CN114419635A - Electronic seal vector diagram identification method based on graphic identification - Google Patents
Electronic seal vector diagram identification method based on graphic identification Download PDFInfo
- Publication number
- CN114419635A CN114419635A CN202210321079.2A CN202210321079A CN114419635A CN 114419635 A CN114419635 A CN 114419635A CN 202210321079 A CN202210321079 A CN 202210321079A CN 114419635 A CN114419635 A CN 114419635A
- Authority
- CN
- China
- Prior art keywords
- edge
- closed
- value
- combination
- subsequence
- 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.)
- Granted
Links
Images
Abstract
The invention relates to the technical field of data processing, in particular to an electronic seal vector diagram identification method based on image identification, which comprises the following steps: carrying out edge identification on the electronic seal bitmap to obtain an edge image; acquiring a closed edge in an edge image, performing edge segment segmentation on the closed edge, and acquiring a characterization value sequence of each closed edge based on a characterization value of an edge segment line form; for any two closed edges, carrying out periodic continuation on the characteristic numerical value sequences of the two closed edges to obtain periodic sequences of any two closed edges; performing sliding window on the periodic sequences of any two closed edges to obtain a plurality of common subsequence combinations; for any two closed edges, selecting any one corresponding consensus subsequence combination, and forming a combination set by the selected consensus subsequence combination to obtain a plurality of combination sets; and identifying to obtain a vector diagram of the electronic seal based on the optimal combination set. The accuracy of the vector diagram can be ensured when the edge of the electronic seal image is more complex.
Description
Technical Field
The invention relates to the field of data processing, in particular to an electronic seal vector diagram identification method based on image identification.
Background
Computer graphics are generally divided into two broad categories, one being vector graphics and one being bitmap. The bitmap is a computer graph which is composed of pixel blocks and can be represented by a matrix, the vector diagram describes the graph by using straight lines and curved lines, the component elements of the graph are composed of straight lines, curved lines, polygons and arcs, the straight lines, the curved lines, the polygons and the arcs are all obtained through mathematical calculation, therefore, vector image files are generally small in size, the storage mode can be seen as mathematical formulas containing parameters, the maximum advantage is that the image cannot be distorted whether the image is enlarged, reduced or stretched in a rotating mode, and the vector diagram is widely applied to the manufacturing of electronic seals due to the characteristic.
The seal bitmap obtained by scanning is converted into a vector diagram, so that the vector diagram has the characteristic of no distortion, and the definition of the electronic seal is ensured. The existing vector diagram identification method generally utilizes an edge identification algorithm to obtain bitmap edge information of an electronic seal and color information of the edge, utilizes curve fitting to obtain the edge, fills colors in the closed edge, and stores or represents an image by using parameter information of each curve and corresponding color information. Therefore, the accuracy of the curve fitting edge information and the number of curve parameters determine the accuracy and the storage amount of the vector diagram identification result. The existing vector diagram identification method generally calls a database of known characters, graphics and other primitives to perform direct fitting when detecting the edge of the stamp image, and the method depends on the adaptability of the database and the actual image, namely when the edge information of the stamp image is too complex, effective parameter adaptation cannot be performed, and further the accuracy is reduced. When the complexity of image elements is reduced, such as only using the basic image composition elements of straight lines, curves and the like, and the accuracy of the parameters of the fitted edge is improved, the number of the parameters is greatly increased due to the excessive number of the elements, and the storage capacity of the parameters is increased. Therefore, for the vector diagram identification process of the complex electronic seal, an optimal image primitive combination can be selected according to the edge morphological characteristics of different electronic seals so as to ensure the accuracy and the massiveness of the vector diagram identification at the same time.
Disclosure of Invention
In order to solve the above technical problems, the present invention aims to provide an electronic seal vector diagram identification method based on image identification, and the adopted technical scheme is as follows:
one embodiment of the invention provides an electronic seal vector diagram identification method based on image identification, which comprises the following specific steps:
acquiring an electronic seal bitmap, and performing edge identification on the electronic seal bitmap to obtain an edge image;
acquiring closed edges in the edge image, performing edge segment segmentation on the closed edges to acquire a characterization value of an edge segment line form, and acquiring a characterization value sequence of each closed edge based on the characterization value of the edge segment line form;
for any two closed edges, carrying out periodic continuation on the characteristic numerical sequence of any two closed edges based on the least common multiple of the lengths of the characteristic numerical sequences of any two closed edges to obtain the periodic sequence of any two closed edges; performing sliding window on the periodic sequences of any two closed edges to obtain a plurality of common subsequence combinations; wherein, the common characteristic numerical value subsequence of the periodic sequences of any two closed edges in a window forms a common subsequence combination;
for any two closed edges, selecting any one corresponding consensus subsequence combination, and forming a combination set by the selected consensus subsequence combination to obtain a plurality of combination sets; calculating the preference degree of each combination set, and determining an optimal combination set based on the preference degree; and identifying and obtaining a vector diagram of the electronic seal based on the optimal combination set.
Further, the obtaining of the preference of the combination set specifically includes:
the closed edge is a node, and any two nodes are connected to form an undirected graph;
calculating a node value according to the number of pixel points of the node corresponding to the closed edge and the number of edge segments;
acquiring a common subsequence combination of closed edges corresponding to the two nodes based on the combination set, and calculating the weight of a node connecting edge according to the sum of the reproduction times of each characteristic numerical value subsequence in the acquired common subsequence combination and the coverage ratio of the characteristic numerical value subsequence covering the closed edges corresponding to the two nodes by the characteristic numerical value subsequence in the common subsequence combination;
and the sum of the weight of the node connecting edge and the product of the node value mean values of the two nodes corresponding to the node connecting edge is the optimization degree of the combination set.
Further, setting the number of the pixel points and the number of the edge segments, and carrying out weighted summation on the number of the pixel points corresponding to the closed edge and the number of the edge segments by the node to obtain a node value.
Further, setting the sum of the reproduction times and the weight of the coverage ratio, and performing weighted summation on the sum of the reproduction times and the coverage ratio to obtain the weight of the node connecting edge.
Further, performing edge segment segmentation on the closed edge, specifically:
each edge pixel on the closed edge is sequentially a pixel to be marked, the direction of the adjacent edge pixel on one side of the pixel to be marked pointing to the pixel to be marked is a first direction, the direction of the pixel to be marked pointing to the adjacent edge pixel on the other side is a second direction, and the pixel to be marked is marked according to the rotating direction and the rotating angle of the pixel to be marked which are rotated to the second direction from the first direction;
and performing edge segment segmentation on the closed edge at the position where the mark is changed.
Further, the line shape comprises a straight line and a curve, the characteristic value of the straight line is a first value, and the characteristic value of the curve is a second value.
The embodiment of the invention at least has the following beneficial effects: the invention can ensure the accuracy of the vector diagram when the edge of the electronic seal image is more complex. In addition, the number of the pixel points and the weight of the number of the edge segments can be adjusted according to the requirements on the vectorization speed and the storage volume, so that a corresponding optimal combination set is obtained, and a vector diagram of the electronic seal is obtained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of 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 other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of the steps of an embodiment of the method of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description, the structure, the features and the effects of the vector diagram recognition method for an electronic seal based on the graphic recognition according to the present invention are provided with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the electronic seal vector diagram identification method based on the image identification in detail with reference to the accompanying drawings.
Referring to fig. 1, a flowchart illustrating steps of a vector diagram recognition method for an electronic seal based on image recognition according to an embodiment of the present invention is shown, where the method includes the following steps:
acquiring an electronic seal bitmap, and performing edge identification on the electronic seal bitmap to obtain an edge image;
acquiring closed edges in the edge image, performing edge segment segmentation on the closed edges to acquire a characterization value of an edge segment line form, and acquiring a characterization value sequence of each closed edge based on the characterization value of the edge segment line form;
for any two closed edges, carrying out periodic continuation on the characteristic numerical sequence of any two closed edges based on the least common multiple of the lengths of the characteristic numerical sequences of any two closed edges to obtain the periodic sequence of any two closed edges; performing sliding window on the periodic sequences of any two closed edges to obtain a plurality of common subsequence combinations; wherein, the common characteristic numerical value subsequence of the periodic sequences of any two closed edges in a window forms a common subsequence combination;
for any two closed edges, selecting any one corresponding consensus subsequence combination, and forming a combination set by the selected consensus subsequence combination to obtain a plurality of combination sets; calculating the preference degree of each combination set, and determining an optimal combination set based on the preference degree; and identifying and obtaining a vector diagram of the electronic seal based on the optimal combination set.
The following steps are specifically developed:
step one, acquiring an electronic seal bitmap, and carrying out edge recognition on the electronic seal bitmap to obtain an edge image.
And acquiring an electronic seal bitmap, carrying out edge identification on the electronic seal bitmap, and determining pixel filling values on two sides of an edge. Because most of electronic seals are binary images, only specific filling values on two sides of the marked edge are needed. Preferably, in the embodiment, the edge recognition is performed on the electronic seal bitmap based on the sobel operator.
And secondly, acquiring closed edges in the edge image, performing edge segment segmentation on the closed edges to acquire the characterization values of the edge segment line morphology, and acquiring the characterization value sequence of each closed edge based on the characterization values of the edge segment line morphology.
(1) And acquiring a closed edge in the edge image.
Each edge pixel on the edge image is respectively used as a seed point to search edge pixels in the neighborhood, if the edge pixels are searched, two edge pixels are combined to be an example, and all the edge pixels forming a closed edge are classified into the same example. Based on this, several closed edges in the edge image can be obtained.
(2) Performing edge segment segmentation on the closed edge: each edge pixel on the closed edge is sequentially a pixel to be marked, the direction of the adjacent edge pixel on one side of the pixel to be marked pointing to the pixel to be marked is a first direction, the direction of the pixel to be marked pointing to the adjacent edge pixel on the other side is a second direction, and the pixel to be marked is marked according to the rotating direction and the rotating angle of the pixel to be marked which are rotated to the second direction from the first direction; and performing edge segment segmentation on the closed edge at the position where the mark is changed.
One edge pixel on the closed edge is selected as a pixel to be marked
Selecting the position in its neighborhood
One side edge pixel is marked as
(two in total, one of them is selectedOne) of the two or more of the above-described elements,
point of direction
Is a first direction, and then selects a location in its neighborhood
Edge pixels on the other side are denoted as
，
Only edge pixels in one, the other neighborhood are already selected
，
Point of direction
The direction of the marking is a second direction, and the marking is carried out on the pixel to be marked according to the rotation direction and the rotation angle from the first direction to the second direction.
As an example, marking the pixels to be marked specifically includes: if the first direction is rotated clockwise and the angle is rotated
A second direction can be obtained when the angle is less than 180 DEG, then
The first flag value is a first flag value, if the first direction is rotated counterclockwise and the rotation angle is rotated
Is less thanA second direction can be obtained at 180 DEG, then
Is a second mark value, if the first direction and the second direction are the same, then
Is the third marker value.
As an example, marking the pixels to be marked specifically includes: if the first direction is rotated counterclockwise and the rotation angle is rotated
A second direction can be obtained when the angle is less than 180 DEG, then
The first flag value is a first flag value, if the first direction is rotated clockwise and the rotation angle is larger than a second flag value
A second direction can be obtained when the angle is less than 180 DEG, then
Is a second mark value, if the first direction and the second direction are the same, then
Is the third marker value.
The specific marking mode can be determined by an implementer, and only the marking values of the straight line pixels and the curve pixels are different, and the marking values of the concave curve pixels and the convex curve pixels are different.
After marking edge pixels on the closed edge, acquiring the positions where the marks change, and regarding the position where each mark changes, taking the midpoint of the connecting line of the edge pixels with different marks corresponding to the position as a dividing point; edge segment segmentation is performed on the closed edge based on a number of segmentation points.
(3) And acquiring a characterization value of the line morphology of the edge segment.
The line form comprises a straight line and a curve, the characteristic value of the straight line is a first value, and the characteristic value of the curve is a second value. As an example, the first value is 0 and the second value is 1.
In one embodiment, the edge segment is input into a line shape recognition network to obtain the shape of the edge segment.
In another embodiment, the rotation angle of the first direction to the second direction corresponding to all edge pixels on the edge segment is calculated
The specific mode of the skewness of (2) is as follows: all will be
Accumulation, corresponding to the first embodiment of labeling pixels to be labeled, clockwise
And marking as positive, and marking as negative anticlockwise, wherein if the accumulation result is less than 45 degrees, the line shape of the edge segment is a straight line, the characterization value is a first value, if the accumulation result is more than 45 degrees, the line shape of the edge segment is a curve, and the characterization value is a second value.
(4) And acquiring a characteristic value sequence of each closed edge based on the characteristic values of the edge segment line morphology.
And for a closed edge, selecting an edge segment with a curve line form as a starting edge segment, and acquiring the characterization values of the edge segments in a clockwise direction to obtain a characterization value sequence of the closed edge. Wherein, the characteristic numerical value sequence of the closed edge is a sequence consisting of 0 or 1.
Step three, for any two closed edges, performing periodic continuation on the characteristic value sequences of any two closed edges based on the least common multiple of the lengths of the characteristic value sequences of any two closed edges to obtain the periodic sequences of any two closed edges; performing sliding window on the periodic sequences of any two closed edges to obtain a plurality of common subsequence combinations; and the characteristic value subsequence shared by the periodic sequences of any two closed edges in a window forms a shared subsequence combination.
(1) And for any two closed edges, performing periodic continuation on the characteristic value sequences of any two closed edges based on the least common multiple of the lengths of the characteristic value sequences of any two closed edges to obtain the periodic sequences of any two closed edges.
For example, the sequence of characterizing values for the closed edge a is: 11100010100, the sequence of characteristic values for the closed edge B is: 1111001010, aligning the two characterization value sequences according to bit and performing cycle extension according to cycle extension to obtain cycle sequences of any two closed edges; specifically, the length of the characteristic value sequence of the closed edge a is 11, the length of the characteristic value sequence of the closed edge B is 10, and the least common multiple is 110, so that when the lengths of the two sequences are extended to 110 bits, the heads and the tails of the two periodic sequences are completely aligned, and the subsequent extension is meaningless.
(2) Performing sliding window on the periodic sequences of any two closed edges to obtain a plurality of common subsequence combinations; and the characteristic value subsequence shared by the periodic sequences of any two closed edges in a window forms a shared subsequence combination.
Setting a window length, specifically, setting the window length as a smaller value of two characteristic value sequence lengths corresponding to any two closed edges, taking the closed edge a and the closed edge B as an example, setting the window length as 10, performing a sliding window on the periodic sequence of any two closed edges, subtracting the characteristic values in the sliding window according to positions, wherein the subtraction result is 0 or 1 or-1, obtaining a subtraction result sequence, wherein all parts (0 appearing continuously twice or more) which are continuously 0 and are not adjacent in the subtraction result sequence are characteristic value subsequences, and the characteristic value subsequences obtained based on one window form a common subsequence combination; for example, within a window, the token value subsequences in a consensus subsequence corresponding to closure edge a and closure edge B are 111, 001010. Wherein the length of the characteristic numerical value subsequence is greater than or equal to 2. And performing sliding window on the periodic sequences of any two closed edges to obtain a plurality of windows, and correspondingly obtaining a plurality of consensus subsequence combinations.
Thus, several consensus subsequence combinations corresponding to any two closed edges are obtained.
Step four, selecting any one corresponding common subsequence combination for any two closed edges, and forming a combination set by the selected common subsequence combinations to obtain a plurality of combination sets; calculating the preference degree of each combination set, and determining an optimal combination set based on the preference degree; and identifying and obtaining a vector diagram of the electronic seal based on the optimal combination set.
(1) And for any two closed edges, selecting any one corresponding consensus subsequence combination, and forming a combination set by the selected consensus subsequence combination to obtain a plurality of combination sets.
For any two closed edges, a plurality of common subsequence combinations are correspondingly arranged, and a combination set can be obtained based on all closed edges by selecting any one of the common subsequence combinations; and if the selection of the common subsequence combinations corresponding to any two closed edges is different, the obtained combination sets are different, and then a plurality of combination sets are obtained.
(2) The preference of each combination set is calculated.
The closed edge is a node, and any two nodes are connected to form an undirected graph; calculating a node value according to the number of pixel points of the node corresponding to the closed edge and the number of edge segments; acquiring a common subsequence combination of closed edges corresponding to the two nodes based on the combination set, and calculating the weight of a node connecting edge according to the sum of the reproduction times of each characteristic numerical value subsequence in the acquired common subsequence combination and the coverage ratio of the characteristic numerical value subsequence covering the closed edges corresponding to the two nodes by the characteristic numerical value subsequence in the common subsequence combination; and the sum of the weight of the node connecting edge and the product of the node value mean values of the two nodes corresponding to the node connecting edge is the optimization degree of the combination set.
In one embodiment, the calculating the node value according to the number of the pixel points of the node corresponding to the closed edge and the number of the edge segments specifically includes: and the sum of the number of the pixels corresponding to the closed edge and the number of the edge segments of the node is a node value.
In another embodiment, the node value is calculated according to the number of the pixel points of the node corresponding to the closed edge and the number of the edge segments, specifically: and setting the weight values of the number of the pixel points and the number of the edge segments, and carrying out weighted summation on the number of the pixel points corresponding to the closed edge and the number of the edge segments by the node to obtain a node value. Wherein, the sum of the two weights is 1. The reason for setting the weight is as follows: when the algorithm speed is concerned, the more the number of the pixel points of the closed edge is, the higher the requirement on the accuracy of the fitting result is, namely the number of the pixel points of the closed edge is the factor influencing the algorithm speed; when the amount of memory is concerned, the more edge segments that close the edge, the more parameters need to be stored, that is, the factor that affects the amount of memory is mainly the number of edge segments.
In one embodiment, the calculation of the weight of the node connecting edge specifically includes: and calculating the weight of the node connecting edge according to the sum of the recurrence times of each characterization numerical value subsequence in the obtained common subsequence combination and the coverage ratio of the characterization numerical value subsequence covering the two nodes corresponding to the closed edge in the common subsequence combination, wherein the sum of the recurrence times and the coverage ratio is the weight of the node connecting edge.
In another embodiment, the calculation of the weight of the node connecting edge specifically includes: and setting the reproduction times and the weight of the coverage ratio, and carrying out weighted summation on the reproduction times and the coverage ratio to obtain the weight of the node connecting edge. Wherein, the sum of the two weights is 1. The reason for setting the weight is as follows: the more the recurrence times are, the faster the algorithm fitting is, and the more favorable the calculation speed is to be improved; and the common subsequence combination can cover the coverage ratio of the characteristic numerical value sequence, and the larger the coverage ratio is, the more saved storage resources are shown, and the advantage of reducing the storage quantity is realized.
In one embodiment, the obtaining of the coverage ratio of the token numerical sequence covering two nodes corresponding to the closed edge in the common subsequence combination is specifically: the sum of the lengths of the two characteristic numerical sequences corresponding to the two nodes over a doubling of the length of the characteristic numerical subsequence in the consensus subsequence combination is the coverage ratio.
It should be noted that the node value and the inter-node edge weight value need to be normalized.
(3) Determining an optimal combination set based on the preference degree; and identifying and obtaining a vector diagram of the electronic seal based on the optimal combination set.
Each combination set is corresponding to a preference degree, and the combination set corresponding to the maximum value of the preference degrees is an optimal combination set. Identifying to obtain a vector diagram of the electronic seal based on the optimal combination set, wherein a characteristic numerical value subsequence in the optimal combination set is used as a vectorized image element, and the vector diagram of the electronic seal is obtained through identification; specifically vectorization procedures the present invention is not described.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (6)
1. A vector diagram identification method of an electronic seal based on image identification is characterized by comprising the following steps:
acquiring an electronic seal bitmap, and performing edge identification on the electronic seal bitmap to obtain an edge image;
acquiring closed edges in the edge image, performing edge segment segmentation on the closed edges to acquire a characterization value of an edge segment line form, and acquiring a characterization value sequence of each closed edge based on the characterization value of the edge segment line form;
for any two closed edges, carrying out periodic continuation on the characteristic numerical sequence of any two closed edges based on the least common multiple of the lengths of the characteristic numerical sequences of any two closed edges to obtain the periodic sequence of any two closed edges; performing sliding window on the periodic sequences of any two closed edges to obtain a plurality of common subsequence combinations; wherein, the common characteristic numerical value subsequence of the periodic sequences of any two closed edges in a window forms a common subsequence combination;
for any two closed edges, selecting any one corresponding consensus subsequence combination, and forming a combination set by the selected consensus subsequence combination to obtain a plurality of combination sets; calculating the preference degree of each combination set, and determining an optimal combination set based on the preference degree; and identifying and obtaining a vector diagram of the electronic seal based on the optimal combination set.
2. The method according to claim 1, wherein the obtaining of the preference of the combined set specifically comprises:
the closed edge is a node, and any two nodes are connected to form an undirected graph;
calculating a node value according to the number of pixel points of the node corresponding to the closed edge and the number of edge segments;
acquiring a common subsequence combination of closed edges corresponding to the two nodes based on the combination set, and calculating the weight of a node connecting edge according to the sum of the reproduction times of each characteristic numerical value subsequence in the acquired common subsequence combination and the coverage ratio of the characteristic numerical value subsequence covering the closed edges corresponding to the two nodes by the characteristic numerical value subsequence in the common subsequence combination;
and the sum of the weight of the node connecting edge and the product of the node value mean values of the two nodes corresponding to the node connecting edge is the optimization degree of the combination set.
3. The method of claim 2, wherein weights of the number of the pixels and the number of the edge segments are set, and the number of the pixels corresponding to the closed edge and the number of the edge segments are weighted and summed to obtain a node value.
4. The method of claim 3, wherein a weight of the sum of the number of repetitions and the coverage ratio is set, and the sum of the number of repetitions and the coverage ratio are weighted and summed to obtain a weight of a node connecting edge.
5. The method according to claim 4, characterized in that the edge segmentation is performed on the closed edge, in particular:
each edge pixel on the closed edge is sequentially a pixel to be marked, the direction of the adjacent edge pixel on one side of the pixel to be marked pointing to the pixel to be marked is a first direction, the direction of the pixel to be marked pointing to the adjacent edge pixel on the other side is a second direction, and the pixel to be marked is marked according to the rotating direction and the rotating angle of the pixel to be marked which are rotated to the second direction from the first direction;
and performing edge segment segmentation on the closed edge at the position where the mark is changed.
6. The method of claim 5, wherein the line morphology comprises a line and a curve, the characteristic value of the line being a first value and the characteristic value of the curve being a second value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210321079.2A CN114419635B (en) | 2022-03-30 | 2022-03-30 | Electronic seal vector diagram identification method based on graphic identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210321079.2A CN114419635B (en) | 2022-03-30 | 2022-03-30 | Electronic seal vector diagram identification method based on graphic identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114419635A true CN114419635A (en) | 2022-04-29 |
| CN114419635B CN114419635B (en) | 2022-06-24 |
Family
ID=81262822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210321079.2A Active CN114419635B (en) | 2022-03-30 | 2022-03-30 | Electronic seal vector diagram identification method based on graphic identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114419635B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330584A (en) * | 2022-10-17 | 2022-11-11 | 北京点聚信息技术有限公司 | Anti-copying encryption method for electronic seal |
| CN115346227A (en) * | 2022-10-17 | 2022-11-15 | 景臣科技(南通)有限公司 | Method for vectorizing electronic file based on layout file |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504738A (en) * | 2014-12-20 | 2015-04-08 | 乐清咔咔网络科技有限公司 | Information seal and image information processing method of seal impression thereof |
| CN106846344A (en) * | 2016-12-14 | 2017-06-13 | 国家海洋局第二海洋研究所 | A kind of image segmentation optimal identification method based on the complete degree in edge |
| CN107610138A (en) * | 2017-10-20 | 2018-01-19 | 四川长虹电器股份有限公司 | A kind of bill seal regional sequence dividing method |
-
2022
- 2022-03-30 CN CN202210321079.2A patent/CN114419635B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504738A (en) * | 2014-12-20 | 2015-04-08 | 乐清咔咔网络科技有限公司 | Information seal and image information processing method of seal impression thereof |
| CN106846344A (en) * | 2016-12-14 | 2017-06-13 | 国家海洋局第二海洋研究所 | A kind of image segmentation optimal identification method based on the complete degree in edge |
| WO2018107939A1 (en) * | 2016-12-14 | 2018-06-21 | 国家海洋局第二海洋研究所 | Edge completeness-based optimal identification method for image segmentation |
| CN107610138A (en) * | 2017-10-20 | 2018-01-19 | 四川长虹电器股份有限公司 | A kind of bill seal regional sequence dividing method |
Non-Patent Citations (6)
| Title |
|---|
| PARTHA PRATIM ROY 等: "Seal Detection and Recognition: An Approach for Document Indexing", 《2009 10TH INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION》 * |
| PARTHA PRATIM ROY 等: "Seal Detection and Recognition: An Approach for Document Indexing", 《2009 10TH INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION》, 2 October 2009 (2009-10-02) * |
| 欧阳欢 等: "多特征融合决策的发票印章识别", 《计算机工程与设计》 * |
| 欧阳欢 等: "多特征融合决策的发票印章识别", 《计算机工程与设计》, 30 September 2018 (2018-09-30) * |
| 谢艳文 等: "基于２－邻域局部结构的矢量图符号模糊识别方法", 《计算机辅助设计与图形学报》 * |
| 谢艳文 等: "基于２－邻域局部结构的矢量图符号模糊识别方法", 《计算机辅助设计与图形学报》, 31 October 2014 (2014-10-31) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330584A (en) * | 2022-10-17 | 2022-11-11 | 北京点聚信息技术有限公司 | Anti-copying encryption method for electronic seal |
| CN115346227A (en) * | 2022-10-17 | 2022-11-15 | 景臣科技(南通)有限公司 | Method for vectorizing electronic file based on layout file |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114419635B (en) | 2022-06-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114419635B (en) | Electronic seal vector diagram identification method based on graphic identification | |
| CN109934826B (en) | Image feature segmentation method based on graph convolution network | |
| CN108154510A (en) | Method for detecting surface defects of products, device and computer readable storage medium | |
| US6173075B1 (en) | Drawing pixmap to vector conversion | |
| AU2006252019B2 (en) | Method and Apparatus for Dynamic Connector Analysis | |
| US6185341B1 (en) | Image processing using vector data to reduce noise | |
| CN111523521A (en) | Remote sensing image classification method for double-branch fusion multi-scale attention neural network | |
| WO2006102014A1 (en) | Fast graph cuts: a weak shape assumption provides a fast, exact method for graph cuts segmentation | |
| CN114494259A (en) | Cloth defect detection method based on artificial intelligence | |
| CN112819793A (en) | Container damage identification method, device, equipment and readable access medium | |
| JP2013536960A (en) | System and method for synthesizing portrait sketches from photographs | |
| CN111652240B (en) | CNN-based image local feature detection and description method | |
| Ranwez et al. | Order independent homotopic thinning for binary and grey tone anchored skeletons | |
| CN112215859B (en) | Texture boundary detection method based on deep learning and adjacency constraint | |
| CN116229205A (en) | Lipstick product surface defect data augmentation method based on small sample characteristic migration | |
| CN113609984A (en) | Pointer instrument reading identification method and device and electronic equipment | |
| CN100375124C (en) | A skeletonized object rebuild method | |
| CN115033721A (en) | Image retrieval method based on big data | |
| CN114359739B (en) | Target identification method and device | |
| CN116309157A (en) | Qin warrior texture restoration method combining recursive reasoning and feature re-optimization | |
| Ren et al. | A new and fast contour-filling algorithm | |
| CN111461139B (en) | Multi-target visual saliency layered detection method in complex scene | |
| CN114491841A (en) | Machining feature recognition method based on NX secondary development and graph neural network | |
| CN113011415A (en) | Improved target detection method and system based on Grid R-CNN model | |
| AU2009250986B2 (en) | Rendering piece-wise smooth image from colour values along paths |
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 |