is a flowchart illustrating a method for segmenting characters of a fuzzy license plate according to an embodiment of the present invention;
FIG. 2 is a fuzzy license plate gray scale map of a character to be segmented according to an embodiment of the present invention;
FIG. 3 is a binarized image of the grayscale image of FIG. 2;
FIG. 4 is a schematic diagram illustrating the effect of the binarized image shown in FIG. 3 after being processed by erosion expansion;
FIG. 5 is a histogram of the binarized image shown in FIG. 4;
FIG. 6 is a graph of the number axis of the histogram conversion of FIG. 5;
FIG. 7 is a graph illustrating the optimized effect of the several-axis graph shown in FIG. 6;
FIG. 8 is a schematic view of a particular valley point;
FIG. 9 is a schematic diagram illustrating a segmentation effect of the blurred license plate gray-scale image shown in FIG. 2;
FIG. 10 is a schematic structural diagram of a fuzzy license plate character segmentation apparatus according to an embodiment of the present invention;
FIG. 11 is a block diagram of a character segmentation point determination module in the apparatus shown in FIG. 10;
fig. 12 is another structural diagram of the character segmentation point determination module in the apparatus shown in fig. 10.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 character segmentation method for a fuzzy license plate, which comprises the following steps of:
and S11, preprocessing the blurred license plate to obtain a binary image of the blurred license plate.
The pretreatment in this embodiment includes:
converting the collected original color license plate image into a gray image, then performing horizontal correction and vertical correction to ensure that no ill-conditioned image exists, performing normalization processing on the license plate character image, uniformly scaling the license plate character image to a fixed size such as 136 × 36, width 136 and height 36, and simultaneously processing the boundary of the license plate character image, wherein the processed license plate character image has no upper and lower frames, and the left and right boundaries do not exceed one character width, as shown in fig. 2.
And then, selecting the average number of all pixels of the license plate character image as a threshold value, and performing binarization processing on the image to obtain a binarized image, as shown in fig. 3.
The binarized image was subjected to erosion dilation processing using an erosion dilation kernel of 3 × 3, and the binarized image after the processing is shown in fig. 4.
And S12, calculating a histogram of the binary image.
Projecting the binarized image to the x-axis direction along the y-axis direction in an image coordinate system, calculating the accumulated sum of all position pixels of the y-axis at each x-axis position, dividing the accumulated sum by the height 36 of the license plate character image to obtain the average value of the x-axis position, and making the average value into a histogram, as shown in fig. 5, wherein the x-axis represents the position of the horizontal coordinate of the pixel of the binarized image, and the y-axis represents the pixel average value.
The formula for calculating the generated histogram is as follows:
let the picture be a ═ a(i,j)]36×136,a(i,j)Indicating the pixel value, (i, j) indicating the coordinate at which the pixel value lies, i indicating the x-axis coordinate value, j indicating the y-axis coordinate value, 36 x 136 indicating a height of 36 and a length of 136.
Let the cumulative sum array be B ═ B(i)]136,
b(i)＝∑a(i,j)i∈[0,35],j∈[0,135] (1)
Let the average value of each x-axis position be Bavg ═ c(i)]136,
c(i)＝b(i)÷36 (2)
Is represented by the formulas (1) and (2)
Bavg＝[(∑a(i,j))÷36]136 i∈[0,35],j∈[0,135] (3)
Equation (3) is the equation for calculating the histogram.
And S13, converting the histogram into a corresponding numerical axis graph.
According to the histogram shown in fig. 5, it is possible to convert into a more vivid numerical axis graph as shown in fig. 6, in which the x-axis represents the position of the abscissa of the pixel of the binarized image and the y-axis represents the pixel average value.
And S14, determining character segmentation points of the fuzzy license plate according to the numerical axis graph.
According to fig. 6, the curves in the graph are optimized, that is, the ordinate of the point whose ordinate is smaller than the first threshold Th1 (for example, set to 16) in fig. 6 is reset to 0, so as to obtain a plurality of non-zero curves, as shown in fig. 7, including 4 non-zero curves, and the start point coordinate and the end point coordinate of each non-zero curve on the horizontal axis are respectively recorded, so as to obtain the interval length of each non-zero curve.
Judging whether the interval length of each non-zero curve is greater than a second threshold value Th2, wherein the second threshold value Th2 is related to a character width empirical value, and in the embodiment of the invention, the second threshold value Th2 is 21.
When the interval length of the non-zero curve is larger than a second threshold Th2, searching for a candidate segmentation point of the non-zero curve. For each non-zero curve with the length larger than the second threshold, searching a local minimum value, namely a valley point, respectively calculating the absolute value of the difference between the valley point and peak points (local maximum values) on two sides, and summing the two absolute values, wherein if the calculation result (namely the sum of the two absolute values) is larger than a third threshold Th3, in the embodiment of the invention, 34 is taken as the third threshold Th3, the valley point is a valid valley point, and the valid valley point is taken as a candidate segmentation point, otherwise, the valley point is an invalid valley point and is not a candidate segmentation point of the character segmentation point.
The section boundary points of the non-zero curve are also treated as effective valley points.
For some special valley points, there may be only one side peak point, as shown in fig. 8, the valley point a has only one side peak point, when it is determined whether the valley point is valid, only the absolute value of the difference between the valley point and the one side peak point is calculated, when the absolute value is greater than a fourth threshold Th4, the valley point is valid, otherwise, the valley point is invalid, where Th4 is 1/2 Th 3.
As can be seen from the above, for a non-zero curve with a section length greater than the second threshold, the candidate segmentation points include: the interval demarcation point and the effective valley point of the non-zero curve.
In fig. 7, the bottom points 14 and 8 are invalid bottom points by calculation, and the sum of the absolute values of the differences between the bottom points and the peak points on both sides calculated by these two bottom points is smaller than the third threshold Th3(34), and is therefore an invalid bottom point.
The valley points 2, 7,9, 10 are all valid valley points, all in line with the sum of the absolute values of the differences between the valley point and the peak points on both sides being greater than a third threshold (34), and thus being valid valley points.
The section boundary points 1, 3, 4,5, 6, 11, 12, and 13 are also treated as valid valley points.
Therefore, the candidate segmentation points in fig. 7 include: 1. 2,3, 4,5, 6,7, 9, 10, 11, 12, 13.
After the candidate segmentation point is determined, estimating the character width in the interval according to the interval length of the non-zero curve, wherein the specific method comprises the following steps:
assuming that the length of the interval is char _ dist, the minimum value of the character width in the interval is tmp _ avg _ s, and the maximum value of the character width in the interval is tmp _ avg _ b, the estimation formula of the character width is:
tmp_avg_s＝(char_dist/num1-tmp1)>11？(char_dist/num1-tmp1):11；
meaning if (char _ dist/num1-tmp1) >11
tmp _ avg _ s ═ (char _ dist/num1-tmp1), otherwise tmp _ avg _ s ═ 11. Where num1 represents the number of characters (which may be a non-integer and empirical data), and tmp1 represents the correction width coefficient (empirical data).
tmp_avg_b＝(char_dist/num2+tmp2)>21？(char_dist/num2+tmp2):21；
Meaning, if (char _ dist/num2+ tmp2) >21, tmp _ avg _ b is (char _ dist/num2+ tmp2), otherwise tmp _ avg _ s is 21. Where num2 represents the number of characters (which may be a non-integer and empirical data), and tmp2 represents the correction width coefficient (empirical data).
In the formula, 11 is an empirical value of the minimum width of a character, and 21 is an empirical value of the maximum width of a character.
In the embodiment of the invention, a segmented estimation formula is adopted to estimate the character width, and the method specifically comprises the following steps:
a) the length char _ dist of the segment is in the range of (21, 43), and the character width of the segment is determined by the following method.
tmp_avg_s＝(char_dist/2.2-8)>11？(char_dist/2.2-8):11；
tmp_avg_b＝(char_dist/1.9+8)>21？(char_dist/1.9+8):21；
b) The length char _ dist of the segment is in the range of (43, 61), and the character width of the segment is determined by the following method.
tmp_avg_s＝(char_dist/3.2-8)>11？(char_dist/3.2-8):11；
tmp_avg_b＝(char_dist/2.8+8)>21？(char_dist/2.8+8):21；
c) The length char _ dist of the segment is in the range of (61, 74), and the character width of the segment is determined by the following method.
tmp_avg_s＝(char_dist/4.4-8)>11？(char_dist/4.4-8):11；
tmp_avg_b＝(char_dist/4+8)>21？(char_dist/4+8):21；
d) The length char _ dist of the segment is in the range of (74, 97), and the character width of the segment is determined by the following method.
tmp_avg_s＝(char_dist/5-8)>11？(char_dist/5-8):11；
tmp_avg_b＝(char_dist/5+9)>21？(char_dist/5+9):21；
e) The segment length char _ dist is in the range of (97,136), and the character width of the segment is determined by the following method.
tmp_avg_s＝(char_dist/6-8)>11？(char_dist/6-8):11；
tmp_avg_b＝(char_dist/6+9)>21？(char_dist/6+9):21；
And finally, determining the character segmentation points of the interval from the candidate segmentation points by using the estimated character width. The length between any two character segmentation points must be within the estimated character width.
Specifically, when the section length of the non-zero curve is less than or equal to a second threshold, the section boundary point is directly taken as the character dividing point of the section.
Further, taking the numerical axis graph shown in fig. 7 as an example, it is described which candidate division points are character division points.
The segment including the division candidate points 1,2, and 3 has a segment length of 33 since the abscissa of the point 1 is 1, the abscissa of the point 2 is 16, and the abscissa of the point 3 is 34, and the character width range calculated from the estimation formula is (11, 26) in the range of (21, 43), and the character division points of the segment can be determined to be 1,2, and 3 using the abscissas of 1,2, and 3 and the estimated character width range.
The segment including the division candidate points 4 and 5 has an abscissa of 41 for the point 4 and an abscissa of 59 for the point 5, and therefore the segment length is 18 and is smaller than the second threshold (21), and therefore the segment is one character width and the character division points are 4 and 5.
A section including division candidate points 6,7, 9, 10, and 11, where the abscissa of the point 6 is 62, the abscissa of the point 7 is 81, the abscissa of the point 9 is 98, the abscissa of the point 10 is 105, and the abscissa of the point 11 is 116, and therefore the section length is 54, and in the range of (43, 61), the character width range calculated according to the estimation formula is (11, 21), and using the abscissas of 6,7, 9, 10, and 11 and the estimated character width range, it can be determined that the character division points of the section are 6,7, 9, and 11, and 10 is not a character division point.
The interval including the division candidate points 12,13, in which the abscissa of the point 12 is 118 and the abscissa of the point 13 is 135, is 17, and is smaller than the second threshold (21), and is therefore one character width, and the character division points are 12, 13.
Therefore, the character segmentation points in fig. 7 include: 1. 2,3, 4,5, 6,7, 9,11, 12, 13.
And S15, segmenting the fuzzy license plate according to the character segmentation points, wherein the segmented characters are respectively (1,2), (2,3), (4,5), (6,7), (7,9), (9,11), (12,13), as shown in FIG. 9.
The character segmentation method of the fuzzy license plate provided by the embodiment of the invention comprises the steps of preprocessing the fuzzy license plate to obtain a binary image of the fuzzy license plate; then calculating a histogram of the binary image; then converting the histogram into a corresponding numerical axis curve graph and determining character segmentation points of the fuzzy license plate according to the numerical axis curve graph; and finally, segmenting the fuzzy license plate according to the character segmentation points. Compared with the prior art, the character segmentation method for the fuzzy license plate can accurately determine character segmentation points and improve the accuracy of character segmentation of the fuzzy license plate.
An embodiment of the present invention further provides a device for segmenting characters of a fuzzy license plate, as shown in fig. 10, the device includes:
the system comprises a preprocessing module 101, a storage module and a display module, wherein the preprocessing module 101 is used for preprocessing a fuzzy license plate to obtain a binary image of the fuzzy license plate;
a calculating module 102, configured to calculate a histogram of the binarized image;
a graph conversion module 103, configured to convert the histogram into a corresponding numerical axis graph;
the character segmentation point determination module 104 is used for determining character segmentation points of the fuzzy license plate according to the numerical axis curve graph;
and the character segmentation module 105 is used for segmenting the fuzzy license plate according to the character segmentation points.
Alternatively, as shown in fig. 11, the character segmentation point determination module 104 includes:
a curve optimization unit 1041, configured to set a vertical coordinate of a point, in the log graph, where the vertical coordinate is smaller than the first threshold to zero, to obtain a plurality of non-zero curves;
a determining unit 1042, configured to determine whether an interval length of each non-zero curve is greater than a second threshold;
a searching unit 1043, configured to search a candidate segmentation point of the non-zero curve when an interval length of the non-zero curve is greater than a second threshold;
an estimating unit 1044, configured to estimate a character width in the interval according to the length of the interval of the non-zero curve;
a first determining unit 1045, configured to determine a character segmentation point of the interval by using the abscissa of the candidate segmentation point and the estimated character width.
Optionally, the searching unit 1043 is configured to search an interval boundary point and an effective valley point of the non-zero curve.
Optionally, as shown in fig. 12, the character segmentation point determining module 104 further includes:
a second determining unit 1046, configured to determine, when a length of the interval of the non-zero curve is less than or equal to a second threshold, the interval boundary point as a character segmentation point of the interval.
The character segmentation device for the fuzzy license plate provided by the embodiment of the invention is used for preprocessing the fuzzy license plate to obtain a binary image of the fuzzy license plate; then calculating a histogram of the binary image; then converting the histogram into a corresponding numerical axis curve graph and determining character segmentation points of the fuzzy license plate according to the numerical axis curve graph; and finally, segmenting the fuzzy license plate according to the character segmentation points. Compared with the prior art, the character segmentation device for the fuzzy license plate can accurately determine character segmentation points and improve the accuracy of character segmentation of the fuzzy license plate.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. A character segmentation method for a fuzzy license plate is characterized by comprising the following steps:
preprocessing a fuzzy license plate to obtain a binary image of the fuzzy license plate;
calculating a histogram of the binary image;
converting the histogram into a corresponding numerical axis graph;
determining character segmentation points of the fuzzy license plate according to the numerical axis graph;
segmenting the fuzzy license plate according to the character segmentation points;
wherein the determining character segmentation points of the fuzzy license plate according to the log graph comprises:
zeroing the ordinate of a point of which the ordinate is smaller than a first threshold value in the numerical axis curve graph to obtain a plurality of non-zero curves;
judging whether the interval length of each non-zero curve is greater than a second threshold value, and searching candidate segmentation points of the non-zero curves when the interval length of the non-zero curves is greater than the second threshold value;
estimating the character width in the interval according to the interval length of the non-zero curve;
and determining the character segmentation points of the interval by using the abscissa of the candidate segmentation points and the estimated character width.
2. The method of claim 1, wherein the candidate segmentation points of the non-zero curve comprise: the interval demarcation point and the effective valley point of the non-zero curve.
3. The method of claim 1, wherein determining the character segmentation points of the fuzzy license plate from the log graph further comprises:
and when the interval length of the non-zero curve is less than or equal to a second threshold value, determining the interval boundary point as the character segmentation point of the interval.
4. A character segmentation device for a fuzzy license plate is characterized by comprising:
the system comprises a preprocessing module, a storage module and a display module, wherein the preprocessing module is used for preprocessing a fuzzy license plate to obtain a binary image of the fuzzy license plate;
the calculation module is used for calculating a histogram of the binary image;
the graph conversion module is used for converting the histogram into a corresponding numerical axis curve graph;
the character segmentation point determination module is used for determining character segmentation points of the fuzzy license plate according to the numerical axis curve graph;
the character segmentation module is used for segmenting the fuzzy license plate according to the character segmentation points;
wherein the character segmentation point determination module includes:
the curve optimization unit is used for setting the vertical coordinate of a point with the vertical coordinate smaller than a first threshold value in the numerical axis curve graph to zero to obtain a plurality of non-zero curves;
the judging unit is used for judging whether the interval length of each non-zero curve is greater than a second threshold value;
the searching unit is used for searching candidate segmentation points of the non-zero curve when the interval length of the non-zero curve is larger than a second threshold value;
the estimation unit is used for estimating the character width in the interval according to the interval length of the non-zero curve;
and the first determining unit is used for determining the character segmentation points of the interval by using the abscissa of the candidate segmentation point and the estimated character width.
5. The apparatus of claim 4, wherein the searching unit is configured to search an interval boundary point and a valid valley point of the non-zero curve.
6. The apparatus of claim 4, wherein the character segmentation point determination module further comprises:
and the second determining unit is used for determining the section boundary point as the character segmentation point of the section when the section length of the non-zero curve is less than or equal to a second threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710203959.9A CN108664978B (en) | 2017-03-30 | 2017-03-30 | Character segmentation method and device for fuzzy license plate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710203959.9A CN108664978B (en) | 2017-03-30 | 2017-03-30 | Character segmentation method and device for fuzzy license plate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108664978A CN108664978A (en) | 2018-10-16 |
| CN108664978B true CN108664978B (en) | 2021-10-26 |
Family
ID=63786387
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710203959.9A Active CN108664978B (en) | 2017-03-30 | 2017-03-30 | Character segmentation method and device for fuzzy license plate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108664978B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101154271A (en) * | 2006-09-30 | 2008-04-02 | 电子科技大学中山学院 | License plate character segmentation method based on fast area labeling algorithm and license plate large-spacing locating method |
| CN103324930A (en) * | 2013-06-28 | 2013-09-25 | 浙江大学苏州工业技术研究院 | License plate character segmentation method based on grey level histogram binaryzation |
| CN104992175A (en) * | 2015-07-10 | 2015-10-21 | 河海大学常州校区 | Over-merged steel billet character segmentation method based on tree type relationship |
| CN105260735A (en) * | 2015-10-10 | 2016-01-20 | 安徽清新互联信息科技有限公司 | License number character segmentation method based on typical position |
| US20160292528A1 (en) * | 2015-03-31 | 2016-10-06 | Nidec Sankyo Corporation | Character segmentation device, character recognition device and character segmentation method |
| CN106127207A (en) * | 2016-08-12 | 2016-11-16 | 万永秀 | Registration number character dividing method |
| CN106203417A (en) * | 2016-07-12 | 2016-12-07 | 北京科技大学 | A kind of adhesion character alienable RMB crown word number identification method |
| CN106529592A (en) * | 2016-11-07 | 2017-03-22 | 湖南源信光电科技有限公司 | License plate recognition method based on mixed feature and gray projection |
-
2017
- 2017-03-30 CN CN201710203959.9A patent/CN108664978B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101154271A (en) * | 2006-09-30 | 2008-04-02 | 电子科技大学中山学院 | License plate character segmentation method based on fast area labeling algorithm and license plate large-spacing locating method |
| CN103324930A (en) * | 2013-06-28 | 2013-09-25 | 浙江大学苏州工业技术研究院 | License plate character segmentation method based on grey level histogram binaryzation |
| US20160292528A1 (en) * | 2015-03-31 | 2016-10-06 | Nidec Sankyo Corporation | Character segmentation device, character recognition device and character segmentation method |
| CN104992175A (en) * | 2015-07-10 | 2015-10-21 | 河海大学常州校区 | Over-merged steel billet character segmentation method based on tree type relationship |
| CN105260735A (en) * | 2015-10-10 | 2016-01-20 | 安徽清新互联信息科技有限公司 | License number character segmentation method based on typical position |
| CN106203417A (en) * | 2016-07-12 | 2016-12-07 | 北京科技大学 | A kind of adhesion character alienable RMB crown word number identification method |
| CN106127207A (en) * | 2016-08-12 | 2016-11-16 | 万永秀 | Registration number character dividing method |
| CN106529592A (en) * | 2016-11-07 | 2017-03-22 | 湖南源信光电科技有限公司 | License plate recognition method based on mixed feature and gray projection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108664978A (en) | 2018-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3309703B1 (en) | Method and system for decoding qr code based on weighted average grey method | |
| CN106683073B (en) | License plate detection method, camera and server | |
| US9613266B2 (en) | Complex background-oriented optical character recognition method and device | |
| US20060245650A1 (en) | Precise grayscale character segmentation apparatus and method | |
| CN111340749B (en) | Image quality detection method, device, equipment and storage medium | |
| CN111723634B (en) | Image detection method and device, electronic equipment and storage medium | |
| CN109729298B (en) | Image processing method and image processing apparatus | |
| CN111950394A (en) | Method and device for predicting lane change of vehicle and computer storage medium | |
| US20170017836A1 (en) | Character segmenting apparatus, character recognition apparatus, and character segmenting method | |
| WO2018025336A1 (en) | Degradation detection device, degradation detection method, and computer-readable recording medium | |
| CN112508913A (en) | Cable section edge detection method based on image detection | |
| JP2011165170A (en) | Object detection device and program | |
| CN112101139B (en) | Human shape detection method, device, equipment and storage medium | |
| CN106951831B (en) | Pedestrian detection tracking method based on depth camera | |
| CN111062415B (en) | Target object image extraction method and system based on contrast difference and storage medium | |
| CN117218672A (en) | Deep learning-based medical records text recognition method and system | |
| Boiangiu et al. | Handwritten documents text line segmentation based on information energy | |
| CN108664978B (en) | Character segmentation method and device for fuzzy license plate | |
| Chang et al. | An efficient thresholding algorithm for degraded document images based on intelligent block detection | |
| KR20090083184A (en) | Method for dectecting front and rear vehicle by using image | |
| CN107886524B (en) | Method for identifying motion track of elevator door | |
| CN116188826A (en) | Template matching method and device under complex illumination condition | |
| CN113538337B (en) | Detection method, detection device and computer readable storage medium | |
| CN115376106A (en) | Vehicle type identification method, device, equipment and medium based on radar map | |
| Yuan et al. | A robust vanishing point estimation method for lane detection |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |