# US 2014/0208192 A1 · Footnote Detection in a Fixed Format Document > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2014/0208192 A1 - Document ID: US20140208192A1 - Status: Published application - Kind: A1 - Application number: 13/745,127 - Filed: 2013-01-18 - Published: 2014-07-24 - Earliest claimed: 2013-01-18 - Assignees: Microsoft Corporation - Applicants: Microsoft Corporation - Inventors: Drazen Zaric; Milos Raskovic - Web page: https://patentgrep.com/patent/US20140208192A1 ## CPC - G06F 40/134: Hyperlinking - G06F 16/382: {using citations (hypermedia G06F16/94)} - G06F 40/103: Formatting, i.e. changing of presentation of documents (automatic justification G06F40/189; automatic line break hyphenation G06F40/191) - G06F 40/169: Annotation, e.g. comment data or footnotes - G06F 40/30: Semantic analysis - G06F 40/106: Display of layout of documents; Previewing - G06V 30/414: Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Footnote detection is provided. When converting a fixed format document to a flow format document, footnotes, which may include a footnote description and a footnote reference mark, may be detected and linked with a corresponding footnote reference mark in a main document flow on a page. Additionally, a footnote numbering scheme may be determined and applied to detected footnotes. By providing footnote detection when converting a fixed format document to a flow format document, footnotes may be statically positioned at the end of a page containing the corresponding footnote reference mark. ## Claims ### Claim 1 A method for detecting one or more footnotes in a document, the method comprising: detecting a footnote zone on a page in the document; detecting one or more paragraphs in the footnote zone; determining which of the one or more paragraphs belong to a footnote description of a footnote; detecting a corresponding footnote reference mark in a main document flow on the page for each footnote; and linking each footnote with the corresponding footnote reference mark in the main document flow on the page. ### Claim 2 (depends on 1) The method of claim 1, wherein detecting a footnote zone on a page comprises detecting an area at the bottom of a region and including the area at the bottom of the region to the footnote zone. ### Claim 3 (depends on 1) The method of claim 1, wherein detecting a footnote zone on a page comprises detecting a footnote separator line and including the area below the footnote separator line to the footnote zone. ### Claim 4 (depends on 3) The method of claim 3, wherein detecting a footnote separator line comprises detecting a line node and determining: the line node has one child element; the child element is a graphic aggregation object; and the height of the line node is less than the average text height of text on the page. ### Claim 5 (depends on 1) The method of claim 1, wherein determining which of the one or more paragraphs belong to a footnote description of a footnote comprises: analyzing each of the one or more paragraphs starting with last paragraph in the footnote zone; if a footnote separator line is not found and if the paragraph starts with a superscript number, determining the paragraph is a first paragraph of a footnote description of a footnote; and if a footnote separator line is found and if the paragraph starts with a number or any text written in superscript, determining the paragraph is a first paragraph of a footnote description of a footnote. ### Claim 6 (depends on 5) The method of claim 5, further comprising: determining a number of paragraphs in a footnote description; and if a footnote description comprises more than three paragraphs, discarding the footnotes in the footnote zone. ### Claim 7 (depends on 1) The method of claim 1, wherein detecting a corresponding footnote reference mark in a main document flow on the page for each footnote comprises: extracting a footnote reference mark from each footnote description; and analyzing the main document flow on the page for a superscript text run comprising the extracted footnote reference mark. ### Claim 8 (depends on 7) The method of claim 7, further comprising: for each footnote description, if the extracted footnote reference mark is not found in a superscript text run, analyzing the main document flow on the page for a non-superscript text run comprising the extracted footnote reference mark. ### Claim 9 (depends on 7) The method of claim 7, further comprising: if a corresponding footnote reference mark is found for less than half of the footnotes on a page, discarding the footnotes. ### Claim 10 (depends on 1) The method of claim 1, further comprising: detecting a footnote numbering scheme; and applying the footnote numbering scheme to each footnote. ### Claim 11 (depends on 10) The method of claim 10, where detecting a footnote numbering scheme comprises analyzing each footnote reference mark on all pages of the document to determine if the footnote reference mark is a valid element in a numbering scheme. ### Claim 12 (depends on 11) The method of claim 11, further comprising determining a numbering scheme that covers a longest subsequence of footnote reference marks. ### Claim 13 (depends on 11) The method of claim 11, further comprising: if a footnote reference mark does not fit into a determined numbering scheme, determining the footnote reference mark to be a custom footnote reference mark to be reconstructed literally. ### Claim 14 (depends on 1) The method of claim 1, further comprising statically positioning each of the one or more footnotes at the end of the page containing the corresponding footnote reference mark. ### Claim 15 A system for detecting one or more footnotes in a document, the system comprising: one or more processors; and a memory coupled to the one or more processors, the one or more processors operable to: detect a footnote zone on a page in the document; detect one or more paragraphs in the footnote zone; determine which of the one or more paragraphs belong to a footnote description of a footnote; detect a corresponding footnote reference mark in a main document flow on the page for each footnote; and link each footnote with the corresponding footnote reference mark in the main document flow on the page. ### Claim 16 (depends on 15) The system of claim 15, wherein the one or more processors are further operable to: detect a footnote numbering scheme; and apply the footnote numbering scheme to each footnote. ### Claim 17 (depends on 15) The system of claim 15, wherein the footnote zone comprises an area at the bottom of a region and an area below a footnote separator line. ### Claim 18 (depends on 15) The system of claim 15, wherein the one or more processors are further operable to: analyze each of the one or more paragraphs starting with last paragraph in the footnote zone; if a footnote separator line is not found and if the paragraph starts with a superscript number, determine the paragraph is a first paragraph of a footnote description of a footnote; and if a footnote separator line is found and if the paragraph starts with a number or any text written in superscript, determine the paragraph is a first paragraph of a footnote description of a footnote. ### Claim 19 (depends on 15) The system of claim 15, wherein the one or more processors are further operable to: extract a footnote reference mark from each footnote description; analyze the main document flow on the page for a superscript text run comprising the extracted footnote reference mark; and if the extracted footnote reference mark is not found in a superscript text run, analyze the main document flow on the page for a non-superscript text run comprising the extracted footnote reference mark. ### Claim 20 A computer readable medium containing computer executable instructions which, when executed by a computer, perform a method for detecting one or more footnotes in a document, the method comprising: detecting a footnote zone on a page in the document; detecting one or more paragraphs in the footnote zone; determining which of the one or more paragraphs in the footnote zone belong to a footnote description of a footnote; extracting a footnote reference mark from each footnote description; and searching the main document flow on the page for a match to the extracted footnote reference mark; and linking each footnote with the matching footnote reference mark in the main document flow on the page.