Intuit. has been granted a patent for a method of enhancing document images using machine learning. The approach involves training two independent models on unpaired dirty and clean images, enabling effective image enhancement while addressing privacy concerns. The enhanced images can subsequently be used for optical character recognition (OCR). GlobalData’s report on Intuit gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
According to GlobalData’s company profile on Intuit, Dynamic premium pricing was a key innovation area identified from patents. Intuit's grant share as of June 2024 was 62%. Grant share is based on the ratio of number of grants to total number of patents.
Document image enhancement using unpaired machine learning models
The granted patent US12045967B1 outlines a computer-implemented method and system for enhancing document images using machine learning (ML) techniques. The process begins with obtaining an electronic document image generated from scanning a physical document. The enhancement involves translating the image from a source space to a latent space using a first implicit probabilistic model, which is trained specifically for this purpose. Subsequently, the image is translated from the latent space to a target space through a second implicit probabilistic model, which is trained independently from the first. Both models utilize unpaired training data consisting of two sets of document images, allowing for independent data augmentation techniques, such as sub-windowing and slide-windowing, to generate additional images. The enhanced document image is then provided to an object character recognition (OCR) engine for further processing.
In addition to the method, the patent also describes a computing system that includes processors and memory to execute the aforementioned operations. The system is designed to perform the same steps as the method, ensuring that the electronic document image is enhanced effectively before being processed by the OCR engine. The independent training of the two probabilistic models and the specific data augmentation strategies are key features of this invention, aimed at improving the quality of document images for OCR applications. The claims emphasize the importance of processing one set of images at a time to optimize the enhancement process, thereby enhancing the overall efficiency and accuracy of document image recognition tasks.
To know more about GlobalData’s detailed insights on Intuit, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
