Patent drawing symbol description output method and device / system therefor

Abstract

A patent drawing code recognition method according to one embodiment of the present invention may include the steps of: learning a plurality of patent drawing samples to construct a drawing code position recognition model and a drawing code recognition model; inputting a patent drawing to be recognized as a drawing code; recognizing the position of the drawing code included in the patent drawing using the drawing code position recognition model; cutting out the drawing code at the recognized position from the patent drawing as an image fragment; and recognizing the drawing code included in the image fragment using the drawing code recognition model.

Description

[Technical Field]

[0001] This specification proposes a patent drawing reference numbers description output method, device and system therefor. [Background technology]

[0002] Knowledge information content such as papers and patent documents generally consists of a large number of pages. Such content includes drawings, mathematical formulas, and text explaining the drawings. In particular, due to formatting limitations, drawings and associated text are often placed on different pages. This forces readers of the content to read the content multiple times, flipping back and forth between pages to understand the explanations for the drawings and mathematical formulas, resulting in unnecessary wastage of time and effort in acquiring knowledge information.

[0003] Recently, to solve this problem, a method has been proposed that uses optical character recognition (OCR) technology to precisely recognize letters and / or numbers in knowledge information content, thereby improving the speed of accessing required information within the content. However, this method also only provides a simple character-based search function, making it difficult to quickly obtain information related to drawings.

[0004] However, to date, there has been no solution that can quickly acquire knowledge information by utilizing content format structure information to extract link information between drawings and content and providing a drawing-centric content analysis method. Summary of the Invention [Problem to be solved by the invention]

[0005] Therefore, a technical problem to be solved by the present invention is to provide a solution for efficiently analyzing knowledge information content centered on drawing information.

[0006] Specifically, the technical problem that the present invention aims to solve is to provide a method and apparatus for identifying characters (especially, explanations of drawing symbols) associated with drawing symbols through optical character reading and / or machine learning of the drawing symbols.

[0007] Another technical problem that the present invention aims to solve is to provide a method for searching for drawings based on characters, as well as for searching for characters (especially, drawing symbol descriptions) based on drawing symbols.

[0008] Specifically, another technical problem that the present invention aims to solve is to provide a drawing-centric content analysis method by filtering all drawings containing drawing codes based on characters (especially, the description of the drawing code) associated with the drawing code.

[0009] Another technical problem that the present invention aims to solve is to provide intuitive drawing-related information to a user by locating an explanation of a matching drawing symbol in an area where the drawing symbol is displayed on the drawing.

[0010] Specifically, another technical problem that the present invention aims to solve is to provide a solution for positioning / moving the description of the drawing symbols to an appropriate area in response to changes in the state of the drawing.

[0011] The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned can be clearly understood by a person skilled in the art of the present invention from the following description. [Means for solving the problem]

[0012] A patent drawing code recognition method according to one embodiment of the present invention may include the steps of: learning a plurality of patent drawing samples to construct a drawing code position recognition model and a drawing code recognition model; inputting a patent drawing to be recognized as a drawing code; recognizing the position of the drawing code included in the patent drawing using the drawing code position recognition model; cutting out the drawing code at the recognized position from the patent drawing as an image fragment; and recognizing the drawing code included in the image fragment using the drawing code recognition model. [Effects of the Invention]

[0013] According to one embodiment of the present invention, by displaying the symbols displayed in the drawings instead of the description of the symbols, a user can easily and intuitively obtain information about the invention from the drawings. In particular, since all related drawings can be filtered based on the description of the symbols, a drawing-centric analysis of knowledge information content can be performed by performing an intensive analysis of the filtered drawings.

[0014] According to another embodiment of the present invention, not only the drawing but also the sentence and paragraph containing the description of the drawing symbol can be visually identified based on the description of the drawing symbol, so that the user can quickly access desired information through a large amount of text using the description of the drawing symbol as a medium.

[0015] Furthermore, according to yet another embodiment of the present invention, even when a drawing is enlarged, reduced, moved, or enlarged, the explanation of the drawing symbols can be positioned in a highly readable area within the changed drawing, thereby providing convenience in analyzing knowledge information content using drawings in various analysis environments.

[0016] The effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description. [Brief explanation of the drawings]

[0017] [Figure 1] 1 illustrates an example of a patent drawing provided in accordance with one embodiment of the present invention. [Figure 2] 2 is a flowchart illustrating a method for recognizing drawing symbols according to an embodiment of the present invention. [Figure 3] 1 is a diagram illustrating a method for constructing a recognition model for a drawing symbol according to an embodiment of the present invention; [Figure 4] 1 is a flowchart illustrating a method for extracting a description of a drawing symbol according to an embodiment of the present invention. [Figure 5] 10A and 10B are diagrams illustrating an example of extracting the description of drawing symbols according to an embodiment of the present invention; [Figure 6] 1 is a flowchart illustrating a method for outputting a description of a drawing code corresponding to a drawing code according to an embodiment of the present invention. [Figure 7] 1 is a flowchart illustrating a method for outputting a description of a drawing symbol using a Scalable Vector Graphics (SVG) image according to an embodiment of the present invention. [Figure 8] 10A and 10B are diagrams illustrating a method for outputting a description of a drawing symbol using an SVG image according to an embodiment of the present invention; [Figure 9] 10A and 10B are diagrams illustrating a method for outputting a description of a drawing symbol using an SVG image according to an embodiment of the present invention; [Figure 10] FIG. 1 is a diagram illustrating a patent document related to an embodiment of the present invention. [Figure 11] 10 is a diagram illustrating a drawing interface in which drawing symbols and description of drawing symbols are linked to each other according to an embodiment of the present invention; [Figure 12] 10 is a diagram illustrating a drawing interface in which drawing symbols and description of drawing symbols are linked to each other according to an embodiment of the present invention; [Figure 13] FIG. 10 is a diagram illustrating a keyword setting interface according to an embodiment of the present invention. [Figure 14] 10 is a diagram illustrating an example of inter-category linkage using the description of drawing symbols as a medium according to an embodiment of the present invention; [Figure 15]10 is a diagram illustrating an example of inter-category linkage using the description of drawing symbols as a medium according to an embodiment of the present invention; [Figure 16] 1 is a diagram illustrating a patent information search system according to an embodiment of the present invention. [Figure 17] FIG. 2 is a block diagram of a web server according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] The technology described below can be modified in various ways and can have various embodiments, so a specific embodiment will be illustrated in the drawings and described in detail. However, this does not limit the technology described below to a specific embodiment, and it should be understood that the technology described below includes all modifications, equivalents, and alternatives that fall within the spirit and technical scope of the technology described below.

[0019] Terms such as "first," "second," "A," and "B" may be used to describe various components, but the components are not limited by these terms and are used solely to distinguish one component from another. For example, a first component may be designated as a second component, and similarly, a second component may be designated as a first component, without departing from the scope of the technology described below. The term "and / or" includes a combination of multiple related listed items or any of multiple related listed items. For example, "A and / or B" may be interpreted as meaning "at least one of A or B." Also, " / " may be interpreted as "and" or "or."

[0020] In the terms used in this specification, singular expressions should be understood to include plural expressions unless the context clearly dictates otherwise, and terms such as "comprise" should be understood to mean the presence of a stated feature, number, step, operation, component, part, or combination thereof, but not to exclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0021] Before proceeding to a detailed description of the drawings, it should be clarified that the division of components in this specification is merely a division according to the main function of each component. That is, two or more components described below may be integrated into one component, or one component may be divided into two or more components according to more specific functions. Furthermore, each component described below may additionally perform some or all of the functions performed by other components in addition to its own main function, and some of the main functions performed by each component may be exclusively performed by other components.

[0022] Furthermore, in performing a method or method of operation, the steps of achieving the method may occur in a different order than specified, unless the context clearly dictates a particular order, i.e., the steps may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in reverse order.

[0023] FIG. 1 illustrates an example of a patent drawing provided in accordance with one embodiment of the present invention.

[0024] More specifically, FIG. 1(a) is an example of a typical patent drawing, and FIG. 1(b) is an example of a patent drawing in which the drawing numbers are replaced with the description of the drawing numbers according to one embodiment of the present invention.

[0025] Knowledge information content, such as papers and patent documents, often uses drawings as a means of explaining information to users more easily and intuitively. Users can more easily and efficiently understand the content of the knowledge information content by simultaneously understanding the drawings and the explanations for the drawings. However, knowledge information content is generally composed of a large number of pages, and users must read the pages multiple times to simultaneously understand the drawings and the explanations for the drawings. This has been a significant obstacle to users' easy and efficient understanding of the content of the knowledge information content.

[0026] In particular, in the case of patent document drawings, which are among knowledge information contents, it is common for each element to be described with a drawing symbol, as shown in Figure 1(a). These drawing symbols are assigned arbitrarily by the inventor for the convenience of describing the drawings, and therefore are characterized by lack of regularity and great variation from patent document to patent document. Therefore, in order for users to understand the contents of the drawings (especially the explanations for each drawing symbol), it is very important to simultaneously understand the drawings and the explanations of the drawings. However, this not only causes inconvenience to users, but also increases the consumption of time and money.

[0027] Therefore, in this specification, we propose a patent drawing provision technology that directly replaces each drawing code with a corresponding drawing code explanation, as shown in Figure 1(b), so that users can understand the invention more easily and efficiently through drawings.

[0028] This patent drawing provision technology can be broadly divided into three technologies as follows: 1. Drawing code recognition technology 2. Technology for extracting explanations of drawing symbols 3. Technology to replace recognized drawing symbols with extracted drawing symbol explanations

[0029] The above-mentioned techniques 1 to 3 may be integrated into one technique and performed sequentially depending on the embodiment, or may be performed / used independently as individual techniques. The above-mentioned techniques 1 to 3 will be described in detail below with reference to the following drawings.

[0030] The patent drawing providing technology proposed in this specification will be described based on the case where it is applied to a website that provides a patent document search service. Therefore, the embodiments described below can be interpreted as being performed by a web server that manages / operates a patent document search website. However, without being limited thereto, the embodiments of this specification can also be applied to programs, applications, etc. that provide patent document search services. In this case, the website can be described as a program or application, and the web server can be described as a program server or application server, respectively.

[0031] In this specification, patent documents are used as an example of knowledge information content, but the present invention is not limited to this and can of course be expanded and applied to various knowledge information content such as papers and books.

[0032] FIG. 2 is a flowchart illustrating a method for recognizing a drawing code according to an embodiment of the present invention.

[0033] Referring to Figure 2, first, the web server can learn from a plurality of patent drawing samples to build a drawing code location recognition model and a drawing code recognition model (S201). The drawing code location recognition model and the drawing code recognition model can all be built based on deep learning technology.

[0034] More specifically, the web server can recognize the positions of drawing symbols contained in multiple patent drawing samples using deep learning technology. An example of the deep learning technology used here is FCN (Fully Convolutional Network). FCN is a deep learning model useful for determining whether text exists within image pixels and is a modified model derived from CNN (Convolutional Neural Network). Unlike existing CNNs, FCNs use only convolutional layers (i.e., convolutionalization) instead of fully connected layers. Due to this feature, FCNs, unlike CNNs, do not lose image position information, making them extremely useful for recognizing the position information of objects (especially text) contained within an image.

[0035] The web server can use this FCN to recognize the positions of drawing symbols included in multiple patent drawing samples and extract common features from the recognized positions of the drawing symbols. For example, the web server can extract common features of the positions of drawing symbols, such as not being located in the center of the drawing, not being located on the periphery of the drawing, and not being located overlapping each other. The web server can also learn the positions of the drawing symbols and extract various common features as a result of learning, including, but not limited to, the above. The web server can build a drawing symbol position recognition model based on the extracted common features. The drawing symbol position recognition model can receive patent drawings and recognize and output the positions of drawing symbols included in the patent drawings based on the extracted common features.

[0036] That is, the web server can learn the positions of drawing symbols contained in a plurality of patent drawing samples, extract common features, and build a drawing symbol position recognition model based on the extracted common features.

[0037] The web server that has constructed the location recognition model of the drawing code can construct a recognition model of the drawing code using the location recognition model of the drawing code, which will be described in more detail below with reference to FIG.

[0038] FIG. 3 is a diagram illustrating a method for constructing a recognition model for a drawing symbol according to an embodiment of the present invention.

[0039] To build a drawing symbol recognition model, the web server first uses a previously built drawing position recognition model to recognize the positions of drawing symbols 10-14, 16, and 18 included in the patent drawing sample 301. The web server can then cut out and collect the drawing symbols 10-14, 16, and 18 at the recognized positions from the patent drawing sample 301 into image fragments 302. The web server can recognize the drawing symbols 10-14, 16, and 18 included in each of the collected image fragments 302 using image character recognition technology (e.g., deep learning technology (particularly, C-RNN) and / or OCR (Optical character recognition)).

[0040] The reason for dividing drawing codes 10-14, 16, and 18 into small images and recognizing them is that recognizing drawing codes 10-14, 16, and 18 in image fragment 302 containing only drawing codes not only has a much higher recognition rate and accuracy, but also a much faster recognition speed, than recognizing drawing codes 10-14, 16, and 18 in the entire patent drawing 301. In fact, when a recognition simulation was conducted by dividing drawing codes into small images according to this method, the recognition accuracy was recorded as 99.43%, and the drawing code recognition time was recorded as within 0.02 seconds per drawing, making it possible to even recognize drawing codes in real time for each drawing.

[0041] The web server can construct a recognition model of the drawing codes by extracting common features from the recognized drawing codes 10 to 14, 16, and 18. The common features that the web server can derive include, for example, a feature that the drawing codes are composed of numbers, English letters, or a combination thereof, and a feature that the drawing codes are five characters or less in length, but are not limited thereto.

[0042] The web server can build a recognition model of the drawing code based on the extracted common features, and the recognition model of the drawing code can recognize and output the drawing code included in each image piece based on the extracted common features.

[0043] That is, the web server can learn the drawing codes included in the image fragments, extract common features, and build a recognition model for the drawing codes based on the extracted common features.

[0044] The drawing code position recognition model and drawing code recognition model thus constructed are used to recognize drawing codes in patent drawings selected / input by the user.

[0045] 2 again, the web server may then input / select a patent drawing to be recognized by the drawing code (S202). More specifically, the web server may receive a selection / input for a specific patent drawing (or a specific patent drawing) from a user device, which is a client device.

[0046] Next, the web server recognizes the position of the drawing codes included in the input / selected patent drawing using the drawing code position recognition model established in step S201 (S203), and then extracts the drawing codes at the recognized positions from the patent drawing to obtain image fragments including the drawing codes (S204). The web server then recognizes the drawing codes included in the obtained image fragments using the drawing code recognition model established in step S201 (S205). In this embodiment, the web server may generate a single image by collecting a predetermined number of image fragments (e.g., 200) and recognize the drawing codes included in the multiple image fragments included in the generated image using image character recognition technology (e.g., deep learning technology (e.g., C-RNN) and / or OCR (Optical Character Recognition)). The detailed description of steps S203 to S205 is essentially the same as that described in detail with reference to FIG. 3, and therefore, a repeated description will be omitted.

[0047] According to the embodiment proposed in Figures 2 and 3, the web server first determines the location of the drawing code and then selectively recognizes only the drawing code at a specific location. This not only significantly reduces the recognition time (i.e., improves recognition speed) compared to the conventional technology that recognized drawing codes for the entire drawing, but also significantly improves the recognition accuracy of the drawing code.

[0048] FIG. 4 is a flowchart illustrating a method for extracting a description of a drawing symbol according to an embodiment of the present invention.

[0049] According to one embodiment of the present invention, a web server can extract the description of a drawing symbol from a patent specification. Patent documents can be broadly categorized into patent specifications and drawings, and the description of a drawing symbol can be extracted from the patent specification. A detailed description of the structure of patent documents will be described in detail below with reference to FIG. 10. In particular, the web server can extract the description of a drawing symbol corresponding to the drawing symbol recognized by the embodiment proposed in FIGS. 2 and 3 from the patent specification.

[0050] For this purpose, the web server may first establish rules for extracting the description of the drawing symbols (S401). The rules for extracting the description of the drawing symbols may be established in various embodiments / methods, but in this specification, taking into consideration that the format of patent documents differs depending on the country in which the patent document is filed, the following establishment embodiment is proposed based on text mining technology.

[0051] The web server can first classify multiple patent specification samples by filing country and then extract common features for each of the classified filing countries, based on at least one of the following: the relative position of the drawing code description relative to the drawing code, the format applied to the drawing code description or the drawing code, and the filing year.

[0052] For example, common features for Korea include the fact that the description of the drawing code precedes the drawing code and that the drawing code is enclosed in parentheses. Common features for the United States include the fact that the description of the drawing code precedes the drawing code, that the drawing code is formatted in bold, and that, unlike Korea, the drawing code is not separated by parentheses. Some filing countries may have a "Code Description" category in which the drawing code and its description are created separately. This can also be extracted as a common feature, and the "Code Description" category can be used later to extract the drawing code description. Furthermore, by studying various patent specification samples, the web server can extract a variety of common features for each filing country.

[0053] The web server can then establish a rule (or model) for extracting the description of the drawing symbols based on the common features extracted in this way, and then use this rule to extract the description of the drawing symbols from the patent specification (S402).

[0054] More specifically, if the drawing code description extraction rule is established based on the relative position of the drawing code description to the drawing code (or if the drawing code description extraction rule includes a rule for the relative position of the drawing code description), the web server can search for the drawing code contained in the patent specification. Furthermore, the web server can predict the position of the drawing code description corresponding to the searched drawing code based on the established relative position rule, and extract the characters at the predicted position as the drawing code description.

[0055] For example, if a rule is established that the description of a drawing code must be placed before the drawing code, the web server can search for drawing code 16 in a patent specification containing the following description, and then extract the character "bolt" that appears before drawing code 16 as the description of the drawing code for drawing code 16.

[0056] -Bolt 16

[0057] As in the above example, when the description of a drawing code is composed of one word, there is little possibility of an error in extracting the description of the drawing code, but when it corresponds to a compound word composed of multiple words, there may be ambiguity as to which words should be extracted as the description of the drawing code. To resolve this ambiguity, the present specification proposes an embodiment shown in FIG.

[0058] FIG. 5 is a diagram illustrating an example of extracting the description of a drawing code according to an embodiment of the present invention.

[0059] In the present invention, which extracts the description of a drawing symbol based on its relative position relative to the drawing symbol, a problem may arise in that it is unclear to what extent the description of the drawing symbol should be extracted in the case of a compound word such as "organic light emitting display 10" illustrated in the drawings. For example, if the description of the drawing symbol is extracted based on the rule that the description of the drawing symbol is placed before the drawing symbol, "display (1)," "light emitting display (2)," "organic light emitting display (3)," etc. can be extracted as the description of the drawing symbol from the example of FIG.

[0060] Therefore, in this specification, in order to resolve such ambiguity / ambiguity, when multiple drawing symbol descriptions are extracted, an embodiment for selecting an accurate drawing symbol description from among them is proposed as follows.

[0061] As an example, if multiple drawing code descriptions are extracted, the web server may search the extracted drawing code descriptions in the patent specification, and determine and extract the drawing code description that is most frequently searched for as the final drawing code description. Taking the present drawing as an example, if the web server searches the patent specification for "display," "emissive display," and "organic light emitting display" respectively, and finds that "organic light emitting display" is searched three times, "emissive display" is searched zero times, and "display" is searched once, then "organic light emitting display" may be extracted as the final drawing code description.

[0062] In this case, the number of words extracted by the web server as candidates for the description of a drawing code may be limited to a predetermined number based on the drawing code description data accumulated to date. More specifically, the web server may create a database for patent drawing search by storing the results of recognition of drawing codes and drawing code descriptions in a database. In this case, the web server may compile statistics on the number of compound words extracted as description of a drawing code based on the drawing code description data accumulated to date in the database, and set the number of words with a statistically low extraction rate / probability as the limit number. For example, if the analysis of the drawing code description data accumulated to date shows that the rate at which compound words of five or more words are extracted as description of a drawing code is 0.0001%, the web server may limit the number of words that can be extracted as description of a drawing code to a maximum of four. In this case, when the web server extracts description of a drawing code according to the extraction rules for description of a drawing code, it may extract only a maximum of four words as description of a drawing code.

[0063] Although not shown in this flowchart, the web server can establish error extraction rules to determine whether there are errors in the extracted drawing code descriptions in order to further improve the accuracy of the extraction of drawing code descriptions, and can correct / complement any errors that are found.

[0064] More specifically, the web server can extract drawing symbol descriptions from a sample of patent specifications based on the established drawing symbol description extraction rules, and can search the extracted drawing symbol descriptions from the patent specifications. If there are drawing symbol descriptions that were not found in the patent specifications as a result of the search, the web server can classify the drawing symbol descriptions as drawing symbol descriptions containing errors. The web server can extract (i.e., learn) common features from the drawing symbol descriptions containing errors and establish error extraction rules (or models). The web server can use the established error extraction rules to determine whether the drawing symbol descriptions extracted from the patent specifications contain errors.

[0065] Examples of common features include the feature of incorrectly extracting the number 0 as the alphabet o, O, or D, the feature of incorrectly extracting the number 9 as the number 0, the feature of extracting non-noun parts of speech such as adjectives like "~na", conjunctions, and adverbs, and the feature of extracting symbols such as !, @, #, $, %, ^, (, ).

[0066] If the web server determines that there is an error in the description of the drawing code based on the established error detection rules, it can complement / correct the extracted description of the drawing code according to a predetermined method. For example, if the web server finds an error in which a number / alphabet is mistakenly recognized as an alphabet / number, it can replace the number / alphabet with the corresponding alphabet / number, and if the web server finds an error in which a part of speech or symbol that is not a noun is included, it can delete the part of speech or symbol.

[0067] Furthermore, the web server can build a drawing code description extraction model with improved performance by learning established rules for extracting drawing code descriptions and error extraction rules based on deep learning technology (CNN, RNN (Recurrent Neural Network), or a combination of these). In particular, the web server can build a drawing code description extraction model with extremely high recognition rate and accuracy by utilizing the correct answer set for drawing codes and drawing code descriptions that has been accumulated to date in the patent drawing search database that has already been built, and use it to extract drawing code descriptions.

[0068] Meanwhile, although not mentioned separately above, in order to extract corresponding drawing codes and their descriptions, drawing codes must be recognized with high accuracy not only in drawings but also in the specification. This is because the drawing description extraction rules proposed in this specification are premised on extracting drawing code descriptions based on "drawing codes." In other words, the specification and drawings can be considered to be interrelated with each other through the drawing codes. Therefore, the web server can determine whether drawing codes have been accurately extracted in each of the specification and drawings, and can complement drawing codes that are determined to have been inaccurately extracted. This can further improve the matching accuracy between drawing codes and their descriptions.

[0069] In a first embodiment, the web server can improve the accuracy of recognizing drawing codes for drawings by cross-matching drawing codes extracted from drawings with drawing codes included in the patent specification. More specifically, the web server can search for drawing codes recognized in patent drawings in the above-described embodiment from the patent specification corresponding to the patent drawing. If a drawing code recognized in a patent drawing is found in the patent specification, the web server can determine the recognized drawing code as an appropriate drawing code and set it as the final drawing code. Conversely, if the recognized drawing code is not found in the patent specification, the web server can determine the drawing code as an inappropriate drawing code and search the patent specification for characters / words / terms that have a shape similarity to the drawing code at a predetermined rate or more and set it as the final drawing code.

[0070] For example, if 36D is recognized in the drawing but 36D is not found in the actual specification, the web server can determine 360, which has the highest shape similarity to 36D, as the final drawing code.

[0071] Similarly, in a second embodiment, the web server can improve the accuracy of recognizing drawing symbols for a specification by cross-matching drawing symbols recognized in the specification with drawing symbols recognized in the drawings. More specifically, the web server can search for drawing symbols recognized in a patent specification from among drawing symbols recognized in the patent drawings corresponding to the specification as in the above embodiment. If a drawing symbol recognized in the patent specification is found among drawing symbols recognized in the patent drawings, the web server can determine the recognized drawing symbol as an appropriate drawing symbol and set it as the final drawing symbol. Conversely, if the recognized drawing symbol is not found among drawing symbols recognized in the patent drawings, the web server can determine the recognized drawing symbol as an inappropriate drawing symbol and search for characters / words / terms with a shape similarity to the recognized drawing symbol at a predetermined rate or higher among drawing symbols recognized in the patent drawings and set it as the final drawing symbol.

[0072] For example, if 36D is recognized in the patent specification but 36D is not found among the drawing symbols recognized in the actual patent drawings, the web server can determine 360, which has the highest shape similarity to 36D, as the final drawing symbol from among the drawing symbols recognized in the patent drawings.

[0073] That is, according to the first embodiment, the web server can complement the drawing symbols recognized in the patent drawings by matching them with the drawing symbols recognized in the patent specification, and according to the second embodiment, the web server can complement the drawing symbols recognized in the patent specification by matching them with the drawing symbols recognized in the patent drawings.

[0074] The first and second embodiments described above can be used selectively or in combination depending on the purpose and effect. When used in combination, only the drawing codes that match between the patent drawings and the patent specification can be determined / extracted / confirmed as the final drawing codes, which significantly reduces the probability of drawing code recognition errors compared to when the first and second embodiments are applied selectively.

[0075] Having considered the method for recognizing the reference symbols and their descriptions in detail above, we will now propose a method for outputting the recognized reference symbols in place of their descriptions.

[0076] In particular, the method of outputting the drawing symbol explanation proposed in this specification is different from the conventional technology in that it is not related to changes in the state of the drawing itself (e.g., movement, rotation, enlargement or reduction), and the relative position within the drawing is always maintained / fixed the same, so that even if the user changes the state of the drawing itself, the drawing symbol explanation can always accurately track and replace the position of the drawing symbol.

[0077] FIG. 6 is a flowchart illustrating a method for outputting a description of a drawing code corresponding to a drawing code according to an embodiment of the present invention.

[0078] 6, the web server first recognizes the size of the patent drawing and the position of the drawing code contained in the patent drawing, and obtains the relative position coordinates of the drawing code within the patent drawing (S601). That is, the web server obtains the relative position coordinates of the drawing code with respect to the size of the patent drawing. The reason for obtaining the relative position coordinates in this way is to accurately track the position of the drawing code even if the state of the drawing changes, as mentioned above.

[0079] Next, the web server may set the relative position coordinates acquired in the previous step as the relative position coordinates of the drawing code description corresponding to the drawing code (S602).

[0080] Finally, the web server can output the drawing symbol explanation at the set relative position coordinates (S603). At this time, the drawing symbol explanation can be output in the form of an icon / GUI (Graphical User Interface) with an opaque background color, and since it is output at the same relative position coordinates as the drawing, it covers / covers at least a portion of the drawing symbol. In other words, the drawing symbol explanation is output at the same position as the drawing symbol in the drawing, and therefore replaces the drawing symbol. For example, in FIG. 1(a), all drawing symbols are replaced by the drawing symbol explanation as in FIG. 1(b).

[0081] In this embodiment, the case where the description of the drawing symbol is output to cover / enclose at least a portion of the drawing symbol is illustrated, but the present invention is not limited thereto, and the description of the drawing symbol may be output anywhere as long as it is in a position corresponding to the drawing symbol (e.g., above / below / left / right / diagonally / adjacent to the drawing symbol, including the position that covers / encloses at least a portion). If the description of the drawing symbol is to be output in an area adjacent to the drawing symbol, the web server may set / assign the coordinates obtained by shifting the relative position coordinates of the drawing symbol acquired in step S601 in a predetermined direction and / or a predetermined distance to the relative coordinates of the description of the drawing symbol. Hereinafter, for convenience of explanation, a specific embodiment in which the description of the drawing symbol replaces the drawing symbol will be described in detail based on the embodiment of FIG. 6 described above.

[0082] 7 to 9 are flowcharts illustrating a method for outputting a description of a drawing symbol using an SVG image according to an embodiment of the present invention.

[0083] More specifically, Figure 7 is a flowchart illustrating a method for outputting a description of a drawing symbol using an SVG (Scalable Vector Graphics) image according to one embodiment of the present invention, and Figures 8 and 9 are figures illustrating a method for outputting a description of a drawing symbol using an SVG image according to one embodiment of the present invention.

[0084] 7 and 8, the web server first generates an SVG image 802 of the same size as patent drawing 801, overlaps SVG image 802 on patent drawing 801, and then fixes it to patent drawing 801 (S701). That is, a transparent SVG image 802 of the same size as patent drawing 801 is generated, and then it is fixed to patent drawing 801 while covering the entire patent drawing 801. As a result, SVG image 802 is characterized by being either unrecognizable to the user or changing state in the same way as patent drawing 801 changes. For reference, an SVG image is an image file format based on XML (Extensible Markup Language) for representing two-dimensional vector graphics, and is characterized by its ability to maintain quality even when the image state changes (especially when enlarged).

[0085] 7 and 9, the web server can then write / add / display / assign 903 a reference symbol description 902 to a predetermined relative position coordinate 901 within the SVG image 802 (S702). Here, the predetermined relative position coordinate 901 refers to the relative position coordinates of the reference symbol and the reference symbol description obtained in steps S601 and S602 of Fig. 6. The reference symbol description 902 can be written / added / displayed / assigned within the SVG image 802 in the form of an icon / GUI with an opaque background color.

[0086] As a result, the drawing code description 902 is written / added / displayed / assigned to a position 901 corresponding to the drawing code in the SVG image 802 (903). The web server can output the SVG image 903, in which the drawing code description 902 has been written / added / displayed / assigned, superimposed on the drawing 801, so that the drawing code of the drawing 801 is covered by the drawing code description 902 written / added / displayed / assigned on the SVG image 903 (S703).

[0087] For example, if the reference symbol description 902 for the reference symbol "16" in drawing 801 is "bolt," an SVG image 903 can be generated in which "bolt" 902 is written / added / displayed / assigned at the same position as the reference symbol "16," and as a result of the SVG image 903 being superimposed and output on drawing 801, the reference symbol "16" can be covered by "bolt" 902. As a result, from the user's perspective, the reference symbol "16" appears to be replaced by "bolt" 902.

[0088] That is, in this embodiment, the web server generates an SVG image 903 in which a description 902 of the drawing code is written / added / assigned / displayed at a position 901 corresponding to the drawing code, and can output the generated SVG image 903 by matching / linking / corresponding to the drawing 801. As a result, the drawing code of the drawing 801 is replaced with the description 902 of the drawing code and output.

[0089] Due to the characteristics of SVG images, they can be freely changed without any loss of quality. Since the position coordinates of the reference symbol captions within an SVG image are assigned as objects / components of the image, the position coordinates of the reference symbol captions are automatically scaled according to changes in the state of the SVG image, eliminating the need to recalculate the position coordinates. In a drawing, the reference symbol is included as an image object constituting the drawing image, and in an SVG image, the reference symbol caption is also included as an image object constituting the SVG image. The reference symbol caption is displayed (or written, added, or assigned) at a position corresponding to the reference symbol. Therefore, the position coordinates of the reference symbol and the reference symbol caption, which are components of the image, are automatically scaled according to changes in the state of the two linked images, and any shifts made as a result of scaling remain the same. Therefore, even if the state of a patent drawing is changed (e.g., moved, rotated, enlarged, or reduced), the SVG image fixed on the patent drawing is also changed in the same way. As a result, the relative positions of the reference symbol captions written / added / assigned / displayed on an SVG image remain fixed, even if the SVG image is moved, rotated, enlarged, or reduced.

[0090] When using an SVG image, the location of the drawing code can be displayed in a specific format (e.g., highlighted in yellow) to inform the user of the drawing code that the web server has currently recognized. Later, at the location displayed in this specific format, an explanation of the drawing code can be displayed like a tool tip using HTML.

[0091] According to this embodiment, the speed of replacing the drawing code description is much faster than the conventional method of assigning the drawing code description by tracking / recalculating the position of the drawing code every time according to the change in the state of the drawing. In particular, in the conventional method, the speed of replacement is very slow because the position of the drawing code must be tracked / recalculated every time according to the change in the state of the drawing, and therefore the description of the drawing code can only be replaced / output for one drawing code at a time. However, according to this embodiment, the speed of replacement is so fast that it is possible to replace / output all drawing codes (i.e., multiple drawing codes) included in one drawing at once.

[0092] Meanwhile, although not shown in the drawings, as a result of assigning relative position coordinates to all the drawing symbol descriptions, there may be cases where drawing symbol descriptions are at least partially overlapped on the SVG image. In this case, the web server can arbitrarily adjust the relative position coordinates of at least one of the overlapping drawing symbol descriptions in a direction so that they do not overlap, and then write / display / add / assign it on the SVG image. For example, if a first drawing symbol description and a second drawing symbol description overlap, the web server can move the first drawing symbol description in a first direction and the second drawing symbol description in a second direction opposite to the first direction by a predetermined length.

[0093] In this specification, an "SVG image" has been described as a representative example of an image used to output the description of drawing symbols, but this is not limited thereto, and images of various formats may be used. Therefore, in this specification, an SVG image may be commonly referred to / replaced with "image," and in this case, "image" may refer to images of various formats that do not lose quality even when the state changes, just like an SVG image.

[0094] Meanwhile, although not specifically mentioned in the flowcharts and drawings, if a drawing already includes a description of each drawing symbol, such as a block diagram or flowchart, the web server may apply a highlighting format to the description of the drawing symbol instead of substituting the description of the drawing symbol for the drawing symbol. To this end, the web server may perform an operation in advance to recognize characters included in the drawing.

[0095] We have discussed in detail above the methods of recognizing drawing codes, extracting drawing code descriptions, and substituting drawing codes for drawing code descriptions. Using the extracted / recognized drawing codes and / or drawing code descriptions as a medium, each category of patent documents can be linked / synchronized with each other. As a result, users can selectively search / retrieve desired information using the drawing codes and / or drawing code descriptions, enabling more efficient understanding of inventions. Examples of the linking / synchronization of each category of patent documents will be described below with reference to FIGS. 11 to 15. Prior to this discussion, we will briefly discuss patent document categories as defined herein.

[0096] FIG. 10 is a diagram illustrating a patent document related to an embodiment of the present invention.

[0097] 10, patent document 1000 can be divided into multiple categories 1001 and 1002. Patent document 1000 can be broadly divided into patent specification 1001 and patent drawings 1002, and patent specification 1001 can be further divided into claims 1001-1, detailed description of the invention 1001-2, and / or explanations of symbols (not shown).

[0098] The multiple categories 1001, 1002 thus divided can be divided into multiple areas / windows and output. For example, the patent specification 1001 and the patent drawings 1002 can be divided into different first and second areas within a single window and output. Alternatively, the patent specification 1001 and the patent drawings 1002 can be divided into different first and second windows and output. Alternatively, the patent specification 1001 and the patent drawings 1002 can be divided into areas within a single window and output, and the patent specification 1001 or the patent drawings 1002 can be additionally output as a separate window.

[0099] The reason for dividing and outputting the information by region / window is to allow all categories 1001 and 1002 to appear in the user's field of vision at once, thereby providing convenience for the user to grasp the information more easily and efficiently.

[0100] In this specification, a plurality of categories 1001, 1002 may be linked to each other through the use of drawing symbols and / or the descriptions of the drawing symbols.

[0101] More specifically, when a selection input for a drawing code included in any one of the plurality of categories 1001 and 1002 is received, the web server can search for the drawing code from the plurality / all of the categories 1001 and 1002. Furthermore, the web server can apply a highlighting format to all of the drawing codes searched from the plurality / all of the categories 1001 and 1002 and output them.

[0102] If drawing codes are searched from the patent specification category 1001, the web server may highlight all the searched drawing codes by applying a predetermined format (e.g., underlining, bolding, applying a different text color, highlighting, etc.) In the same vein, if drawing codes are searched from the patent drawing category 1002, the web server may highlight all the searched drawing codes or all the corresponding alternatively outputted drawing code descriptions by applying a predetermined format (e.g., underlining, bolding, applying a different text color, highlighting, etc.).

[0103] For example, if a user selects drawing number "16" in patent specification category 1001, drawings containing drawing number "16" are automatically selected / output from patent drawing category 1002, and then drawing number "16" is highlighted; or, if a user selects drawing number "16" in patent drawing category 1002, all drawing numbers "16" are highlighted in patent specification category 1001, and a scrolling / output operation can be automatically performed on the sentence / paragraph containing drawing number "16."

[0104] In order to receive user input regarding drawing codes, drawing codes can be output in hyperlink format in the patent specification category 1001, and drawing codes (or descriptions of drawing codes) can also be output in selectable GUI / icon format in the patent drawing category 1002.

[0105] 11 and 12 are diagrams illustrating a drawing interface in which drawing symbols and their descriptions are linked to each other according to an embodiment of the present invention.

[0106] The drawing interface proposed in this specification can be output in a drawing category as a user interface that provides various drawing-related functions to the user. Such a drawing interface can basically output a drawing, as shown in the drawing. In addition, the drawing interface can provide various functions related to the drawings, such as a preview and shortcut function for all drawings included in the patent document, a function for changing the state of the drawing (e.g., rotating, moving, zooming in and out of the drawing), and a function for substituting the description of the drawing symbols, for user convenience. Here, the function for substituting the description of the drawing symbols corresponds to the function to which the above-mentioned embodiment is applied.

[0107] In particular, the drawing interface may provide a function of recognizing all drawing symbols of a selected drawing, extracting the descriptions of all corresponding drawing symbols, and listing all extracted drawing symbol descriptions to provide to a user. At this time, corresponding drawing symbols and their descriptions may be output after being matched / paired with each other. The above-described embodiments may be applied to the recognition of drawing symbols and the extraction of drawing symbol descriptions.

[0108] Furthermore, as shown in Fig. 12, when the web server receives a user selection input for at least one drawing symbol description (or drawing symbol description) from the drawing symbol description list output via the drawing interface, it can selectively output only the selected drawing symbol description (or drawing symbol). For example, as shown in Fig. 12, when a user selection input for bolt 16 is received, the web server can output only the drawing symbol description for bolt 16.

[0109] According to this embodiment, the user can selectively view only the desired drawing symbols or the description of the drawing symbols, and can easily and efficiently understand the drawings / inventions. FIG. 13 illustrates an example of a keyword setting interface according to one embodiment of the present invention.

[0110] The drawing interface proposed in this specification may provide a keyword setting interface, which corresponds to a user interface provided via a drawing category to allow at least a portion of the description of a drawing symbol to be set as a keyword.

[0111] As illustrated in the drawings, the keyword setting interface may include an input window for inputting the description of the drawing symbol to be set as a keyword, a color setting window for setting the highlight color of the set keyword, and / or a keyword indicator for indicating the keywords that have been set up to now. However, the keyword setting interface is not limited thereto, and various functions may be added or at least some of the functions described above may be excluded.

[0112] Through these keyword setting interfaces, users can register / set specific drawing code descriptions among the drawing code descriptions as keywords, and the web server can highlight the registered / set keywords in the drawings or highlight drawings containing the keywords so that users can easily find them.

[0113] For example, if a user wishes to specify "Bolt 16" in the drawing symbol description shown in FIG. 11 as a keyword, the user can input "Bolt" into the input window as shown in FIG. 12, set a highlight color, and then press the register button. The web server can then set / register "Bolt" as a keyword and output a keyword indicator in a predetermined area of ​​the keyword setting interface to indicate that "Bolt" has been set / registered as a keyword. Furthermore, the web server can search for "Bolt" within a drawing category and apply a highlight color to the keyword in the drawing that is currently selected and being enlarged and output. Furthermore, the web server can output an indicator in the same color as the keyword highlight color for drawings in which a keyword has been searched among drawings currently being previewed.

[0114] As a result, the user can easily identify keywords in the drawing he / she is currently viewing, and can also easily identify which drawings contain keywords among drawings he / she is not currently viewing, which has the effect of greatly improving the efficiency with which the user can access desired information. In addition, the user can set / register multiple keywords, and in this case, different colors can be set for each keyword, making it easy to distinguish keywords by color.

[0115] 14 and 15 are diagrams illustrating examples of inter-category linkages using the description of drawing symbols as a medium according to an embodiment of the present invention.

[0116] As shown in Fig. 14, when a user's selection input is received for the drawing code description 1402 that is output by covering the drawing code, a selection window for selecting at least one function may be output. In this case, the provided function 1403 may be a function for searching / retrieving the selected drawing code description 1402 within the patent specification. When the user selects this function 1403, the web server may search for the selected drawing code description 1402 or a drawing code corresponding to this 1402 in all categories, and may apply highlighting 1502 to the retrieved drawing code description 1402 and output it as shown in Fig. 15.

[0117] Referring to FIG. 15, when the web server searches for the description of a selected drawing symbol or its corresponding drawing symbol from a patent specification category, it can highlight all the searched drawing symbol descriptions or drawing symbols 1506-1, 1506-2 by applying a predetermined format (e.g., underlining, bolding, applying a different text color, highlighting, etc.).

[0118] And / or, the web server may automatically scroll the web page to the position / page / sentence / paragraph of the description or drawing code 1506-1 of the searched drawing code description or drawing code located at the top of the patent specification category (particularly the claims category and detailed description category).

[0119] Furthermore, the web server may output a first indicator 1505 in the scroll bar area (or an area adjacent to the scroll bar) corresponding to the location of all the searched drawing symbol descriptions or drawing symbols in the overall scroll bar area provided in the patent specification category (particularly, the claims category and the detailed description category). That is, the first indicator 1505 displays the page area where the drawing symbol descriptions or drawing symbols are located in the scroll bar area in a minimap format, allowing the user to move the scroll bar to the position of the first indicator 1505 and more easily access the desired information. Furthermore, the first indicator output in a minimap format allows the user to see at a glance the distribution of all the searched drawing symbol descriptions and / or drawing symbols to which categories and to what extent, which can be useful for determining the importance of elements.

[0120] And / or, the web server may provide / output search windows 1504-1, 1504-2 for searching the description or drawing code of the searched drawing code in the patent specification category (particularly, the claims category and the detailed description category). Thus, the user can immediately search for the paragraph, page, or sentence containing the description or drawing code of the drawing code he / she is searching for through the search windows 1504-1, 1504-2.

[0121] And / or, the web server may search for the selected drawing code description or corresponding drawing code in the patent drawing category, and may provide / output a second indicator 1507 for drawings containing the drawing code description or corresponding drawing code among the drawings currently being previewed. Thus, the user can immediately search / find drawings containing the desired drawing code description by selecting a drawing for which the second indicator 1507 is provided / output.

[0122] Meanwhile, although not shown in the drawings, the web server can automatically select and enlarge and output drawings corresponding to the paragraph / sentence portion currently being read / searched by the user from within the patent specification category. To this end, the web server can grasp the contents of the patent specification category (particularly the "detailed description of the invention" category) and divide areas into sections for each description of each drawing (e.g., paragraph, sentence, page), and automatically select and enlarge and output drawings corresponding to the area being output with the highest ratio on the current web page. For example, if the "detailed description of the invention" for Figure 1 is being output on the user's device, the web server can automatically select Figure 1 and enlarge and output it within the "drawings" category.

[0123] Similarly, if a user selects a specific area (e.g., paragraph, sentence, page) within the patent specification category, the web server can automatically select and enlarge the drawing corresponding to the selected drawing from the drawing category. Furthermore, the web server can automatically replace the drawing symbols included in the selected specific area with the description of the drawing symbols.

[0124] The web server can database (i.e., accumulate as data and store separately in a database) at least one piece of information acquired / recognized by the above-mentioned operation / method / embodiment, and construct a patent drawing search database.

[0125] For example, the web server can group patent documents, patent drawings, the size of the patent drawings, the drawing symbols contained in the patent drawings, the description of the drawing symbols corresponding to each drawing symbol, and / or the relative position coordinates of the drawing symbols into one data set and store this in a database, thereby constructing a patent drawing search database.

[0126] Therefore, users can easily obtain information about the patent drawings they are looking for from patent drawings around the world by entering search terms related to patent documents through the patent drawing search database constructed in this way.

[0127] Furthermore, the web server can update the various models / rules mentioned above by learning various patent drawing data in real time / periodically through the patent drawing search database constructed in this way, thereby improving recognition accuracy / speed.

[0128] FIG. 16 is a diagram illustrating a patent information search system according to an embodiment of the present invention.

[0129] Referring to FIG. 16, the patent information search system proposed in this specification may include a web server and a user device.

[0130] The web server 1601 and the user device 1602 are interconnected mainly via an internet connection to perform communication, and can provide / receive patent information search services via web services / pages.

[0131] The web server 1601 may correspond to a server / device including at least one software and hardware component designed to perform the embodiments proposed herein. In particular, the web server 1601 may provide the patent information search service proposed herein to a user device, which is a client device, via an internet web page.

[0132] The user device 1602 may correspond to a client device that receives a patent information search service provided through a web server. The user device 1602 may receive the patent information search service provided by the web server through an internet web page.

[0133] In this specification, the patent information search system is illustrated as being composed of a web server 1601 and a user device 1602, but is not limited thereto, and the web server 1601 may be substituted for an application server, a program server, or a service providing device depending on the embodiment. Also, in this specification, the subject performing the embodiments is described as the web server 1601, but is not limited thereto, and the web server 1601 may be substituted for a program or application designed to implement the above-described embodiments, and the operation of the web server 1601 may be interpreted as these functions.

[0134] FIG. 17 is a block diagram of a web server according to one embodiment of the present invention.

[0135] Referring to FIG. 17, the web server may include a processor 1710 , a memory unit 1720 and a communication unit 1730 .

[0136] The processor 1710 may communicate with or control other components to perform the embodiments described herein, execute various programs and / or applications stored in the memory unit 1720, and process internal data. The processor 1710 may include at least one central processing unit (CPU), microprocessor unit (MPU), microcontroller unit (MCU), application processor (AP), or any other type of processor known in the art. Therefore, in this specification, a web server may be referred to as a processor.

[0137] The memory unit 1720 refers not only to a built-in digital data storage space such as a flash memory, hard disk drive (HDD), or solid state drive (SSD), but also to an external storage space capable of storing data via a communication connection, such as the cloud. Thus, the memory unit 1720 can store various digital data such as video, audio, photographs, animations, images, text, applications, and programs. In particular, the memory unit 1720 proposed herein can store various knowledge information content (particularly, patent documents) data and can store a patent document search database and / or a patent drawing search database 1720-1 constructed by the processor 1710. Thus, according to an embodiment, the processor 1710 can load various data from the patent document and / or patent drawing search database stored in the memory unit 1720 and perform data processing / output operations.

[0138] The communication unit 1730 can perform communication and send / receive data using at least one wired / wireless communication protocol.

[0139] Embodiments of the present invention may be implemented by various means, such as hardware, firmware, software, or a combination thereof. In the case of a hardware implementation, an embodiment of the present invention may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, etc.

[0140] Furthermore, in the case of a firmware or software implementation, an embodiment of the present invention may be embodied in the form of modules, procedures, functions, etc. that perform the functions or operations described above, and may be recorded on various computer-readable recording media. Here, the recording media may include program instructions, data files, data structures, etc., alone or in combination. The program instructions recorded on the recording media may be specially designed and constructed for the present invention, or may be known and available to those skilled in the art of computer software. For example, recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as compact disk read-only memories (CD-ROMs) and digital video disks (DVDs); magneto-optical media such as floptical disks; and hardware devices specially configured to store and execute program instructions, such as read-only memory (ROM), random access memory (RAM), and flash memory. Examples of program instructions include not only machine language code, such as that produced by a compiler, but also high-level language code that can be executed by a computer using an interpreter, etc. Such hardware devices may be configured to operate as one or more software programs to perform the operations of the present invention, and vice versa.

[0141] Furthermore, the devices and terminals of the present invention may be driven by instructions that cause one or more processors to perform the functions and processes described above. For example, such instructions may include interpreted instructions, such as script instructions, e.g., JavaScript or ECMAScript instructions, executable code, or other instructions stored on a computer-readable medium. Furthermore, the devices and terminals of the present invention may be implemented in a distributed manner across a network, such as a server farm, or may be implemented in a single computing device.

[0142] Additionally, computer programs (also known as programs, software, software applications, scripts, or code) that are included in the devices of the present invention and that perform the methods of the present invention can be written in any form of programming language, including compiled or interpreted, a priori, or procedural languages, and can be deployed in any form, including stand-alone programs, modules, components, subroutines, or other units suitable for use in a computing environment. Computer programs do not necessarily correspond to files in a file system. A program can be stored in a single file provided by a requested program, in multiple interacting files (e.g., a file storing one or more modules, subprograms, or portions of code), or as part of a file containing other programs or data (e.g., one or more scripts stored in a markup language document). A computer program can be deployed to run on a single computer, or on multiple computers located at one site or distributed across multiple sites interconnected by a communications network.

[0143] Although the drawings have been described separately for ease of explanation, it is possible to combine the embodiments shown in the drawings to design a new embodiment. Furthermore, the present invention is not limited to the configurations and methods of the above-described embodiments, and the above-described embodiments may be configured by selectively combining all or part of each embodiment to achieve various modifications.

[0144] Furthermore, although preferred embodiments have been illustrated and described above, the present specification is not limited to the specific embodiments described above, and various modifications are possible for those skilled in the art to which the present invention pertains without departing from the gist of the claims, and these modifications should not be understood individually from the technical ideas and perspectives of the present specification. [Industrial Applicability]

[0145] The present invention can be utilized in various patent search systems / devices / methods. [Explanation of symbols]

[0146] 10 Organic Light-Emitting Display 301 Patent Drawing Samples 302 Image fragments 801 Patent Drawings 802 SVG images 901 Location coordinates 902 Explanation 903 Write / Append / Display / Assign 1000 Patent Documents 1001 Patent Specification 1001-1 Claims 1001-2 Detailed Description of the Invention 1002 Patent Drawings 1402 Drawing symbol explanation 1403 Features 1502 Highlight 1504-1, 1504-2 Search window 1505 First Indicator 1506-1, 1506-2 Drawing code explanation or drawing code 1601 Web Server 1602 User equipment 1710 processor 1720 Memory Unit 1720-1 Patent Document Search Database and / or Patent Drawing Search Database 1730 Communication Unit

Claims

1. In a method for outputting a drawing symbol description performed by a server, the server recognizing the size of the patent drawings; the server obtaining relative position coordinates of the drawing symbols in the patent drawings; the server setting relative position coordinates of the drawing code legend corresponding to the drawing code; the server overlapping the patent drawing with an image of the same size and transparency; the server overlapping the image with the patent drawing and fixing the transparent image to the patent drawing; and the server outputs the reference numeral legend onto the transparent image at a relative position coordinate of the reference numeral legend.

2. The above drawing symbol explanations are as follows:

2. The method for outputting an explanation of a drawing code according to claim 1, wherein the position of the drawing code included in the patent drawing is recognized, and the drawing code at the recognized position is cut out from the patent drawing to obtain an image engraving including the drawing code, which corresponds to the recognized drawing code in the image engraving.

3. 2. The drawing symbol explanation output method according to claim 1, wherein the drawing symbol explanation specifies relative position coordinates within the patent drawing, so that the relative position within the patent drawing is fixed even if the patent drawing is moved, rotated, enlarged, or reduced.

4. the server recognizing the drawing reference numbers included in the patent drawings; The server extracts drawing code legends corresponding to all of the drawing codes; and 2. The method of claim 1, further comprising the step of: the server listing and outputting the extracted drawing symbol descriptions.

5. providing a keyword setting interface for setting the drawing reference numeral legends as keywords by the server; and displaying an indicator corresponding to the keyword on a drawing in which the keyword is included in a drawing code description among the drawings provided in the preview. The method of claim 1 , further comprising:

6. the server displays the detailed description of the invention in an area separated from the patent drawings; In the detailed description of the invention, a step of displaying an indicator in a scroll bar area in a page area where a drawing reference number included in the patent drawing is located; The method of claim 1 , further comprising:

7. providing a function for searching the drawing symbol description within the patent specification; When a user selects a function, searching for a drawing code description in the detailed description of the invention of the patent specification and highlighting the drawing code description; The method of claim 1 , further comprising:

8. A server that executes a method for outputting a description of a drawing symbol, the server includes a processor; The processor is Recognize the size of the patent drawings, Obtaining the relative position coordinates of the drawing symbols in the patent drawings; Set the relative position coordinates of the drawing symbol legend corresponding to the drawing symbol, The server overlays a transparent image on the patent drawing, the image being the same size as the patent drawing; the server overlaps the image with the patent drawing and fixes the transparent image to the patent drawing; The server outputs the reference numeral legend to the transparent image at the relative position coordinates of the reference numeral legend. A server characterized by:

Images