# US 2021/0279532 A1 · Digital image processing > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0279532 A1 - Document ID: US20210279532A1 - Status: Published application - Kind: A1 - Application number: 16/811,490 - Filed: 2020-03-06 - Published: 2021-09-09 - Earliest claimed: 2020-03-06 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Peter Willem Jan Staar; Michele Dolfi; Christoph Auer; Leonidas Georgopoulos; Konstantinos Bekas - Web page: https://patentgrep.com/patent/US20210279532A1 ## CPC - G06F 40/205: Parsing - G06F 18/2148: {characterised by the process organisation or structure, e.g. boosting cascade} - G06F 18/2413: based on distances to training or reference patterns - G06F 40/30: Semantic analysis - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/08: Learning methods - G06V 10/764: using classification, e.g. of video objects - G06V 10/82: using neural networks - G06V 30/19147: {Obtaining sets of training patterns; Bootstrap methods, e.g. bagging or boosting} - G06V 30/1983: {Syntactic or structural pattern recognition, e.g. symbolic string recognition} - G06V 30/413: Classification of content, e.g. text, photographs or tables - 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) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A computer-implemented method for processing a digital image. The digital image comprises one or more text cells, wherein each of the one or more text cells comprises a string and a bounding box. The method comprises receiving the digital image in a first format, the first format providing access to the strings and the bounding boxes of the one more text cells. The methods further comprises encoding the strings of the one or more text cells as visual pattern according to a predefined string encoding scheme and providing the digital image in a second format. The second format comprises the visual pattern of the strings of the one or more text cells. A corresponding system and a related computer program product is provided. ## Claims ### Claim 1 A method comprising: receiving a digital image in a first format, the first format providing access to a strings and a bounding boxes of a one more text cells; encoding the strings of the one or more text cells as visual patterns according to a predefined string encoding scheme; and providing the digital image in a second format, the second format comprising the visual patterns of the strings of the one or more text cells. ### Claim 2 (depends on 1) The method according to claim 1, wherein the string encoding scheme is configured to encode natural language features of the string. ### Claim 3 (depends on 1) The method according to claim 1, the method further comprising performing a digital image processing application with the digital image of the second format. ### Claim 4 (depends on 1) The method according to claim 1, the method further comprising performing a machine learning application with the digital image of the second format. ### Claim 5 (depends on 4) The method according to claim 4, wherein the machine learning application is a deep neural network application. ### Claim 6 (depends on 1) The method according to claim 1, wherein the digital image of the first format and the digital image of the second format comprises paths including corresponding path information, the path information comprising path coordinates. ### Claim 7 (depends on 1) The method according to claim 1, the method further comprising augmenting the bounding boxes of the one or more text cells with the visual patterns of the corresponding strings. ### Claim 8 (depends on 1) The method according to claim 1, the method further comprising superimposing the visual patterns on the bounding boxes of the one or more text cells. ### Claim 9 (depends on 1) The method according to claim 1, wherein the encoding of the strings comprises generating a word embedding of the string; and generating the visual pattern from the word embedding of the string. ### Claim 10 (depends on 1) The method according to claim 1, wherein the predefined string encoding scheme is configured to encode the strings in dependence on the percentage of digits in the string. ### Claim 11 (depends on 1) The method according to claim 1, the method comprising applying within the bounding box of the text cell a first visual pattern if a predefined percentage of the characters in the string are digits; and else applying within the bounding box of the text cell a second visual pattern. ### Claim 12 (depends on 11) The method according to claim 11, wherein the first visual pattern comprises or consists of vertical lines and the second visual pattern comprises or consist of horizonal lines; or the first visual pattern comprises or consists of horizonal lines and the second visual pattern comprises or consist of vertical lines. ### Claim 13 (depends on 11) The method according to claim 11, the method comprising coloring the bounding box of the text cell with a first color if a predefined percentage of the characters in the string are digits; and else coloring the bounding box of the text cell with a second color. ### Claim 14 (depends on 1) The method according to claim 1, wherein the digital image of the first format comprises bitmap resources and their bounding boxes. ### Claim 15 (depends on 14) The method according to claim 14, further comprising performing an optical character recognition on the bitmap resources; identifying text cells including their strings and their bounding boxes in the bitmap resources; and encoding the strings of the text cells of the bitmap images as visual patterns according to the predefined string encoding scheme; ### Claim 16 (depends on 1) The method according to claim 1, wherein the first format is a programmatic data format. ### Claim 17 (depends on 1) The method according to claim 1, wherein the first format is a format according to a Portable Document Format-standard. ### Claim 18 (depends on 1) The method according to claim 1, wherein the first format is selected from the group consisting of: Java Script Object Notification Data Interchange Format; Hypertext Markup Language; and YAML Ain't Markup Language. ### Claim 19 (depends on 4) The method according to claim 4, wherein the machine learning application is configured to identify a predefined set of items within the digital image. ### Claim 20 (depends on 4) The method according to claim 4, wherein the machine learning application is configured to perform tasks selected from the group consisting of: classifying the digital image and/or elements of the digital image based on the visual patterns of the text cells; detecting tables in the digital image; and detecting images in the digital image representing a page layout. ### Claim 21 A system comprising one or more processors for executing computer-readable instructions, the computer-readable instructions controlling the one or more processors to perform operations comprising: receiving a digital image in a first format, the first format providing access to strings and bounding boxes of one more text cells; encoding the strings of the one or more text cells as visual patterns according to a predefined string encoding scheme; and providing the digital image in a second format, the second format comprising the visual patterns of the strings of the one or more text cells. ### Claim 22 A computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform operations comprising: receiving a digital image in a first format, the first format providing access to strings and bounding boxes of one more text cells; encoding the strings of the one or more text cells as visual patterns according to a predefined string encoding scheme; and providing the digital image in a second format, the second format comprising the visual patterns of the strings of the one or more text cells. ### Claim 23 (depends on 22) The computer program product of claim 22, further comprising: receiving digital images in a second format, the second format comprising visual patterns which encode strings of one or more text cells according to a predefined string encoding scheme; and training a cognitive model of the machine learning application with the digital images of the second format. ### Claim 24 A cognitive model for a machine learning application, the cognitive model being encoded on a non-transitory computer readable medium, the machine learning application being configured to process digital images, the digital images comprising one or more text cells, wherein each of the one or more text cells comprises a string and a bounding box, the cognitive model being configured to receive the digital images in a second format, the second format comprising visual patterns which encode strings of one or more text cells according to a predefined string encoding scheme. ### Claim 25 A method for processing a digital image, the method comprising: receiving, by a cognitive model, digital images in a second format, the second format comprising visual patterns which encode strings of one or more text cells according to a predefined string encoding scheme.