CN112287145A - Method, system and device for searching graphic trademark and readable storage medium - Google Patents

Method, system and device for searching graphic trademark and readable storage medium Download PDF

Info

Publication number
CN112287145A
CN112287145A CN201910663424.9A CN201910663424A CN112287145A CN 112287145 A CN112287145 A CN 112287145A CN 201910663424 A CN201910663424 A CN 201910663424A CN 112287145 A CN112287145 A CN 112287145A
Authority
CN
China
Prior art keywords
graphic
trademark
picture
graph
retrieved
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910663424.9A
Other languages
Chinese (zh)
Inventor
卿春梅
唐义波
唐林波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Zhuteng Information Technology Co ltd
Original Assignee
Shanghai Zhuteng Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Zhuteng Information Technology Co ltd filed Critical Shanghai Zhuteng Information Technology Co ltd
Priority to CN201910663424.9A priority Critical patent/CN112287145A/en
Publication of CN112287145A publication Critical patent/CN112287145A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/53—Querying
    • G06F16/532—Query formulation, e.g. graphical querying
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/18—Legal services
    • G06Q50/184—Intellectual property management
    • 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
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Tourism & Hospitality (AREA)
  • Technology Law (AREA)
  • Databases & Information Systems (AREA)
  • Library & Information Science (AREA)
  • Evolutionary Computation (AREA)
  • Economics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a graph trademark retrieval method, which comprises the following steps: acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved; identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic; if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library; carrying out graphic trademark retrieval through graphic codes to obtain a trademark retrieval result; a graphic trademark retrieval system, an apparatus and a readable storage medium are also disclosed. Compared with the existing graph trademark retrieval method based on pure manual extraction, the method overcomes the defects that the traditional trademark graph element codes are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result.

Description

Method, system and device for searching graphic trademark and readable storage medium
Technical Field
The invention relates to the technical field of intellectual property inquiry, in particular to a graph trademark retrieval method.
Background
The trademark is a mark of a company, a product or a service, is integrated with the commodity quality, the service quality and the operation management of an enterprise, plays a very important role in the industrial and commercial society, is an important attribute of the company and the product thereof, and has uniqueness. In order to protect the trademark legally, it must be officially registered with the trademark office. With the development of economy and the acceleration of globalization process in China, the number of trademarks is gradually increased year by year. Preventing duplicate registration or similar trademark registration is a core problem of trademark management. In order to protect the legitimate rights and interests of the registered trademark and to attack the illegal act of embezzling the registered trademark, the registered trademark needs to be inquired, compared with the registered trademark, and the registered trademark is determined to be different or not similar to the registered trademark, so that the registered trademark has the registration qualification. In the conventional graphic trademark retrieval, the retrieval purpose is basically realized by manually inputting trademark graphic element codes as retrieval conditions. The trademark figure element code is a trademark figure element dividing tool generated according to 'establishing trademark figure element international classification Vienna protocol', and consists of a list of trademark figure elements classified according to major categories, minor categories and groups, wherein the list comprises trademark figure element numbers and trademark figure element names. Thus, each brand graphic element code represents the meaning of the contents of the brand graphic element. The existing trademark graph retrieval can be realized manually, but the existing trademark graph retrieval has obvious defects and disadvantages, which are mainly reflected in that:
1) the trademark graphic element codes need to be divided manually, and the low dividing work efficiency and the huge consumption of work energy are obviously easy to see;
2) even if the accurate trademark graphic element codes are obtained, the number of searched comparison graphic trademarks is huge, the low working efficiency and the huge consumption of working energy of manual comparison and identification are obviously easy to see, and the wide application of the graphic trademark search is limited;
3) the similarity of the trademark images can be different according to subjective judgment of different professionals, and the difference can cause the accuracy of a trademark image comparison result.
At present, there is also a method for searching trademarks through image recognition, but the method adopts the comparison of the graph trademarks to be searched with all registered graph trademarks or with all registered graph trademarks of the same category, and because the number of the registered trademarks is too large, the searching speed is greatly influenced, and the working efficiency and accuracy of recognition need to be improved.
Disclosure of Invention
In view of the above-mentioned shortcomings, the present invention provides a graph trademark retrieval method, which can obtain accurate graph codes through image recognition and obtain better retrieval results through the graph codes.
In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:
a method for searching a graphic trademark comprises the following steps:
acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved;
identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic;
if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library;
and searching the graphic trademark through the graphic code to obtain a trademark searching result.
According to one aspect of the present invention, the obtaining the corresponding graphic code according to the first graphic feature and the preset graphic code library comprises: and matching the obtained first graphic characteristics with preset graphic code library characters to obtain the corresponding graphic codes.
According to one aspect of the invention, the acquiring the picture and the graphic feature of the graphic trademark to be retrieved comprises: and acquiring a to-be-retrieved graphic trademark picture input by a user and a first graphic feature which is input by the user and used for describing a graphic element of the to-be-retrieved graphic trademark picture.
According to one aspect of the invention, the identifying the picture of the graphic trademark to be retrieved and confirming whether the picture contains the first graphic feature comprises the following steps:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of the picture, wherein the basic composition elements are graphic elements;
describing the graphic elements through characters to obtain the graphic features of the to-be-retrieved graphic trademark picture, wherein the obtained graphic features of the to-be-retrieved graphic trademark picture are second graphic features;
and comparing the second graphic features with the first graphic features, and judging whether the second graphic features comprise the first graphic features.
According to one aspect of the invention, the identifying the picture of the graphic trademark to be retrieved and confirming whether the picture contains the first graphic feature comprises the following steps:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of a graph trademark picture to be retrieved, wherein the basic composition elements are graph elements;
extracting characteristic points of graphic elements of the graphic trademark picture to be retrieved;
obtaining a graphic element feature point corresponding to the first graphic feature according to the first graphic feature;
and comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are contained in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, and judging whether the graphic trademark picture to be retrieved contains the first graphic characteristic according to the result.
According to one aspect of the invention, the graphic trademark retrieval method comprises the following steps: the method for judging the significance of the obtained first graphic features and sorting the first graphic features according to the significance comprises the following steps:
acquiring the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
judging the significance according to the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
and sorting the judgment results from high to low according to the significance.
According to one aspect of the invention, the graphical trademark retrieval method comprises the following steps:
after identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph feature, if the picture does not contain the first graph feature, the picture of the graph trademark to be retrieved is required to be provided with the first graph feature again or the picture of the graph trademark to be retrieved is subjected to image identification to obtain a second graph feature;
continuing to identify the picture of the graph trademark to be retrieved and confirming whether the picture contains the re-provided first graph characteristic until the picture contains the re-provided first graph characteristic;
and obtaining the corresponding graphic code according to the confirmed first graphic feature and a preset graphic code library or obtaining the corresponding graphic code according to the second graphic feature obtained by image recognition and the preset graphic code library.
According to one aspect of the invention, the identifying the picture of the graphic trademark to be retrieved to obtain the second graphic characteristic comprises: analyzing the composition of the graph and trademark picture to be retrieved, extracting basic composition elements of the picture, wherein the basic composition elements are the graph elements of the graph and trademark picture to be retrieved, and describing the graph elements through characters to obtain a second graph characteristic of the graph and trademark picture to be retrieved.
According to one aspect of the invention, the obtaining of the trademark retrieval result through the graphic trademark retrieval by the graphic code comprises the following steps: searching a graphic trademark through graphic codes to obtain a comparison trademark; and carrying out approximate judgment to obtain a final retrieval conclusion.
According to one aspect of the invention, the performing the approximate judgment to obtain the final retrieval conclusion comprises: and carrying out image recognition on the comparison trademark to extract a graphic element corresponding to the graphic code, carrying out similarity comparison on the obtained graphic element and a graphic element of the to-be-retrieved graphic trademark corresponding to the graphic feature of the to-be-retrieved graphic code to obtain a comparison result, sequencing the comparison result according to the similarity from high to low, synthesizing the similarity comparison results corresponding to all the graphic features of the to-be-retrieved graphic trademark to obtain a final retrieval conclusion, and giving a corresponding registration suggestion.
According to one aspect of the invention, the performing the approximate judgment to obtain the final retrieval conclusion comprises: and synthesizing the similarity comparison results corresponding to all the graphic features of the graphic trademark to be retrieved, combining the significance of all the graphic features of the graphic trademark to be retrieved to obtain a final retrieval conclusion, and giving corresponding registration suggestions.
According to one aspect of the invention, the graphic trademark retrieval method comprises the following steps: if the graph trademark to be searched contains characters, the character similarity comparison result is obtained after the similarity comparison is carried out on the character search character trademark, and corresponding registration suggestions are given according to the final search conclusion.
A graphical trademark retrieval system, the graphical trademark retrieval system comprising a graphical code library and:
the acquisition module is used for acquiring the picture of the graph trademark to be retrieved and the first graph characteristic and preprocessing the picture;
the identification module is used for identifying the picture of the graph trademark to be retrieved to obtain the graph characteristics contained in the picture;
the confirming module is used for confirming whether the graph characteristics obtained by the identifying module contain the first graph characteristics obtained by the obtaining module;
the code acquisition module is used for acquiring a graphic code corresponding to the graphic trademark to be retrieved according to the confirmed first graphic characteristic and a preset graphic code library;
and the retrieval module is used for carrying out graphic trademark retrieval through the graphic codes to obtain a trademark retrieval result.
In accordance with one aspect of the invention, the retrieval module comprises: the query module is used for searching the graphic trademark through the graphic code to obtain a comparison trademark; and the judging module is used for carrying out approximate judgment to obtain a final retrieval conclusion.
A graphic trademark retrieval device comprising:
a memory for storing a computer program;
a processor for implementing the steps of the graphical trademark retrieval method as described above when executing the computer program.
A readable storage medium having stored thereon a computer program which, when executed, implements the steps of the graphical trademark retrieval method as described above.
The implementation of the invention has the advantages that: the invention relates to a graph trademark retrieval method, which comprises the following steps: acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved; identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic; if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library; carrying out graphic trademark retrieval through graphic codes to obtain a trademark retrieval result; namely, the invention confirms whether the picture of the graph trademark to be searched contains the first graph characteristic submitted by the user by identifying the picture of the graph trademark to be searched to obtain the graph characteristic contained by the picture, or the picture of the graph trademark to be retrieved is extracted and compared with the graph element basic feature points of the first graph features submitted by the user to confirm whether the picture of the graph trademark to be retrieved contains the first graph features submitted by the user, after the picture is confirmed to contain the first graph features, the graph codes corresponding to the graph trademark to be retrieved are obtained through the confirmed first graph features and a preset graph code library, finally, the graph trademark is retrieved through the graph codes to obtain a trademark retrieval result, and the whole process does not manually obtain the graph codes corresponding to the graph trademark to be retrieved, the image code corresponding to the trademark of the graph to be retrieved is automatically obtained by identifying the image of the trademark of the graph to be retrieved and confirming the image characteristics; compared with the existing graph trademark retrieval method for purely manually extracting graph features, the method overcomes the defects that the graph element codes of the traditional trademark are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result; compared with other methods similar to the method for obtaining the graphic codes by establishing a pattern library in which trademark patterns and graphic codes are in one-to-one correspondence and then comparing the patterns in the pattern library with the patterns in the pattern library one by one, the method provided by the invention confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by identifying the pattern of the trademark to be searched to obtain the graphic features contained in the pattern, or confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by extracting the pattern element basic feature points of the pattern of the trademark to be searched and the first graphic features submitted by the user and then performs character matching on the confirmed first graphic features and the preset pattern code library to obtain the graphic codes corresponding to the trademark to be searched without performing a large amount of image identification and comparison, the workload is greatly reduced, and the calculation and storage resources occupied by the pattern library are not needed; matched graphic codes can be obtained very quickly through character matching, and meanwhile, compared with the existing self-defined patterns and graphic code libraries, the obtained retrieval result is more in line with the standard of trademark registration, and the retrieval accuracy is improved. Further, through the comparison trade mark of accurate acquisition, can greatly reduce the quantity that the trade mark was compared, promote comparison efficiency effectively, accurate comparison data can promote the precision of comparing.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic diagram of a graph and trademark retrieval method according to a first embodiment of the present invention;
fig. 2 is a schematic diagram of a to-be-retrieved graphic trademark picture according to a first embodiment of the present invention;
FIG. 3 is a diagram illustrating a part of the trademark retrieval result obtained in the first embodiment of the present invention;
fig. 4 is a schematic diagram illustrating a graphic trademark retrieval method according to a second embodiment of the present invention;
fig. 5 is a schematic diagram of a to-be-retrieved graphic trademark picture according to a second embodiment of the present invention;
FIG. 6 is a diagram illustrating a part of the trademark retrieval results obtained in the second embodiment of the present invention;
fig. 7 is a schematic diagram illustrating a graphical trademark retrieval method according to a third embodiment of the present invention;
fig. 8 is a schematic diagram illustrating a graphic trademark retrieval method according to a fourth embodiment of the present invention;
fig. 9 is a schematic view of a trademark picture of a graph to be retrieved according to a fourth embodiment of the present invention;
FIG. 10 is a diagram illustrating a part of the trademark retrieval result obtained in the fourth embodiment of the present invention;
fig. 11 is a schematic diagram illustrating a graphic trademark retrieval method according to a fifth embodiment of the present invention;
fig. 12 is a schematic diagram of a trademark picture of a graph to be retrieved according to the fifth embodiment of the present invention;
FIG. 13 is a diagram illustrating an approximate result of the character trademark retrieval according to the fifth embodiment of the present invention;
FIG. 14 is a diagram of a graphical trademark retrieval system according to a sixth embodiment of the present invention;
fig. 15 is a schematic diagram of a graphic trademark retrieval 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.
Example one
As shown in fig. 1, 2 and 3, a graphic trademark retrieval method includes the steps of:
step S1: acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved;
the step S1 of obtaining the picture and the first graphic feature of the graphic trademark to be retrieved may specifically be: acquiring a to-be-retrieved graphic trademark picture input by a user and a first graphic feature which is input by the user and used for describing a graphic element of the to-be-retrieved graphic trademark picture, and automatically acquiring the picture and a first graphic feature content by a system; for example, in this embodiment, as shown in fig. 2, the obtained to-be-retrieved graphic trademark picture submitted by the user obtains that the first graphic feature submitted by the user is "monkey".
In practical applications, when providing the picture of the trademark to be registered, the user may submit the category of the commodity or the service to be registered at the same time, for example, the category of the commodity or the service submitted by the user to be registered is the 09 th category in this embodiment.
Step S2: identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic;
the step S2 of recognizing the picture of the to-be-retrieved graphic trademark and determining whether the picture contains the first graphic feature may specifically include the following steps:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of the picture, wherein the basic composition elements are graphic elements;
describing the graphic elements through characters to obtain the graphic features of the to-be-retrieved graphic trademark picture, wherein the obtained graphic features of the to-be-retrieved graphic trademark picture are second graphic features;
and comparing the second graphic features with the first graphic features, and judging whether the second graphic features comprise the first graphic features.
In practical application, image preprocessing is required to be performed on the graph and trademark picture to be retrieved before identification, and the steps of graying, drying removal, normalization and the like can be included.
The image recognition technology adopted in step S2 is a prior art based on picture content recognition in the field of computer vision technology, so that the specific implementation process for recognizing the preprocessed picture to obtain the graphic features included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture.
For example, if the graph and trademark picture to be retrieved of the user obtained in step S1 in this embodiment is subjected to image recognition in step S2, it can be obtained that the basic constituent element is a monkey pattern, the monkey pattern is a graphic element of the picture, and the monkey pattern is described by text and then is a "monkey", and in this embodiment, the monkey pattern is submitted by the user, and therefore, it is necessary to compare and determine the "monkey" with the monkey to confirm that the first graphic feature "monkey" submitted by the user is included in the graph and trademark picture to be retrieved.
The identification of the preprocessed picture to obtain the graphic features contained in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps. For example, step S2 can be implemented by:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of a graph trademark picture to be retrieved, wherein the basic composition elements are graph elements;
extracting characteristic points of graphic elements of the graphic trademark picture to be retrieved;
according to the first graph feature obtained in the past, obtaining a feature point of a first graph element corresponding to the first graph feature;
according to the first graphic feature obtained before, obtaining the feature point of the first graphic element corresponding to the first graphic feature, and 2 ways can be adopted: one way is as follows: obtaining a graphic element corresponding to the first graphic feature according to the first graphic feature, and extracting a feature point of the graphic element corresponding to the first graphic feature through image processing; the other mode is as follows: a mapping database with mutually corresponding graphic features, graphic elements and graphic element feature points can be preset, and the feature points of the graphic elements corresponding to the first graphic features can be obtained by mapping and matching the first graphic features and the mapping database;
and comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are contained in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, and judging whether the graphic trademark picture to be retrieved contains the first graphic characteristic according to the result.
The steps can be specifically as follows: comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic trademark picture to be retrieved, if the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, obtaining that the graphic trademark picture to be retrieved includes the first graphic characteristic, otherwise, obtaining that the graphic trademark picture to be retrieved does not include the first graphic characteristic.
And determining whether the to-be-retrieved graphic trademark picture contains the first graphic feature according to the result.
Step S3: if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library;
if the first graphic feature is included, obtaining the corresponding graphic code according to the first graphic feature and a preset graphic code library comprises: matching the confirmed first graphic characteristics with preset graphic code library characters to obtain corresponding graphic codes; if the first graphic feature submitted by the user in step S1 in this embodiment is "monkey", the image recognition in step S2 confirms that the trademark of the graphic to be retrieved submitted by the user includes the first graphic feature "monkey" submitted by the user, and at this time, the first graphic feature "monkey" submitted by the user can be directly matched with the preset graphic code library text to obtain the corresponding graphic code. The method specifically comprises the following steps: matching description is carried out according to the confirmed first graphic characteristics and preset graphic code library characters, the description is similar or synonymous or comprises and is contained when the description is matched, and then the corresponding graphic codes are correspondingly obtained; for example, in this embodiment, it is determined according to step S2 that the graphic trademark picture to be retrieved by the user includes the first graphic feature submitted by the user as "monkey", and matching is performed in the preset graphic coding library according to the monkey, the matched characters are described as "primate, monkey, orangutan", and under the animal directory corresponding to the 3 codes, the graphic code corresponding to the monkey is "3.5.19", so that the graphic code whose graphic feature is the monkey of the graphic trademark picture to be retrieved by the user is "3.5.19".
In practical applications, if the picture of the graphic trademark to be retrieved is identified and whether the picture of the graphic trademark to be retrieved includes the first graphic feature is determined to be executed in step S2, and the result is that the picture of the graphic trademark to be retrieved does not include the first graphic feature submitted by the user, any one of the following 2 steps may be executed:
(1) asking the customer to resubmit the first graphical feature and confirming that the newly submitted first graphical feature continues to perform step S2; the steps can be repeated until the first graphic features submitted by the user are contained in the graphic trademark to be retrieved submitted by the user; for example, in this embodiment, the to-be-retrieved graphic trademark picture submitted by the user is shown in fig. 2, if the first graphic feature submitted by the user is a child, the second graphic feature is obtained as a monkey after the picture is identified in step S2, and the child is not consistent with the monkey, so that it can be determined that the first graphic feature submitted by the user is not included in the to-be-retrieved graphic trademark picture provided by the user, and the user is required to provide the correct first graphic feature again; if the first graphical feature submitted by the user for the second time is not included in the picture, the customer may be asked to provide it again, or the following step (2) may be performed.
(2) Directly identifying the picture of the graph trademark to be retrieved submitted by the user, and obtaining a second graph characteristic contained in the picture; then, step S3 is executed to obtain the graphics code corresponding to the second graphics feature, and then step S4 is directly executed.
The image recognition technology adopted for directly recognizing the picture of the graph trademark to be retrieved submitted by the user and obtaining the second graph feature contained in the picture is the prior art based on picture content recognition in the technical field of computer vision, so that the specific implementation process for recognizing the preprocessed picture and obtaining the second graph feature contained in the picture can comprise the following steps:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize the graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain second graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the graphic trademark to be retrieved.
The identification of the preprocessed picture to obtain the second graphic feature included in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps.
Step S4: carrying out graphic trademark retrieval through graphic codes to obtain a trademark retrieval result;
in step S4, the obtaining of the trademark search result by performing graphic trademark search through graphic coding may specifically be: all trademarks containing graphic features corresponding to the graphic codes are obtained through the graphic codes in a trademark library of a trademark office, and the trademark library becomes a search result. Compared with other graph trademark retrieval methods, the graph to be registered is compared with all graph trademarks in the trademark library to obtain a comparison result, the graph code is limited to obtain the trademarks corresponding to the graph codes, the retrieval number is greatly reduced, meanwhile, because the trademark office is a graph code maker and an examination standard editor of the graph codes, the graph code library in the trademark office is used, the obtained result is extremely accurate, and the accuracy is greatly improved. For example, in this embodiment, it is determined according to step S2 that the graph and trademark picture to be retrieved by the user includes the first graph feature "monkey" submitted by the user, the graph code corresponding to the first graph feature "monkey" is "3.5.19", and all trademarks related to the graph code are obtained from the trademark library through the graph code "3.5.19", which becomes the retrieval result. In practical application, when the graph and trademark is searched through the graph codes, the quasi-registration information of the graph and trademark to be searched is obtained, the classification group of the commodity or service where the graph and trademark is located is confirmed, and the graph and trademark is searched according to the classification group and the graph codes to obtain a search result; by adopting a commodity or service classification mode, more accurate data can be obtained. For example, in the present embodiment, if the trademark of the graphic to be registered by the user is class 09 is obtained at step S1, a partial search result obtained after the search of class 09 is performed by the graphic code "3.5.19" as shown in fig. 3.
The embodiment confirms whether the picture of the graph trademark to be retrieved contains the first graph characteristic submitted by the user by identifying the picture of the graph trademark to be retrieved to obtain the graph characteristic contained in the picture, or the picture of the graph trademark to be retrieved is extracted and compared with the graph element basic feature points of the first graph features submitted by the user to confirm whether the picture of the graph trademark to be retrieved contains the first graph features submitted by the user, after the picture is confirmed to contain the first graph features, the graph codes corresponding to the graph trademark to be retrieved are obtained through the confirmed first graph features and a preset graph code library, finally, the graph trademark is retrieved through the graph codes to obtain a trademark retrieval result, and the whole process does not manually obtain the graph codes corresponding to the graph trademark to be retrieved, the image code corresponding to the trademark of the graph to be retrieved is automatically obtained by identifying the image of the trademark of the graph to be retrieved and confirming the image characteristics; compared with the existing graph trademark retrieval method for purely manually extracting graph features, the method overcomes the defects that the graph element codes of the traditional trademark are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result; compared with other methods similar to the method for obtaining the graphic codes by establishing a pattern library in which trademark patterns and graphic codes are in one-to-one correspondence and then comparing the patterns in the pattern library with the patterns in the pattern library one by one, the method provided by the invention confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by identifying the pattern of the trademark to be searched to obtain the graphic features contained in the pattern, or confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by extracting the pattern element basic feature points of the pattern of the trademark to be searched and the first graphic features submitted by the user and then performs character matching on the confirmed first graphic features and the preset pattern code library to obtain the graphic codes corresponding to the trademark to be searched without performing a large amount of image identification and comparison, the workload is greatly reduced, and the calculation and storage resources occupied by the pattern library are not needed; matched graphic codes can be obtained very quickly through character matching, and meanwhile, compared with the existing self-defined patterns and graphic code libraries, the obtained retrieval result is more in line with the standard of trademark registration, and the retrieval accuracy is improved.
Example two
As shown in fig. 4, 5 and 6, a graphic trademark retrieval method includes the steps of:
step S1: acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved;
the step S1 of obtaining the picture and the first graphic feature of the graphic trademark to be retrieved may specifically be: acquiring a to-be-retrieved graphic trademark picture input by a user and a first graphic feature which is input by the user and used for describing a graphic element of the to-be-retrieved graphic trademark picture, and automatically acquiring the picture and a first graphic feature content by a system; for example, in this embodiment, as shown in fig. 5, the obtained to-be-retrieved graphic trademark picture submitted by the user is obtained, and the first graphic features submitted by the user are obtained as "girl" and "basket".
In practical applications, when providing the picture of the trademark to be registered, the user may submit the category of the commodity or the service to be registered at the same time, for example, the category of the commodity or the service submitted by the user to be registered is the 09 th category in this embodiment.
Step S2: identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic;
the step S2 of recognizing the picture of the to-be-retrieved graphic trademark and determining whether the picture contains the first graphic feature may specifically include the following steps:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of the picture, wherein the basic composition elements are graphic elements;
describing the graphic elements through characters to obtain the graphic features of the to-be-retrieved graphic trademark picture, wherein the obtained graphic features of the to-be-retrieved graphic trademark picture are second graphic features;
and comparing the second graphic features with the first graphic features, and judging whether the second graphic features comprise the first graphic features.
In practical application, before identification, image preprocessing is required to be performed on a trademark picture of a graph to be retrieved, which can comprise the steps of graying, drying removal, normalization and the like.
The image recognition technology adopted in step S2 is a prior art based on picture content recognition in the field of computer vision technology, so that the specific implementation process for recognizing the preprocessed picture to obtain the graphic features included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture.
As shown in fig. 5, the image recognition is performed on the to-be-retrieved graphic trademark picture of the user obtained in step S1 in this embodiment, so as to obtain the patterns of which the basic constituent elements are girl patterns and basket patterns, respectively, and the girl patterns and the basket patterns are distributed as one graphic element of the picture, and the two patterns are described by words and are respectively "girl" and "basket", whereas the first graphic features submitted by the user obtained in step S1 in this embodiment are "girl" and "basket", and therefore, it is necessary to compare and judge the "girl" and "basket" respectively, and confirm that the first graphic features submitted by the user are included in the to-be-retrieved graphic trademark picture.
The identification of the preprocessed picture to obtain the graphic features contained in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps. For example, step S2 can be implemented by:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of a graph trademark picture to be retrieved, wherein the basic composition elements are graph elements;
extracting characteristic points of graphic elements of the graphic trademark picture to be retrieved;
according to the first graph feature obtained in the past, obtaining a feature point of a first graph element corresponding to the first graph feature;
according to the first graphic feature obtained before, obtaining the feature point of the first graphic element corresponding to the first graphic feature, and 2 ways can be adopted: one way is as follows: obtaining a graphic element corresponding to the first graphic feature according to the first graphic feature, and extracting a feature point of the graphic element corresponding to the first graphic feature through image processing; the other mode is as follows: a mapping database with mutually corresponding graphic features, graphic elements and graphic element feature points can be preset, and the feature points of the graphic elements corresponding to the first graphic features can be obtained by mapping and matching the first graphic features and the mapping database;
and comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are contained in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, and judging whether the graphic trademark picture to be retrieved contains the first graphic characteristic according to the result.
The steps can be specifically as follows: comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic trademark picture to be retrieved, if the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, obtaining that the graphic trademark picture to be retrieved includes the first graphic characteristic, otherwise, obtaining that the graphic trademark picture to be retrieved does not include the first graphic characteristic.
And determining whether the to-be-retrieved graphic trademark picture contains the first graphic feature according to the result.
The image recognition technology adopted in step S2 is a prior art based on picture content recognition in the field of computer vision technology, and thus, a specific implementation process for recognizing the preprocessed picture to obtain the feature points included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a feature point extraction model capable of extracting basic feature points of different types of graphic elements according to the training result;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on the segmented pictures through the feature point extraction model to obtain basic feature points of the graphic elements in each segmented picture.
Step S3: carrying out significance judgment on the obtained first graph characteristics and sequencing according to significance;
the step of judging the significance of the obtained first graphic features and sorting the first graphic features according to the significance comprises the following steps:
acquiring the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
judging the significance according to the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
and sorting the judgment results from high to low according to the significance.
For example, the embodiment confirms the first graphic features "girl" and "basket" submitted by the user as the graphic features contained in the graphic trademark picture to be retrieved according to step S2, and according to the overall proportion of the girl and the basket in the trademark picture, the saliency arrangement of the girl and the basket can be obtained as follows: girls > baskets. For example, in the present embodiment, a highly conspicuous girl is used as a graphic feature to be retrieved later.
In practical applications, the second graphic features obtained by image recognition may also be subjected to saliency sorting through the above steps.
Step S4: if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library;
if the first graphic feature is included, obtaining the corresponding graphic code according to the first graphic feature and a preset graphic code library comprises: matching the confirmed first graphic characteristics with preset graphic code library characters to obtain corresponding graphic codes; if the first graphic features submitted by the user are "girl" and "basket" as obtained in step S1 of this embodiment, it is recognized and confirmed through the image in step S2 that the graphic trademark picture to be retrieved submitted by the user includes the first graphic features "girl" and "basket" submitted by the user, and at this time, the first graphic features "girl" and "basket" submitted by the user can be directly matched with the preset graphic code library text characters respectively to obtain the corresponding graphic codes. The method specifically comprises the following steps: matching description is carried out according to the confirmed first graphic characteristics and preset graphic code library characters, the description is similar or synonymous or comprises and is contained when the description is matched, and then the corresponding graphic codes are correspondingly obtained; for example, in this embodiment, it is determined according to step S2 that the to-be-retrieved graphic trademark picture of the user includes the first graphic feature submitted by the user as "girl" and "basket", the matching is performed in the preset graphic code library according to the girl, the matched text is described as "girl", the corresponding graphic code is "2.5.3" under the human directory corresponding to the 2 code, and therefore, the graphic code of the girl is "2.5.3" can be obtained; matching is carried out in a preset pattern code library according to the basket, the matched characters are described as ' basket, pack basket and prime, the corresponding pattern code is ' A19.1.12 ' under the catalogue corresponding to the 19 code, and therefore, the pattern code of the basket is ' A19.1.12 ' can be obtained. Meanwhile, in the present embodiment, according to the saliency determination of step S3, the present embodiment uses the girl' S graphic code to perform the subsequent step S6.
In practical applications, if the picture of the graphic trademark to be retrieved is identified and whether the picture of the graphic trademark to be retrieved includes the first graphic feature is determined to be executed in step S2, and the result is that the picture of the graphic trademark to be retrieved does not include the first graphic feature submitted by the user, any one of the following 2 steps may be executed:
(1) asking the customer to resubmit the first graphical feature and confirming that the newly submitted first graphical feature continues to perform step S2; the steps can be repeated until the first graphic features submitted by the user are contained in the graphic trademark to be retrieved submitted by the user; for example, in this embodiment, if the to-be-retrieved graphic trademark picture submitted by the user is shown in fig. 5, and if the first graphic feature submitted by the user is a monkey, the step S2 is performed to perform picture recognition to obtain second graphic features, namely a girl and a basket, so that it can be determined that the first graphic feature submitted by the user is not included in the to-be-retrieved graphic trademark picture provided by the user, and the user is required to provide the correct first graphic feature again; if the first graphical feature submitted by the user for the second time is not included in the picture, the customer may be asked to provide it again, or the following step (2) may be performed.
(2) Directly identifying the picture of the graph trademark to be retrieved submitted by the user, and obtaining a second graph characteristic contained in the picture; then, step S4 is executed to obtain the graphics code corresponding to the second graphics feature, and then step S5 is directly executed.
The image recognition technology adopted for directly recognizing the picture of the graph trademark to be retrieved submitted by the user and obtaining the second graph feature contained in the picture is the prior art based on picture content recognition in the technical field of computer vision, so that the specific implementation process for recognizing the preprocessed picture and obtaining the second graph feature contained in the picture can comprise the following steps:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the to-be-retrieved graphic trademark picture.
The identification of the preprocessed picture to obtain the second graphic feature included in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps.
Step S5: carrying out graphic trademark retrieval through graphic codes to obtain a trademark retrieval result;
in step S5, the obtaining of the trademark search result by performing graphic trademark search through graphic coding may specifically be: all trademarks containing graphic features corresponding to the graphic codes are obtained through the graphic codes in a trademark library of a trademark office, and the trademark library becomes a search result. Compared with other graph trademark retrieval methods, the graph to be registered is compared with all graph trademarks in the trademark library to obtain a comparison result, the graph code is limited to obtain the trademarks corresponding to the graph codes, the retrieval number is greatly reduced, meanwhile, because the trademark office is a graph code maker and an examination standard editor of the graph codes, the graph code library in the trademark office is used, the obtained result is extremely accurate, and the accuracy is greatly improved. For example, in step S2 of this embodiment, it is determined that the to-be-retrieved graphic trademark picture of the user includes the first graphic feature submitted by the user as a girl and a basket, the obtained graphic code of the girl is "2.5.3", the obtained code of the basket is "A19.1.12", and all trademarks including the graphic feature corresponding to the graphic code can be obtained from the trademark library according to "2.5.3" and "A19.1.12", respectively, to become a retrieval result; or simultaneously acquiring all trademarks containing the graphic features corresponding to the graphic codes from the trademark library according to the 2.5.3 and the A19.1.12 to obtain the search result. In practical application, when the graph and trademark is searched through the graph codes, the quasi-registration information of the graph and trademark to be searched is obtained, the classification group of the commodity or service where the graph and trademark is located is confirmed, and the graph and trademark is searched according to the classification group and the graph codes to obtain a search result; by adopting a commodity or service classification mode, more accurate data can be obtained. For example, the graphical icon to be retrieved by the user obtained in step S1 of the present embodiment is supposed to be registered in the ninth category of the goods or services, so that the present embodiment performs a retrieval in the ninth category to obtain a retrieval result.
Meanwhile, in the present embodiment, it is necessary to determine which graphics code to search in combination with the saliency feature ranking, but in the present embodiment, the graphics code with high saliency is used for searching. In the embodiment, in step S3, a saliency array of the girl and the basket is obtained, and the saliency of the girl significantly exceeds that of the basket, so that the embodiment uses a graph search on the girl to obtain a trademark search result, as shown in fig. 6, which is a partial search result obtained after the ninth type search through a girl graph code "2.5.3".
The embodiment confirms whether the picture of the graph trademark to be retrieved contains the first graph characteristic submitted by the user by identifying the picture of the graph trademark to be retrieved to obtain the graph characteristic contained in the picture, or the picture of the graph trademark to be retrieved is extracted and compared with the graph element basic feature points of the first graph features submitted by the user to confirm whether the picture of the graph trademark to be retrieved contains the first graph features submitted by the user, after the picture is confirmed to contain the first graph features, the graph codes corresponding to the graph trademark to be retrieved are obtained through the confirmed first graph features and a preset graph code library, finally, the graph trademark is retrieved through the graph codes to obtain a trademark retrieval result, and the whole process does not manually obtain the graph codes corresponding to the graph trademark to be retrieved, the image code corresponding to the trademark of the graph to be retrieved is automatically obtained by identifying the image of the trademark of the graph to be retrieved and confirming the image characteristics; compared with the existing graph trademark retrieval method for purely manually extracting graph features, the method overcomes the defects that the graph element codes of the traditional trademark are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result; compared with other methods similar to the method for obtaining the graphic codes by establishing a pattern library in which trademark patterns and graphic codes are in one-to-one correspondence and then comparing the patterns in the pattern library with the patterns in the pattern library one by one, the method provided by the invention confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by identifying the pattern of the trademark to be searched to obtain the graphic features contained in the pattern, or confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by extracting the pattern element basic feature points of the pattern of the trademark to be searched and the first graphic features submitted by the user and then performs character matching on the confirmed first graphic features and the preset pattern code library to obtain the graphic codes corresponding to the trademark to be searched without performing a large amount of image identification and comparison, the workload is greatly reduced, and the calculation and storage resources occupied by the pattern library are not needed; matched graphic codes can be obtained very quickly through character matching, and meanwhile, compared with the existing self-defined patterns and graphic code libraries, the obtained retrieval result is more in line with the standard of trademark registration, and the retrieval accuracy is improved. Furthermore, by judging the significance, the graph characteristics and the graph codes of the graph trademarks with complex composition elements can be selected for retrieval through the significance, and then the retrieval results are obtained through the graph codes, so that a large amount of repeated or meaningless retrieval work is reduced, and the retrieval efficiency is greatly improved.
EXAMPLE III
As shown in fig. 7, 2 and 3, a graphic trademark retrieval method includes the steps of:
step S1: acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved;
the step S1 of obtaining the picture and the first graphic feature of the graphic trademark to be retrieved may specifically be: acquiring a to-be-retrieved graphic trademark picture input by a user and a first graphic feature which is input by the user and used for describing a graphic element of the to-be-retrieved graphic trademark picture, and automatically acquiring the picture and a first graphic feature content by a system; for example, in this embodiment, as shown in fig. 2, the obtained to-be-retrieved graphic trademark picture submitted by the user obtains that the first graphic feature submitted by the user is "monkey".
In practical applications, when providing the picture of the trademark to be registered, the user may submit the category of the commodity or the service to be registered at the same time, for example, the category of the commodity or the service submitted by the user to be registered is the 09 th category in this embodiment.
Step S2: identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic;
the step S2 of recognizing the picture of the to-be-retrieved graphic trademark and determining whether the picture contains the first graphic feature may specifically include the following steps:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of the picture, wherein the basic composition elements are graphic elements;
describing the graphic elements through characters to obtain the graphic features of the to-be-retrieved graphic trademark picture, wherein the obtained graphic features of the to-be-retrieved graphic trademark picture are second graphic features;
and comparing the second graphic features with the first graphic features, and judging whether the second graphic features comprise the first graphic features.
In practical application, before identification, image preprocessing is required to be performed on a trademark picture of a graph to be retrieved, which can comprise the steps of graying, drying removal, normalization and the like.
The image recognition technology adopted in step S2 is a prior art for recognizing based on picture content in the field of computer vision technology, so that the specific implementation process for recognizing the preprocessed picture to obtain the second graphic feature included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the to-be-retrieved graphic trademark picture.
For example, if the graph and trademark picture to be retrieved of the user obtained in step S1 in this embodiment is subjected to image recognition in step S2, it can be obtained that the basic constituent element is a monkey pattern, the monkey pattern is a graphic element of the picture, and the monkey pattern is described by text and then is a "monkey", and in this embodiment, the monkey pattern is submitted by the user, and therefore, it is necessary to compare and determine the "monkey" with the monkey to confirm that the first graphic feature "monkey" submitted by the user is included in the graph and trademark picture to be retrieved.
The identification of the preprocessed picture to obtain the graphic features contained in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps. For example, step S2 can be implemented by:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of a graph trademark picture to be retrieved, wherein the basic composition elements are graph elements;
extracting characteristic points of graphic elements of the graphic trademark picture to be retrieved;
according to the first graph feature obtained in the past, obtaining a feature point of a first graph element corresponding to the first graph feature;
according to the first graphic feature obtained before, obtaining the feature point of the first graphic element corresponding to the first graphic feature, and 2 ways can be adopted: one way is as follows: obtaining a graphic element corresponding to the first graphic feature according to the first graphic feature, and extracting a feature point of the graphic element corresponding to the first graphic feature through image processing; the other mode is as follows: a mapping database with mutually corresponding graphic features, graphic elements and graphic element feature points can be preset, and the feature points of the graphic elements corresponding to the first graphic features can be obtained by mapping and matching the first graphic features and the mapping database;
and comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are contained in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, and judging whether the graphic trademark picture to be retrieved contains the first graphic characteristic according to the result.
The steps can be specifically as follows: comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic trademark picture to be retrieved, if the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, obtaining that the graphic trademark picture to be retrieved includes the first graphic characteristic, otherwise, obtaining that the graphic trademark picture to be retrieved does not include the first graphic characteristic.
And determining whether the to-be-retrieved graphic trademark picture contains the first graphic feature according to the result.
Step S3: if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library;
if the first graphic feature is included, obtaining the corresponding graphic code according to the first graphic feature and a preset graphic code library comprises: matching the confirmed first graphic characteristics with preset graphic code library characters to obtain corresponding graphic codes; if the first graphic feature submitted by the user in step S1 in this embodiment is "monkey", the image recognition in step S2 confirms that the trademark of the graphic to be retrieved submitted by the user includes the first graphic feature "monkey" submitted by the user, and at this time, the first graphic feature "monkey" submitted by the user can be directly matched with the preset graphic code library text to obtain the corresponding graphic code. The method specifically comprises the following steps: matching description is carried out according to the confirmed first graphic characteristics and preset graphic code library characters, the description is similar or synonymous or comprises and is contained when the description is matched, and then the corresponding graphic codes are correspondingly obtained; for example, in this embodiment, it is determined according to step S2 that the graphic trademark picture to be retrieved by the user includes the first graphic feature submitted by the user as "monkey", and matching is performed in the preset graphic coding library according to the monkey, the matched characters are described as "primate, monkey, orangutan", and under the animal directory corresponding to the 3 codes, the graphic code corresponding to the monkey is "3.5.19", so that the graphic code whose graphic feature is the monkey of the graphic trademark picture to be retrieved by the user is "3.5.19".
In practical applications, if the picture of the graphic trademark to be retrieved is identified and whether the picture of the graphic trademark to be retrieved includes the first graphic feature is determined to be executed in step S2, and the result is that the picture of the graphic trademark to be retrieved does not include the graphic feature submitted by the user, any one of the following 2 steps may be executed:
(1) asking the customer to resubmit the first graphical feature and confirming that the newly submitted first graphical feature continues to perform step S2; the steps can be repeated until the first graphic features submitted by the user are contained in the graphic trademark to be retrieved submitted by the user; for example, in this embodiment, the to-be-retrieved graphic trademark picture submitted by the user is as shown in fig. 2, and if the first graphic feature submitted by the user is a child, after the picture is identified and confirmed in step S2, it may be determined that the first graphic feature submitted by the user is not included in the to-be-retrieved graphic trademark picture provided by the user, and the user is required to provide a correct first graphic feature again; if the first graphical feature submitted by the user for the second time is not included in the picture, the customer may be asked to provide it again, or the following step (2) may be performed.
(2) Directly identifying the picture of the graph trademark to be retrieved submitted by the user, and obtaining a second graph characteristic contained in the picture; then, step S3 is executed to obtain the graphics code corresponding to the second graphics feature, and then step S4 is directly executed.
The image recognition technology adopted for directly recognizing the picture of the graph trademark to be retrieved submitted by the user and obtaining the second graph feature contained in the picture is the prior art based on picture content recognition in the technical field of computer vision, so that the specific implementation process for recognizing the preprocessed picture and obtaining the second graph feature contained in the picture can comprise the following steps:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the to-be-retrieved graphic trademark picture.
The identification of the preprocessed picture to obtain the second graphic feature included in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps.
Step S4: searching a graphic trademark through graphic codes to obtain a comparison trademark;
in step S4, the obtaining of the comparison trademark by performing graphic trademark retrieval through graphic coding may specifically be: all trademarks containing graphic features corresponding to the graphic codes are obtained through the graphic codes in a trademark library of a trademark office and become comparison trademarks. Compared with other graph trademark retrieval methods, the graph to be registered is compared with all graph trademarks in the trademark library to obtain a comparison result, the graph code is limited to obtain the trademarks corresponding to the graph codes, the retrieval number is greatly reduced, meanwhile, because the trademark office is a graph code maker and an examination standard editor of the graph codes, the graph code library in the trademark office is used, the obtained result is extremely accurate, and the accuracy is greatly improved. For example, in this embodiment, it is determined according to step S2 that the graphic trademark picture to be retrieved by the user includes a graphic feature "monkey" submitted by the user, the graphic code corresponding to the first graphic feature "monkey" is "3.5.19", and all trademarks related to the graphic code are obtained from the trademark library through the graphic code "3.5.19" to become the retrieval result. In practical application, when the graph and trademark is searched through the graph codes, the quasi-registration information of the graph and trademark to be searched is obtained, the classification group of the commodity or service where the graph and trademark is located is confirmed, and the graph and trademark is searched according to the classification group and the graph codes to obtain a search result; by adopting a commodity or service classification mode, more accurate data can be obtained. For example, in the present embodiment, if the trademark of the graph to be registered by the user is obtained as the 09 th class at step S1, the trademark is compared with the part obtained after searching the 09 th class by the graph code "3.5.19" as shown in fig. 3.
Step S5: and carrying out approximate judgment to obtain a final retrieval conclusion.
The step of performing approximate judgment to obtain a final retrieval conclusion comprises the following steps: and carrying out image recognition on the comparison trademark to extract a graphic element corresponding to the graphic code, carrying out similarity comparison on the obtained graphic element and a graphic element of the to-be-retrieved graphic trademark corresponding to the graphic feature of the graphic code to obtain a comparison result, and sequencing the comparison result according to the sequence of similarity from high to low. In this embodiment, the graphic code of the graphic element monkey pattern of the graphic trademark to be retrieved corresponding to the graphic feature is obtained through step S3, and the comparison trademark is obtained through the graphic code; by identifying the figure element of figure code '3.5.19' contained in the comparison trademark, comparing and approximately judging the figure element with the monkey as the figure element of the figure trademark to be retrieved, the similarity between the figure trademark to be retrieved and each comparison trademark can be obtained, and then the similarity is sequenced.
In practical application, the performing the approximate judgment to obtain the final retrieval conclusion includes: and synthesizing the similarity comparison results corresponding to all the graphic features of the graphic trademark to be retrieved to obtain a final retrieval conclusion, and giving a corresponding registration suggestion.
The embodiment confirms whether the picture of the graph trademark to be retrieved contains the first graph characteristic submitted by the user by identifying the picture of the graph trademark to be retrieved to obtain the graph characteristic contained in the picture, or the picture of the graph trademark to be retrieved is extracted and compared with the graph element basic feature points of the first graph features submitted by the user to confirm whether the picture of the graph trademark to be retrieved contains the first graph features submitted by the user, after the picture is confirmed to contain the first graph features, the graph codes corresponding to the graph trademark to be retrieved are obtained through the confirmed first graph features and a preset graph code library, finally, the graph trademark is retrieved through the graph codes to obtain a trademark retrieval result, and the whole process does not manually obtain the graph codes corresponding to the graph trademark to be retrieved, the image code corresponding to the trademark of the graph to be retrieved is automatically obtained by identifying the image of the trademark of the graph to be retrieved and confirming the image characteristics; compared with the existing graph trademark retrieval method for purely manually extracting graph features, the method overcomes the defects that the graph element codes of the traditional trademark are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result; compared with other methods similar to the method for obtaining the graphic codes by establishing a pattern library in which trademark patterns and graphic codes are in one-to-one correspondence and then comparing the patterns in the pattern library with the patterns in the pattern library one by one, the method provided by the invention confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by identifying the pattern of the trademark to be searched to obtain the graphic features contained in the pattern, or confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by extracting the pattern element basic feature points of the pattern of the trademark to be searched and the first graphic features submitted by the user and then performs character matching on the confirmed first graphic features and the preset pattern code library to obtain the graphic codes corresponding to the trademark to be searched without performing a large amount of image identification and comparison, the workload is greatly reduced, and the calculation and storage resources occupied by the pattern library are not needed; matched graphic codes can be obtained very quickly through character matching, and meanwhile, compared with the existing self-defined patterns and graphic code libraries, the obtained retrieval result is more in line with the standard of trademark registration, and the retrieval accuracy is improved. Compared with the existing graph trademark retrieval method for purely and manually extracting the graph features, the method provided by the invention can automatically compare the graph features with the graph elements to obtain an approximate judgment result, and obtain a retrieval conclusion and a registration suggestion according to the approximate judgment result to realize artificial intelligent trademark retrieval. Further, through the comparison trade mark of accurate acquisition, can greatly reduce the quantity that the trade mark was compared, promote comparison efficiency effectively, accurate comparison data can promote the precision of comparing.
Example four
As shown in fig. 8, 9 and 10, a graphic trademark retrieval method includes the steps of:
step S1: acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved;
the step S1 of obtaining the picture and the first graphic feature of the graphic trademark to be retrieved may specifically be: acquiring a to-be-retrieved graphic trademark picture input by a user and a first graphic feature which is input by the user and used for describing a graphic element of the to-be-retrieved graphic trademark picture, and automatically acquiring the picture and a first graphic feature content by a system; for example, in this embodiment, as shown in fig. 9, the obtained to-be-retrieved graphic trademark picture submitted by the user obtains first graphic features submitted by the user as "girl" and "monkey".
In practical applications, when providing the picture of the trademark to be registered, the user may submit the category of the commodity or the service to be registered at the same time, for example, the category of the commodity or the service submitted by the user to be registered is the 09 th category in this embodiment.
Step S2: identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic;
the step S2 of recognizing the picture of the to-be-retrieved graphic trademark and determining whether the picture contains the first graphic feature may specifically include the following steps:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of the picture, wherein the basic composition elements are graphic elements;
describing the graphic elements through characters to obtain the graphic features of the to-be-retrieved graphic trademark picture, wherein the obtained graphic features of the to-be-retrieved graphic trademark picture are second graphic features;
and comparing the second graphic features with the first graphic features, and judging whether the second graphic features comprise the first graphic features.
In practical application, before identification, image preprocessing is required to be performed on a trademark picture of a graph to be retrieved, which can comprise the steps of graying, drying removal, normalization and the like.
The image recognition technology adopted in step S2 is a prior art for recognizing based on picture content in the field of computer vision technology, so that the specific implementation process for recognizing the preprocessed picture to obtain the second graphic feature included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the trademark picture of the graph to be retrieved.
As shown in fig. 9, the image recognition is performed on the graph and trademark picture to be retrieved of the user obtained in step S1 in this embodiment, so as to obtain basic components of the graph and trademark picture, which are respectively girl pattern, monkey pattern and basket pattern, the girl pattern, monkey pattern and basket pattern are distributed as one of the graphic elements of the picture, and the two patterns are respectively "girl", "monkey" and "basket" after being described by characters, which are the second graphic features, and the first graphic features submitted by the user obtained in step S1 in this embodiment are "girl" and "monkey", so that it is necessary to respectively compare and determine "girl" and "monkey" and determine whether the first graphic features submitted by the user are included in the graph and trademark picture to be retrieved.
The identification of the preprocessed picture to obtain the second graphic feature included in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps. For example, step S2 can be implemented by:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of a graph trademark picture to be retrieved, wherein the basic composition elements are graph elements;
extracting characteristic points of graphic elements of the graphic trademark picture to be retrieved;
according to the first graph feature obtained in the past, obtaining a feature point of a first graph element corresponding to the first graph feature;
according to the first graphic feature obtained before, obtaining the feature point of the first graphic element corresponding to the first graphic feature, and 2 ways can be adopted: one way is as follows: obtaining a graphic element corresponding to the first graphic feature according to the first graphic feature, and extracting a feature point of the graphic element corresponding to the first graphic feature through image processing; the other mode is as follows: a mapping database with mutually corresponding graphic features, graphic elements and graphic element feature points can be preset, and the feature points of the graphic elements corresponding to the first graphic features can be obtained by mapping and matching the first graphic features and the mapping database;
and comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are contained in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, and judging whether the graphic trademark picture to be retrieved contains the first graphic characteristic according to the result.
The steps can be specifically as follows: comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic trademark picture to be retrieved, if the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, obtaining that the graphic trademark picture to be retrieved includes the first graphic characteristic, otherwise, obtaining that the graphic trademark picture to be retrieved does not include the first graphic characteristic.
And determining whether the to-be-retrieved graphic trademark picture contains the first graphic feature according to the result.
The image recognition technology adopted in step S2 is a prior art based on picture content recognition in the field of computer vision technology, and thus, a specific implementation process for recognizing the preprocessed picture to obtain the feature points included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a feature point extraction model capable of extracting basic feature points of different types of graphic elements according to the training result;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on the segmented pictures through the feature point extraction model to obtain basic feature points of the graphic elements in each segmented picture.
Step S3: carrying out significance judgment on the obtained first graph characteristics and sequencing according to significance;
the step of judging the significance of the obtained first graphic features and sorting the first graphic features according to the significance comprises the following steps:
acquiring the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
judging the significance according to the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
and sorting the judgment results from high to low according to the significance.
The graphical features include a first graphical feature submitted by a user and a second graphical feature obtained by image recognition.
For example, in the present embodiment, it is confirmed according to step S2 that the first graphic features "girl", "monkey" and "basket" submitted by the user are graphic features included in the graphic trademark picture to be retrieved, and according to the overall proportion of girl and basket in the trademark picture, the saliency arrangement of girl, monkey and basket can be obtained as follows: girls > monkeys > hand baskets. For example, in the present embodiment, a highly conspicuous girl and a monkey are used as the graphic features to be retrieved later.
In practical applications, the second graphic features obtained by image recognition may also be subjected to saliency sorting through the above steps.
Step S4: if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library;
if the first graphic feature is included, obtaining the corresponding graphic code according to the first graphic feature and a preset graphic code library comprises: matching the confirmed first graphic characteristics with preset graphic code library characters to obtain corresponding graphic codes; if the first graphic features submitted by the user in step S1 in this embodiment are "girl" and "monkey", the image recognition in step S2 confirms that the trademark of the graph to be retrieved submitted by the user includes the first graphic features "girl" and "monkey" submitted by the user, and at this time, the first graphic features "girl" and "monkey" submitted by the user can be directly matched with the preset graphic code library characters to obtain the corresponding graphic codes. The method specifically comprises the following steps: matching description is carried out according to the confirmed first graphic characteristics and preset graphic code library characters, the description is similar or synonymous or comprises and is contained when the description is matched, and then the corresponding graphic codes are correspondingly obtained; for example, in this embodiment, the first graphic features of the graphic trademark picture to be retrieved by the user, which is confirmed according to step S2, are "girl" and "monkey", the matching is performed in the preset graphic code library according to the girl, the matched text is described as "girl", the corresponding graphic code is "2.5.3" under the human directory corresponding to the 2 code, and therefore, the graphic code of the girl is "2.5.3" can be obtained; according to the monkey, matching is carried out in a preset graph code library, matched characters are described as primates, monkeys and orangutans, the characters are positioned in a directory of animals corresponding to the 3 codes, and the corresponding graph code is 3.5.19, so that the graph code of the monkey is 3.5.19. Meanwhile, in the present embodiment, according to the significance determination of step S3, the graphical codes of girls and monkeys are adopted for the present embodiment to perform the subsequent step S6.
In practical applications, if the picture of the graphic trademark to be retrieved is identified and whether the picture of the graphic trademark to be retrieved includes the first graphic feature is determined to be executed in step S2, and the result is that the picture of the graphic trademark to be retrieved does not include the first graphic feature submitted by the user, any one of the following 2 steps may be executed:
(1) asking the customer to resubmit the first graphical feature and confirming that the newly submitted first graphical feature continues to perform step S2; the steps can be repeated until the first graphic features submitted by the user are contained in the graphic trademark to be retrieved submitted by the user; for example, in this embodiment, if the to-be-retrieved graphic trademark picture submitted by the user is shown in fig. 9, and if the first graphic feature submitted by the user is a lion, the second graphic feature is a monkey obtained by performing picture recognition through step S2, and after comparison, it can be determined that the first graphic feature submitted by the user is not included in the to-be-retrieved graphic trademark picture provided by the user, and the user is required to provide a correct first graphic feature again; if the first graphical feature submitted by the user for the second time is not included in the picture, the customer may be asked to provide it again, or the following step (2) may be performed.
(2) Directly identifying the picture of the graph trademark to be retrieved submitted by the user, and obtaining a second graph characteristic contained in the picture; then, step S4 is executed to obtain the graphics code corresponding to the second graphics feature, and then step S5 is directly executed.
The image recognition technology adopted for directly recognizing the picture of the graph trademark to be retrieved submitted by the user and obtaining the second graph feature contained in the picture is the prior art based on picture content recognition in the technical field of computer vision, so that the specific implementation process for recognizing the preprocessed picture and obtaining the second graph feature contained in the picture can comprise the following steps:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the graphic trademark to be retrieved.
The identification of the preprocessed picture to obtain the second graphic feature included in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps.
Step S5: searching a graphic trademark through graphic codes to obtain a comparison trademark;
in step S5, the obtaining of the comparison trademark by performing graphic trademark retrieval through graphic coding may specifically be: all trademarks containing graphic features corresponding to the graphic codes are obtained through the graphic codes in a trademark library of a trademark office and become comparison trademarks. Compared with other graph trademark retrieval methods, the graph to be registered is compared with all graph trademarks in the trademark library to obtain a comparison result, the graph code is limited to obtain the trademarks corresponding to the graph codes, the retrieval number is greatly reduced, meanwhile, because the trademark office is a graph code maker and an examination standard editor of the graph codes, the graph code library in the trademark office is used, the obtained result is extremely accurate, and the accuracy is greatly improved. For example, in this embodiment, it is determined that the to-be-retrieved graphic trademark picture of the user includes the first graphic feature girl and the monkey submitted by the user according to step S2, the obtained graphic code of the girl is "2.5.3", the obtained code of the monkey is "3.5.19", and all trademarks including the graphic feature corresponding to the graphic code can be obtained from the trademark library according to "2.5.3" and "3.5.19", respectively, so as to become comparison trademarks. In practical application, when the graph and trademark is searched through the graph codes, the quasi-registration information of the graph and trademark to be searched is obtained, the classification group of the commodity or service where the graph and trademark is located is confirmed, and the graph and trademark is searched according to the classification group and the graph codes to obtain a search result; by adopting a commodity or service classification mode, more accurate data can be obtained. For example, the graphical icon to be retrieved by the user obtained in step S1 of the present embodiment is supposed to be registered in the ninth category of the goods or services, so that the present embodiment performs a retrieval in the ninth category to obtain a retrieval result.
Meanwhile, in the present embodiment, it is necessary to determine which graphics code to search in combination with the saliency feature ranking, but in the present embodiment, the graphics code with high saliency is used for searching. In this embodiment, in step S3, the saliency arrays of the girl, the monkey and the basket are obtained, and the saliency of the girl and the monkey significantly exceeds that of the basket, so that the embodiment uses the graphic search on the girl and the monkey to obtain the comparison trademark, as shown in fig. 10, which is the partial comparison trademark obtained after the search on the ninth category through the graphic codes "2.5.3" and "3.5.19".
Step S6: and carrying out approximate judgment to obtain a final retrieval conclusion.
The step of performing approximate judgment to obtain a final retrieval conclusion comprises the following steps: carrying out image recognition on the comparison trademark to extract a graphic element corresponding to the graphic code, carrying out similarity comparison on the obtained graphic element and a graphic element of the to-be-retrieved graphic trademark corresponding to the graphic feature of the graphic code to obtain a comparison result, and sequencing the comparison result according to the sequence of similarity from high to low; in this embodiment, the graphic codes of the girl and monkey graphic features of the graphic trademark to be retrieved, which are confirmed by step S4, and the comparison trademarks are obtained through the graphic codes; by identifying the graphic elements of which the graphic codes contained in the comparison trademark are 2.5.3 and 3.5.19, comparing and approximately judging the graphic elements with the graphic elements of the to-be-retrieved graphic trademark, namely girls and monkeys, so that the similarity between the to-be-retrieved graphic trademark and each comparison trademark can be obtained, and then carrying out similarity sequencing; and then, combining all the graphic feature significances of the graphic trademarks to be searched to obtain a final searching conclusion and giving corresponding registration suggestions.
The embodiment confirms whether the picture of the graph trademark to be retrieved contains the first graph characteristic submitted by the user by identifying the picture of the graph trademark to be retrieved to obtain the graph characteristic contained in the picture, or the picture of the graph trademark to be retrieved is extracted and compared with the graph element basic feature points of the first graph features submitted by the user to confirm whether the picture of the graph trademark to be retrieved contains the first graph features submitted by the user, after the picture is confirmed to contain the first graph features, the graph codes corresponding to the graph trademark to be retrieved are obtained through the confirmed first graph features and a preset graph code library, finally, the graph trademark is retrieved through the graph codes to obtain a trademark retrieval result, and the whole process does not manually obtain the graph codes corresponding to the graph trademark to be retrieved, the image code corresponding to the trademark of the graph to be retrieved is automatically obtained by identifying the image of the trademark of the graph to be retrieved and confirming the image characteristics; compared with the existing graph trademark retrieval method for purely manually extracting graph features, the method overcomes the defects that the graph element codes of the traditional trademark are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result; compared with other methods similar to the method for obtaining the graphic codes by establishing a pattern library in which trademark patterns and graphic codes are in one-to-one correspondence and then comparing the patterns in the pattern library with the patterns in the pattern library one by one, the method provided by the invention confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by identifying the pattern of the trademark to be searched to obtain the graphic features contained in the pattern, or confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by extracting the pattern element basic feature points of the pattern of the trademark to be searched and the first graphic features submitted by the user and then performs character matching on the confirmed first graphic features and the preset pattern code library to obtain the graphic codes corresponding to the trademark to be searched without performing a large amount of image identification and comparison, the workload is greatly reduced, and the calculation and storage resources occupied by the pattern library are not needed; matched graphic codes can be obtained very quickly through character matching, and meanwhile, compared with the existing self-defined patterns and graphic code libraries, the obtained retrieval result is more in line with the standard of trademark registration, and the retrieval accuracy is improved. And automatically comparing according to the graphic features and the graphic elements to obtain an approximate judgment result, and obtaining a retrieval conclusion and a registration suggestion according to the approximate judgment result to realize artificial intelligent trademark retrieval. Further, through the comparison trade mark of accurate acquisition, can greatly reduce the quantity that the trade mark was compared, promote comparison efficiency effectively, accurate comparison data can promote the precision of comparing. Furthermore, by judging the significance, for the graphic trademark with complex constituent elements, the graphic features and the graphic codes thereof can be selected for retrieval through the significance, and the retrieval result is obtained through the graphic codes, so that a large amount of repeated or meaningless retrieval work is reduced, and the retrieval efficiency is greatly improved.
EXAMPLE five
As shown in fig. 11, 12, 3 and 13, a graphic trademark retrieval method includes the steps of:
step S1: acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved;
the step S1 of obtaining the picture and the first graphic feature of the graphic trademark to be retrieved may specifically be: acquiring a to-be-retrieved graphic trademark picture input by a user and a first graphic feature which is input by the user and used for describing a graphic element of the to-be-retrieved graphic trademark picture, and automatically acquiring the picture and a first graphic feature content by a system; for example, in this embodiment, as shown in fig. 12, the obtained to-be-retrieved graphic trademark picture submitted by the user obtains that the first graphic feature submitted by the user is "monkey".
In practical applications, when providing the picture of the trademark to be registered, the user may submit the category of the commodity or the service to be registered at the same time, for example, the category of the commodity or the service submitted by the user to be registered is the 09 th category in this embodiment.
Step S2: identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic;
the step S2 of recognizing the picture of the to-be-retrieved graphic trademark and determining whether the picture contains the first graphic feature may specifically include the following steps:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of the picture, wherein the basic composition elements are graphic elements;
describing the graphic elements through characters to obtain the graphic features of the to-be-retrieved graphic trademark picture, wherein the obtained graphic features of the to-be-retrieved graphic trademark picture are second graphic features;
and comparing the second graphic features with the first graphic features, and judging whether the second graphic features comprise the first graphic features.
In practical application, before identification, image preprocessing is required to be performed on a trademark picture of a graph to be retrieved, which can comprise the steps of graying, drying removal, normalization and the like.
The image recognition technology adopted in step S2 is a prior art based on picture content recognition in the field of computer vision technology, so that the specific implementation process for recognizing the preprocessed picture to obtain the graphic features included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the trademark picture of the graph to be retrieved.
For example, if the graph and trademark picture to be retrieved of the user obtained in step S1 in this embodiment is subjected to image recognition in step S2, it can be obtained that the basic constituent element is a monkey pattern, the monkey pattern is a graphic element of the picture, and the monkey pattern is described by text and then is a "monkey", and in this embodiment, the monkey pattern is submitted by the user, and therefore, it is necessary to compare and determine the "monkey" with the monkey to confirm that the first graphic feature "monkey" submitted by the user is included in the graph and trademark picture to be retrieved.
The identification of the preprocessed picture to obtain the second graphic feature included in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps. For example, step S2 can be implemented by:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of a graph trademark picture to be retrieved, wherein the basic composition elements are graph elements;
extracting characteristic points of graphic elements of the graphic trademark picture to be retrieved;
according to the first graph feature obtained in the past, obtaining a feature point of a first graph element corresponding to the first graph feature;
according to the first graphic feature obtained before, obtaining the feature point of the first graphic element corresponding to the first graphic feature, and 2 ways can be adopted: one way is as follows: obtaining a graphic element corresponding to the first graphic feature according to the first graphic feature, and extracting a feature point of the graphic element corresponding to the first graphic feature through image processing; the other mode is as follows: a mapping database with mutually corresponding graphic features, graphic elements and graphic element feature points can be preset, and the feature points of the graphic elements corresponding to the first graphic features can be obtained by mapping and matching the first graphic features and the mapping database;
and comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are contained in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, and judging whether the graphic trademark picture to be retrieved contains the first graphic characteristic according to the result.
The steps can be specifically as follows: comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic trademark picture to be retrieved, if the characteristic points of the graphic element corresponding to the first graphic characteristic are included in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, obtaining that the graphic trademark picture to be retrieved includes the first graphic characteristic, otherwise, obtaining that the graphic trademark picture to be retrieved does not include the first graphic characteristic.
And determining whether the to-be-retrieved graphic trademark picture contains the first graphic feature according to the result.
The image recognition technology adopted in step S2 is a prior art based on picture content recognition in the field of computer vision technology, and thus, a specific implementation process for recognizing the preprocessed picture to obtain the feature points included in the picture may include the steps of:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a feature point extraction model capable of extracting basic feature points of different types of graphic elements according to the training result;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on the segmented pictures through the feature point extraction model to obtain basic feature points of the graphic elements in each segmented picture.
Step S3: if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library;
if the first graphic feature is included, obtaining the corresponding graphic code according to the first graphic feature and a preset graphic code library comprises: matching the confirmed first graphic characteristics with preset graphic code library characters to obtain corresponding graphic codes; if the first graphic feature submitted by the user in step S1 in this embodiment is "monkey", the image recognition in step S2 confirms that the trademark of the graphic to be retrieved submitted by the user includes the first graphic feature "monkey" submitted by the user, and at this time, the first graphic feature "monkey" submitted by the user can be directly matched with the preset graphic code library text to obtain the corresponding graphic code. The method specifically comprises the following steps: matching description is carried out according to the confirmed first graphic characteristics and preset graphic code library characters, the description is similar or synonymous or comprises and is contained when the description is matched, and then the corresponding graphic codes are correspondingly obtained; for example, in this embodiment, it is determined according to step S2 that the graphic trademark picture to be retrieved by the user includes the first graphic feature submitted by the user as "monkey", and matching is performed in the preset graphic coding library according to the monkey, the matched characters are described as "primate, monkey, orangutan", and under the animal directory corresponding to the 3 codes, the graphic code corresponding to the monkey is "3.5.19", so that the graphic code whose graphic feature is the monkey of the graphic trademark picture to be retrieved by the user is "3.5.19".
In practical applications, if the picture of the graphic trademark to be retrieved is identified and whether the picture of the graphic trademark to be retrieved includes the first graphic feature is determined to be executed in step S2, and the result is that the picture of the graphic trademark to be retrieved does not include the graphic feature submitted by the user, any one of the following 2 steps may be executed:
(1) asking the customer to resubmit the first graphical feature and confirming that the newly submitted first graphical feature continues to perform step S2; the steps can be repeated until the first graphic features submitted by the user are contained in the graphic trademark to be retrieved submitted by the user; for example, in this embodiment, the to-be-retrieved graphic trademark picture submitted by the user is as shown in fig. 15, and the first graphic feature submitted by the user is a child, after the picture is identified and confirmed in step S2, it may be determined that the first graphic feature submitted by the user is not included in the to-be-retrieved graphic trademark picture provided by the user, and the user is required to provide a correct first graphic feature again; if the first graphical feature submitted by the user for the second time is not included in the picture, the customer may be asked to provide it again, or the following step (2) may be performed.
(2) Directly identifying the picture of the graph trademark to be retrieved submitted by the user, and obtaining a second graph characteristic contained in the picture; then, step S3 is executed to obtain the graphics code corresponding to the second graphics feature, and then step S4 is directly executed.
The image recognition technology adopted for directly recognizing the picture of the graph trademark to be retrieved submitted by the user and obtaining the second graph feature contained in the picture is the prior art based on picture content recognition in the technical field of computer vision, so that the specific implementation process for recognizing the preprocessed picture and obtaining the second graph feature contained in the picture can comprise the following steps:
1) pre-constructing a training database, collecting and inputting N images respectively containing different types of graphic elements by a computer, constructing a training image set for machine learning training, and obtaining training results of the different types of graphic elements;
2) constructing a picture recognition model according to the training result, wherein the picture recognition model can recognize graphic elements in the picture and obtain characters describing the graphic elements, namely the picture recognition model can recognize the graphic elements in the picture and obtain graphic features describing the graphic elements;
3) analyzing the preprocessed picture, extracting basic composition elements of the picture, namely graph elements, and then segmenting the picture according to the graph elements;
4) and carrying out picture identification on each divided picture through a picture identification model to obtain an identification result of the graphic elements in each divided picture, namely obtaining the graphic features of the graphic elements in each divided picture, wherein the obtained graphic features of the graphic elements in each divided picture are the second graphic features of the to-be-retrieved graphic trademark picture.
The identification of the preprocessed picture to obtain the second graphic feature included in the picture can be realized by adopting different image identification methods and steps according to different conditions besides the steps.
Step S4: searching a graphic trademark through graphic codes to obtain a comparison trademark;
in step S4, the obtaining of the comparison trademark by performing graphic trademark retrieval through graphic coding may specifically be: all trademarks containing graphic features corresponding to the graphic codes are obtained through the graphic codes in a trademark library of a trademark office and become comparison trademarks. Compared with other graph trademark retrieval methods, the graph to be registered is compared with all graph trademarks in the trademark library to obtain a comparison result, the graph code is limited to obtain the trademarks corresponding to the graph codes, the retrieval number is greatly reduced, meanwhile, because the trademark office is a graph code maker and an examination standard editor of the graph codes, the graph code library in the trademark office is used, the obtained result is extremely accurate, and the accuracy is greatly improved. For example, in this embodiment, it is determined according to step S2 that the graphic trademark picture to be retrieved by the user includes a graphic feature "monkey" submitted by the user, the graphic code corresponding to the first graphic feature "monkey" is "3.5.19", and all trademarks related to the graphic code are obtained from the trademark library through the graphic code "3.5.19" to become the retrieval result. In practical application, when the graph and trademark is searched through the graph codes, the quasi-registration information of the graph and trademark to be searched is obtained, the classification group of the commodity or service where the graph and trademark is located is confirmed, and the graph and trademark is searched according to the classification group and the graph codes to obtain a search result; by adopting a commodity or service classification mode, more accurate data can be obtained. For example, in the present embodiment, if the trademark of the graph to be registered by the user is obtained as the 09 th class at step S1, the trademark is compared with the part obtained after searching the 09 th class by the graph code "3.5.19" as shown in fig. 3.
Step S5: and carrying out approximate judgment to obtain a final retrieval conclusion.
The step of performing approximate judgment to obtain a final retrieval conclusion comprises the following steps: and carrying out image recognition on the comparison trademark to extract a graphic element corresponding to the graphic code, carrying out similarity comparison on the obtained graphic element and a graphic element of the to-be-retrieved graphic trademark corresponding to the graphic feature of the graphic code to obtain a comparison result, and sequencing the comparison result according to the sequence of similarity from high to low. In this embodiment, the graphic code of the graphic element monkey pattern of the graphic trademark to be retrieved, which is confirmed in step S3, corresponds to the graphic feature, and the comparison trademark is obtained through the graphic code; by identifying the figure element of figure code '3.5.19' contained in the comparison trademark, comparing and approximately judging the figure element with the monkey as the figure element of the figure trademark to be retrieved, the similarity between the figure trademark to be retrieved and each comparison trademark can be obtained, and then the similarity is sequenced.
In practical application, the performing the approximate judgment to obtain the final retrieval conclusion includes: and synthesizing the similarity comparison results corresponding to all the graphic features of the graphic trademark to be retrieved to obtain a final retrieval conclusion, and giving a corresponding registration suggestion.
Step S6: if the graph trademark to be retrieved contains characters, carrying out similarity comparison through the character retrieval character trademark to obtain a character similarity comparison result, and giving a corresponding registration suggestion by combining a final retrieval conclusion;
in this embodiment, as shown in fig. 12, the trademark to be registered submitted by the user contains a "fun" character in addition to the first graphic feature, and is a mixed trademark of graphics and characters, it is necessary to search for an approximate trademark of "fun" and obtain a comparison trademark, and as shown in fig. 13, after the judgment of the degree of approximation, "fun" is the same as several character trademarks previously applied, so the passage rate of the application of the "fun" character is extremely low, and the passage rate of the combined trademark can be judged to be extremely low by the trademark examination standard; therefore, the retrieval result is obtained by combining the graph retrieval, and the user can be recommended to split the combined trademark to register the combined trademark.
The embodiment confirms whether the picture of the graph trademark to be retrieved contains the first graph characteristic submitted by the user by identifying the picture of the graph trademark to be retrieved to obtain the graph characteristic contained in the picture, or the picture of the graph trademark to be retrieved is extracted and compared with the graph element basic feature points of the first graph features submitted by the user to confirm whether the picture of the graph trademark to be retrieved contains the first graph features submitted by the user, after the picture is confirmed to contain the first graph features, the graph codes corresponding to the graph trademark to be retrieved are obtained through the confirmed first graph features and a preset graph code library, finally, the graph trademark is retrieved through the graph codes to obtain a trademark retrieval result, and the whole process does not manually obtain the graph codes corresponding to the graph trademark to be retrieved, the image code corresponding to the trademark of the graph to be retrieved is automatically obtained by identifying the image of the trademark of the graph to be retrieved and confirming the image characteristics; compared with the existing graph trademark retrieval method for purely manually extracting graph features, the method overcomes the defects that the graph element codes of the traditional trademark are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result; compared with other methods similar to the method for obtaining the graphic codes by establishing a pattern library in which trademark patterns and graphic codes are in one-to-one correspondence and then comparing the patterns in the pattern library with the patterns in the pattern library one by one, the method provided by the invention confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by identifying the pattern of the trademark to be searched to obtain the graphic features contained in the pattern, or confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by extracting the pattern element basic feature points of the pattern of the trademark to be searched and the first graphic features submitted by the user and then performs character matching on the confirmed first graphic features and the preset pattern code library to obtain the graphic codes corresponding to the trademark to be searched without performing a large amount of image identification and comparison, the workload is greatly reduced, and the calculation and storage resources occupied by the pattern library are not needed; matched graphic codes can be obtained very quickly through character matching, and meanwhile, compared with the existing self-defined patterns and graphic code libraries, the obtained retrieval result is more in line with the standard of trademark registration, and the retrieval accuracy is improved. And automatically comparing according to the graphic features and the graphic elements to obtain an approximate judgment result, and obtaining a retrieval conclusion and a registration suggestion according to the approximate judgment result to realize artificial intelligent trademark retrieval. Further, through the comparison trade mark of accurate acquisition, can greatly reduce the quantity that the trade mark was compared, promote comparison efficiency effectively, accurate comparison data can promote the precision of comparing. By combining the character retrieval and the trademark examination standard, a registration suggestion with higher reference value can be given, and the use experience of a user is improved.
EXAMPLE six
As shown in fig. 14, a graphic trademark retrieval system includes a graphic code library 5 and:
the acquisition module 1 is used for acquiring the picture and the first graphic characteristic of the graphic trademark to be retrieved and preprocessing the picture and the first graphic characteristic;
the identification module 2 is used for identifying the picture of the graph trademark to be retrieved to obtain the graph characteristics or the characteristic points contained in the picture;
the confirming module 3 is used for confirming whether the graphic features or the contained feature points obtained by the identifying module contain the first graphic features or the corresponding feature points obtained by the obtaining module;
the code acquisition module 4 is used for acquiring a graphic code corresponding to the graphic trademark to be retrieved according to the confirmed first graphic feature and a preset graphic code library;
and the retrieval module 6 is used for carrying out graphic trademark retrieval through the graphic codes to obtain a trademark retrieval result.
The acquisition module acquires the picture of the graph trademark to be retrieved and the first graph characteristic provided by the user from the content input by the user, and simultaneously preprocesses the acquired picture; the identification module identifies the preprocessed picture obtained by the obtaining module to obtain the graphic characteristics contained in the picture; the confirmation module acquires the recognized graphic features from the recognition module and simultaneously acquires the first graphic features provided by the user from the acquisition module, and compares the first graphic features and the first graphic features to confirm; the code acquisition module is respectively connected with the confirmation module and the graphic code library, and can acquire the graphic codes through the graphic code library. The retrieval module comprises an inquiry module, and can retrieve the graphics trademark through the graphics code acquired by the code acquisition module to acquire a retrieval result.
The operation principle of the graph trademark retrieval system is as follows:
acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved; identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic; if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library; and searching the graphic trademark through the graphic code to obtain a trademark searching result.
The embodiment confirms whether the picture of the graph trademark to be retrieved contains the first graph characteristic submitted by the user by identifying the picture of the graph trademark to be retrieved to obtain the graph characteristic contained in the picture, or the picture of the graph trademark to be retrieved is extracted and compared with the graph element basic feature points of the first graph features submitted by the user to confirm whether the picture of the graph trademark to be retrieved contains the first graph features submitted by the user, after the picture is confirmed to contain the first graph features, the graph codes corresponding to the graph trademark to be retrieved are obtained through the confirmed first graph features and a preset graph code library, finally, the graph trademark is retrieved through the graph codes to obtain a trademark retrieval result, and the whole process does not manually obtain the graph codes corresponding to the graph trademark to be retrieved, the image code corresponding to the trademark of the graph to be retrieved is automatically obtained by identifying the image of the trademark of the graph to be retrieved and confirming the image characteristics; compared with the existing graph trademark retrieval method for purely manually extracting graph features, the method overcomes the defects that the graph element codes of the traditional trademark are only divided manually, the dividing work efficiency is low and the work energy consumption is huge, and obtains a relatively accurate retrieval result; compared with other methods similar to the method for obtaining the graphic codes by establishing a pattern library in which trademark patterns and graphic codes are in one-to-one correspondence and then comparing the patterns in the pattern library with the patterns in the pattern library one by one, the method provided by the invention confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by identifying the pattern of the trademark to be searched to obtain the graphic features contained in the pattern, or confirms whether the pattern of the trademark to be searched contains the first graphic features submitted by the user by extracting the pattern element basic feature points of the pattern of the trademark to be searched and the first graphic features submitted by the user and then performs character matching on the confirmed first graphic features and the preset pattern code library to obtain the graphic codes corresponding to the trademark to be searched without performing a large amount of image identification and comparison, the workload is greatly reduced, and the calculation and storage resources occupied by the pattern library are not needed; matched graphic codes can be obtained very quickly through character matching, and meanwhile, compared with the existing self-defined patterns and graphic code libraries, the obtained retrieval result is more in line with the standard of trademark registration, and the retrieval accuracy is improved.
In practical applications, the retrieving module may further include: the query module is used for searching the graphic trademark through the graphic code to obtain a comparison trademark; and the judging module is used for carrying out approximate judgment to obtain a final retrieval conclusion.
In the following, a graphic trademark retrieval device provided by an embodiment of the present invention is introduced, and a graphic trademark retrieval device described below and a graphic trademark retrieval method and system described above may be referred to each other.
As shown in fig. 15, the graphic trademark retrieval device according to the embodiment of the present invention includes:
a memory 100 for storing a computer program;
a processor 200, configured to implement the steps of the graphical trademark retrieval method according to any of the above embodiments when executing the computer program.
The following describes a readable storage medium provided by an embodiment of the present invention, and a readable storage medium described below and the above-described graphic trademark retrieval method, system, and apparatus may be referred to each other.
A readable storage medium having stored thereon a computer program which, when executed, implements the steps of the graphical trademark retrieval method as described in any of the above embodiments.
The above description is only an 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 disclosed herein are intended to be covered by the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.

Claims (15)

1. A graphic trademark retrieval method is characterized by comprising the following steps:
acquiring a picture and a first graphic characteristic of a graphic trademark to be retrieved;
identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph characteristic;
if the first graphic feature is included, acquiring a corresponding graphic code according to the first graphic feature and a preset graphic code library;
and searching the graphic trademark through the graphic code to obtain a trademark searching result.
2. The graphic trademark retrieval method of claim 1, wherein the obtaining of the corresponding graphic code according to the first graphic feature and a preset graphic code library comprises: and matching the obtained first graphic characteristics with preset graphic code library characters to obtain the corresponding graphic codes.
3. The graphic trademark retrieval method of claim 1, wherein the identifying the picture of the graphic trademark to be retrieved and confirming whether the picture contains the first graphic feature comprises:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of the picture, wherein the basic composition elements are graphic elements;
describing the graphic elements through characters to obtain the graphic features of the to-be-retrieved graphic trademark picture, wherein the obtained graphic features of the to-be-retrieved graphic trademark picture are second graphic features;
and comparing the second graphic features with the first graphic features, and judging whether the second graphic features comprise the first graphic features.
4. The graphic trademark retrieval method of claim 1, wherein the identifying the picture of the graphic trademark to be retrieved and confirming whether the picture contains the first graphic feature comprises:
analyzing the composition of the graph trademark picture to be retrieved;
extracting basic composition elements of a graph trademark picture to be retrieved, wherein the basic composition elements are graph elements;
extracting characteristic points of graphic elements of the graphic trademark picture to be retrieved;
obtaining a graphic element feature point corresponding to the first graphic feature according to the first graphic feature;
and comparing the characteristic points of the graphic element of the graphic trademark picture to be retrieved with the characteristic points of the graphic element corresponding to the first graphic characteristic, judging whether the characteristic points of the graphic element corresponding to the first graphic characteristic are contained in the characteristic points of the graphic element of the graphic trademark picture to be retrieved, and judging whether the graphic trademark picture to be retrieved contains the first graphic characteristic according to the result.
5. The graphic trademark retrieval method of claim 1, wherein the graphic trademark retrieval method includes: the method for judging the significance of the obtained first graphic features and sorting the first graphic features according to the significance comprises the following steps:
acquiring the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
judging the significance according to the positions and proportions of the graphic elements corresponding to the first graphic features in the picture;
and sorting the judgment results from high to low according to the significance.
6. The graphic trademark retrieval method of claim 1, wherein the graphic trademark retrieval method includes the steps of:
after identifying the picture of the graph trademark to be retrieved and confirming whether the picture contains the first graph feature, if the picture does not contain the first graph feature, the picture of the graph trademark to be retrieved is required to be provided with the first graph feature again or the picture of the graph trademark to be retrieved is subjected to image identification to obtain a second graph feature;
continuing to identify the picture of the graph trademark to be retrieved and confirming whether the picture contains the re-provided first graph characteristic until the picture contains the re-provided first graph characteristic;
and obtaining the corresponding graphic code according to the confirmed first graphic feature and a preset graphic code library or obtaining the corresponding graphic code according to the second graphic feature obtained by image recognition and the preset graphic code library.
7. The graphic trademark retrieval method of claim 6, wherein the identifying the picture of the graphic trademark to be retrieved to obtain the second graphic feature comprises: analyzing the composition of the graph and trademark picture to be retrieved, extracting basic composition elements of the picture, wherein the basic composition elements are the graph elements of the graph and trademark picture to be retrieved, and describing the graph elements through characters to obtain a second graph characteristic of the graph and trademark picture to be retrieved.
8. The graphic trademark retrieval method according to any one of claims 1 to 7, wherein the obtaining of the trademark retrieval result through graphic trademark retrieval by graphic coding comprises: searching a graphic trademark through graphic codes to obtain a comparison trademark; and carrying out approximate judgment to obtain a final retrieval conclusion.
9. The graphic trademark retrieval method of claim 8, wherein the performing of the approximate judgment to obtain the final retrieval conclusion comprises: and carrying out image recognition on the comparison trademark to extract a graphic element corresponding to the graphic code, carrying out similarity comparison on the obtained graphic element and a graphic element of the to-be-retrieved graphic trademark corresponding to the graphic feature of the to-be-retrieved graphic code to obtain a comparison result, sequencing the comparison result according to the similarity from high to low, synthesizing the similarity comparison results corresponding to all the graphic features of the to-be-retrieved graphic trademark to obtain a final retrieval conclusion, and giving a corresponding registration suggestion.
10. The graphic trademark retrieval method of claim 9, wherein the performing of the approximate judgment to obtain the final retrieval conclusion comprises: and synthesizing the similarity comparison results corresponding to all the graphic features of the graphic trademark to be retrieved, combining the significance of all the graphic features of the graphic trademark to be retrieved to obtain a final retrieval conclusion, and giving corresponding registration suggestions.
11. The graphic trademark retrieval method of claim 10, wherein the graphic trademark retrieval method includes: if the graph trademark to be searched contains characters, the character similarity comparison result is obtained after the similarity comparison is carried out on the character search character trademark, and corresponding registration suggestions are given according to the final search conclusion.
12. A graphic trademark retrieval system is characterized in that the graphic trademark retrieval system comprises a graphic code library and:
the acquisition module is used for acquiring the picture of the graph trademark to be retrieved and the first graph characteristic and preprocessing the picture;
the identification module is used for identifying the picture of the graph trademark to be retrieved to obtain the graph characteristics contained in the picture;
the confirming module is used for confirming whether the graph characteristics obtained by the identifying module contain the first graph characteristics obtained by the obtaining module;
the code acquisition module is used for acquiring a graphic code corresponding to the graphic trademark to be retrieved according to the confirmed first graphic characteristic and a preset graphic code library;
and the retrieval module is used for carrying out graphic trademark retrieval through the graphic codes to obtain a trademark retrieval result.
13. The graphic trademark retrieval system of claim 12, wherein the retrieval module comprises: the query module is used for searching the graphic trademark through the graphic code to obtain a comparison trademark; and the judging module is used for carrying out approximate judgment to obtain a final retrieval conclusion.
14. A graphic trademark retrieval apparatus, characterized by comprising:
a memory for storing a computer program;
a processor for implementing the steps of the graphical trademark retrieval method of any of claims 1 to 11 when executing the computer program.
15. A readable storage medium, characterized in that the readable storage medium has stored thereon a computer program which, when executed, implements the steps of the graphical trademark retrieval method of any of claims 1 to 11.
CN201910663424.9A 2019-07-22 2019-07-22 Method, system and device for searching graphic trademark and readable storage medium Pending CN112287145A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910663424.9A CN112287145A (en) 2019-07-22 2019-07-22 Method, system and device for searching graphic trademark and readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910663424.9A CN112287145A (en) 2019-07-22 2019-07-22 Method, system and device for searching graphic trademark and readable storage medium

Publications (1)

Publication Number Publication Date
CN112287145A true CN112287145A (en) 2021-01-29

Family

ID=74419109

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910663424.9A Pending CN112287145A (en) 2019-07-22 2019-07-22 Method, system and device for searching graphic trademark and readable storage medium

Country Status (1)

Country Link
CN (1) CN112287145A (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104636428A (en) * 2014-12-24 2015-05-20 四川超凡知识产权服务股份有限公司 Trademark recommendation method and device
CN104809142A (en) * 2014-01-29 2015-07-29 北京瑞天科技有限公司 Trademark inquiring system and method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104809142A (en) * 2014-01-29 2015-07-29 北京瑞天科技有限公司 Trademark inquiring system and method
CN104636428A (en) * 2014-12-24 2015-05-20 四川超凡知识产权服务股份有限公司 Trademark recommendation method and device

Similar Documents

Publication Publication Date Title
US10152650B2 (en) Trademark retrieval method, apparatus and system, and computer storage medium
EP2296097B1 (en) Method and apparatus for representing and searching for an object in an image
Iqbal et al. Combining structure, color and texture for image retrieval: A performance evaluation
EP3660703A1 (en) Method, apparatus, and system for identifying device, storage medium, processor, and terminal
US20010024517A1 (en) Method and device for seeking images based on the content taking into account the content of regions of interest
CN111611936B (en) Automatic identification system for similar vector diagrams in CAD drawings
CN112926601A (en) Image recognition method, device and equipment based on deep learning and storage medium
CN116069969A (en) Image retrieval method, device and storage medium
CN117115569B (en) Automatic object image identification and classification method and system based on machine learning
CN112256909A (en) Method, system and device for searching graphic trademark and readable storage medium
CN116721246B (en) Continuous frame point cloud rapid labeling method and system
CN111611935B (en) Automatic identification method for similar vector diagrams in CAD drawing
CN116661786A (en) Design page generation method and device
CN117150138A (en) Scientific and technological resource organization method and system based on high-dimensional space mapping
CN112287145A (en) Method, system and device for searching graphic trademark and readable storage medium
Gonzalez-Soler et al. Semi-synthetic data generation for tattoo segmentation
CN116257877A (en) Data classification grading method for privacy calculation
CN112949305B (en) Negative feedback information acquisition method, device, equipment and storage medium
CN114494730A (en) Trademark automatic classification processing system based on image recognition
CN114077682A (en) Intelligent recognition matching processing method and system for image retrieval and storage medium
CN112650874A (en) Method for realizing graph trademark retrieval
CN114332599A (en) Image recognition method, image recognition device, computer equipment, storage medium and product
Lengauer et al. Sketch-Aided Retrieval of Incomplete 3D Cultural Heritage Objects.
CN111177301A (en) Key information identification and extraction method and system
JP2001101226A (en) Document group sorter and document group sorting method

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