# US 2023/0053656 A1 · Machine-Learning-Based Identification of Drawing Attributes > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0053656 A1 - Document ID: US20230053656A1 - Status: Published application - Kind: A1 - Application number: 17/408,052 - Filed: 2021-08-20 - Published: 2023-02-23 - Earliest claimed: 2021-08-20 - Assignees: Procore Technologies, Inc. - Applicants: Procore Technologies, Inc. - Inventors: Patavee Meemeng; Gagik Hacobian; Hunter Buckhorn - Web page: https://patentgrep.com/patent/US20230053656A1 ## CPC - G06V 30/422: Technical drawings; Geographical maps - G06V 10/764: using classification, e.g. of video objects - G06V 30/10: Character recognition - G06V 30/414: Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text - G06V 30/416: Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors - G06V 30/19007: {Matching; Proximity measures} - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract An example computing system is configured to: (i) access a drawing associated with a construction project; (ii) identify, in the drawing, a set of candidate textual elements that potentially represent a title of the drawing; (iii) for each candidate textual element, (a) determine a respective dataset comprising values for a set of data variables that are potentially predictive of whether the candidate textual element is the title of the drawing, and (b) input the respective dataset into a machine-learning model that functions to (1) evaluate the respective dataset and (2) output, based on the evaluation, a respective score indicating a likelihood that the candidate textual element represents the title of the drawing; and (iv) based on the respective scores for the candidate textual elements that are output by the machine-learning model, select one given candidate textual element as the title of the drawing. ## Claims ### Claim 1 A computing system comprising: at least one processor; a non-transitory computer-readable medium; and program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor such that the computing system is configured to: access a drawing associated with a construction project; identify, in the drawing, a set of candidate textual elements that potentially represent a title of the drawing; for each candidate textual element in the set of candidate textual elements, (i) determine a respective dataset comprising values for a set of data variables that are potentially predictive of whether the candidate textual element is the title of the drawing, and (ii) input the respective dataset for the candidate textual element into a machine-learning model that functions to (a) evaluate the respective dataset and (b) based on the evaluation, output a respective score indicating a likelihood that the candidate textual element represents the title of the drawing; based on the respective scores for the candidate textual elements in the set of candidate textual elements that are output by the machine-learning model, select one given candidate textual element in the set of candidate textual elements as the title of the drawing; and populate a data attribute field representing the title of the drawing with the given candidate textual element. ### Claim 2 (depends on 1) The computing system of claim 1, wherein identifying the set of candidate textual elements that potentially represent the title of the drawing comprises: identifying one or more rectangular boxes in the drawing; identifying a first subset of candidate textual elements in the one or more rectangular boxes; and including the first subset of candidate textual elements in the set of candidate textual elements. ### Claim 3 (depends on 2) The computing system of claim 2, wherein identifying the set of candidate textual elements that potentially represent the title of the drawing further comprises: applying a mask to the one or more rectangular boxes in the drawing; identifying a second subset of candidate textual elements in one or more targeted portions of the drawing, wherein the one or more targeted portions of the drawing at least partially overlaps with the one or more rectangular boxes; and including the second subset of candidate textual elements in the set of candidate textual elements. ### Claim 4 (depends on 3) The computing system of claim 3, wherein the one or more targeted portions of the drawing includes at least one of (i) a first portion extending upward from a bottom edge of the drawing by no more than 30% of the drawing height, or (ii) a second portion extending leftward from a right edge of the drawing by no more than 30% of the drawing width. ### Claim 5 (depends on 1) The computing system of claim 1, wherein determining the respective dataset comprising values for the set of data variables that are potentially predictive of whether the candidate textual element is the title of the drawing comprises: determining (i) one or more spatial characteristics of the candidate textual element, (ii) one or more linguistic characteristics of the candidate textual element, and (iii) one or more surrounding characteristics of the candidate textual element; and determining the respective dataset based on (i) the one or more spatial characteristics of the candidate textual element, (ii) the one or more linguistic characteristics of the candidate textual element, and (iii) the one or more surrounding characteristics of the candidate textual element. ### Claim 6 (depends on 5) The computing system of claim 5, wherein the one or more spatial characteristics of the candidate textual element comprises at least one of (i) a distance between the candidate textual element and an edge of the drawing, (ii) a location of the textual element within the drawing, (iii) an orientation of the candidate textual element, or (iv) a size of the candidate textual element. ### Claim 7 (depends on 5) The computing system of claim 5, wherein the one or more linguistic characteristics of the candidate textual element comprises at least one of (i) one or more words in the candidate textual element, (ii) one or more parts of speech of the one or more words in the candidate textual element, (iii) information identifying whether the candidate textual element includes a word in a first predefined list of words, (iv) information identifying whether the textual element includes a capitalized character, or (v) information identifying whether the textual element includes a numerical character. ### Claim 8 (depends on 5) The computing system of claim 5, wherein the one or more surrounding characteristics of the candidate textual element comprises at least one of (i) information identifying whether the candidate textual element is located within a rectangular box, (ii) information identifying whether one or more other textual elements are within a threshold distance of the candidate textual element, or (iii) information identifying whether the one or more other textual elements within the threshold distance of the candidate textual element include a word in a second predefined list of words. ### Claim 9 A non-transitory computer-readable medium, wherein the non-transitory computer-readable medium is provisioned with program instructions that, when executed by at least one processor, cause a computing system to: access a drawing associated with a construction project; identify, in the drawing, a set of candidate textual elements that potentially represent a title of the drawing; for each candidate textual element in the set of candidate textual elements, (i) determine a respective dataset comprising values for a set of data variables that are potentially predictive of whether the candidate textual element is the title of the drawing, and (ii) input the respective dataset for the candidate textual element into a machine-learning model that functions to (a) evaluate the respective dataset and (b) based on the evaluation, output a respective score indicating a likelihood that the candidate textual element represents the title of the drawing; based on the respective scores for the candidate textual elements in the set of candidate textual elements that are output by the machine-learning model, select one given candidate textual element in the set of candidate textual elements as the title of the drawing; and populate a data attribute field representing the title of the drawing with the given candidate textual element. ### Claim 10 (depends on 9) The non-transitory computer-readable medium of claim 9, wherein identifying the set of candidate textual elements that potentially represent the title of the drawing comprises: identifying one or more rectangular boxes in the drawing; identifying a first subset of candidate textual elements in the one or more rectangular boxes; and including the first subset of candidate textual elements in the set of candidate textual elements. ### Claim 11 (depends on 10) The non-transitory computer-readable medium of claim 10, wherein identifying the set of candidate textual elements that potentially represent the title of the drawing further comprises: applying a mask to the one or more rectangular boxes in the drawing; identifying a second subset of candidate textual elements in one or more targeted portions of the drawing, wherein the one or more targeted portions of the drawing at least partially overlaps with the one or more rectangular boxes; and including the second subset of candidate textual elements in the set of candidate textual elements. ### Claim 12 (depends on 11) The non-transitory computer-readable medium of claim 11, wherein the one or more targeted portions of the drawing includes at least one of (i) a first portion extending upward from a bottom edge of the drawing by no more than 30% of the drawing height, or (ii) a second portion extending leftward from a right edge of the drawing by no more than 30% of the drawing width. ### Claim 13 (depends on 9) The non-transitory computer-readable medium of claim 9, wherein determining the respective dataset comprising values for the set of data variables that are potentially predictive of whether the candidate textual element is the title of the drawing comprises: determining (i) one or more spatial characteristics of the candidate textual element, (ii) one or more linguistic characteristics of the candidate textual element, and (iii) one or more surrounding characteristics of the candidate textual element; and determining the respective dataset based on (i) the one or more spatial characteristics of the candidate textual element, (ii) the one or more linguistic characteristics of the candidate textual element, and (iii) the one or more surrounding characteristics of the candidate textual element. ### Claim 14 (depends on 13) The non-transitory computer-readable medium of claim 13, wherein: the one or more spatial characteristics of the candidate textual element comprises at least one of (i) a distance between the candidate textual element and an edge of the drawing, (ii) a location of the textual element within the drawing, (iii) an orientation of the candidate textual element, or (iv) a size of the candidate textual element; the one or more linguistic characteristics of the candidate textual element comprises at least one of (i) one or more words in the candidate textual element, (ii) one or more parts of speech of the one or more words in the candidate textual element, (iii) information identifying whether the candidate textual element includes a word in a first predefined list of words, (iv) information identifying whether the textual element includes a capitalized character, or (v) information identifying whether the textual element includes a numerical character; and the one or more surrounding characteristics of the candidate textual element comprises at least one of (i) information identifying whether the candidate textual element is located within a rectangular box, (ii) information identifying whether one or more other textual elements are within a threshold distance of the candidate textual element, or (iii) information identifying whether the one or more other textual elements within the threshold distance of the candidate textual element include a word in a second predefined list of words. ### Claim 15 A method comprising: accessing a drawing associated with a construction project; identifying, in the drawing, a set of candidate textual elements that potentially represent a title of the drawing; for each candidate textual element in the set of candidate textual elements, (i) determining a respective dataset comprising values for a set of data variables that are potentially predictive of whether the candidate textual element is the title of the drawing, and (ii) inputting the respective dataset for the candidate textual element into a machine-learning model that functions to (a) evaluate the respective dataset and (b) based on the evaluation, output a respective score indicating a likelihood that the candidate textual element represents the title of the drawing; based on the respective scores for the candidate textual elements in the set of candidate textual elements that are output by the machine-learning model, selecting one given candidate textual element in the set of candidate textual elements as the title of the drawing; and populating a data attribute field representing the title of the drawing with the given candidate textual element. ### Claim 16 (depends on 15) The method of claim 15, wherein identifying the set of candidate textual elements that potentially represent the title of the drawing comprises: identifying one or more rectangular boxes in the drawing; identifying a first subset of candidate textual elements in the one or more rectangular boxes; and including the first subset of candidate textual elements in the set of candidate textual elements. ### Claim 17 (depends on 16) The method of claim 16, wherein identifying the set of candidate textual elements that potentially represent the title of the drawing further comprises: applying a mask to the one or more rectangular boxes in the drawing; identifying a second subset of candidate textual elements in one or more targeted portions of the drawing, wherein the one or more targeted portions of the drawing at least partially overlaps with the one or more rectangular boxes; and including the second subset of candidate textual elements in the set of candidate textual elements. ### Claim 18 (depends on 17) The method of claim 17, wherein the one or more targeted portions of the drawing includes at least one of (i) a first portion extending upward from a bottom edge of the drawing by no more than 30% of the drawing height, or (ii) a second portion extending leftward from a right edge of the drawing by no more than 30% of the drawing width. ### Claim 19 (depends on 15) The method of claim 15, wherein determining the respective dataset comprising values for the set of data variables that are potentially predictive of whether the candidate textual element is the title of the drawing comprises: determining (i) one or more spatial characteristics of the candidate textual element, (ii) one or more linguistic characteristics of the candidate textual element, and (iii) one or more surrounding characteristics of the candidate textual element; and determining the respective dataset based on (i) the one or more spatial characteristics of the candidate textual element, (ii) the one or more linguistic characteristics of the candidate textual element, and (iii) the one or more surrounding characteristics of the candidate textual element. ### Claim 20 (depends on 19) The method of claim 19, wherein: the one or more spatial characteristics of the candidate textual element comprises at least one of (i) a distance between the candidate textual element and an edge of the drawing, (ii) a location of the textual element within the drawing, (iii) an orientation of the candidate textual element, or (iv) a size of the candidate textual element; the one or more linguistic characteristics of the candidate textual element comprises at least one of (i) one or more words in the candidate textual element, (ii) one or more parts of speech of the one or more words in the candidate textual element, (iii) information identifying whether the candidate textual element includes a word in a first predefined list of words, (iv) information identifying whether the textual element includes a capitalized character, or (v) information identifying whether the textual element includes a numerical character; and the one or more surrounding characteristics of the candidate textual element comprises at least one of (i) information identifying whether the candidate textual element is located within a rectangular box, (ii) information identifying whether one or more other textual elements are within a threshold distance of the candidate textual element, or (iii) information identifying whether the one or more other textual elements within the threshold distance of the candidate textual element include a word in a second predefined list of words.