首页|Researchers Submit Patent Application, 'Systems And Methods For Detecting User Created Circular Shaped Indications Using Machine Learning Models', for Approval (USPTO 20240046676)
Researchers Submit Patent Application, 'Systems And Methods For Detecting User Created Circular Shaped Indications Using Machine Learning Models', for Approval (USPTO 20240046676)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
From Washington, D.C., NewsRx journalists report that a patent application by the inventor Chepel, Maksim (Hartford, CT, US), filed on August 2, 2022, was made available online on February 8, 2024. No assignee for this patent application has been made. News editors obtained the following quote from the background information supplied by the inventors: “Traditionally, text recognition algorithms such as optical character recognition (OCR) algorithms are used to convert images of typed, handwritten, or printed text into machine-encoded text. For instance, a document such as a scanned document, a photo, an image, or any other type of file may include text. The OCR algorithms may be used to convert the document into machine-encoded text that is usable in a plurality of different applications for an enterprise organization. For instance, an enterprise organization may use text recognition algorithms to extract content from numerous different documents fairly quickly. In certain documents, a person may provide user selections such as marking a checkbox or other features to indicate their selection. For marking a checkbox or circling in an oval, standard algorithms may be used to detect the user selection. However, in some instances, the person may seek to select a design or certain text (e.g., “F” or “M” for gender) within a document. For instance, the person may circle an option (e.g., “M”), but these circles may vary between user selections (e.g., a first circular selection may be different from a second circular selection, even in the same document). These user created indications that are circling text or designs may be difficult for a system using standard algorithms to process and assess the user’s selections as they are not pre-defined beforehand. Accordingly, there remains a technical need to be able to detect user created indications.”
AlgorithmsArtificial IntelligenceCyborgsEmerging TechnologiesMachine LearningOptical Character RecognitionPatent Application
NETL
NSTL
