Abstract
From Washington, D.C., NewsRx journalists report that a patent application by the inventors DUAN, Yue (Allen, TX, US); HO, Chi-San (Allen, TX, US); PRICE, Micah (The Colony, TX, US), filed on October 3, 2023, was made available online on February 1, 2024. The patent’s assignee is Capital One Services LLC (McLean, Virginia, United States). News editors obtained the following quote from the background information supplied by the inventors: “Image classification models may be used to classify or categorize images based on their parameters and/or extracted attributes. Using machine learning methods, such as convolutional neural networks (CNNs), it is possible to categorize images. For example, images may be manipulated to be run in a CNN to either recognize or classify the images. Further, CNNs or-similar machine learning methods-may be employed to extract image parameters that are then correlated with other parameters or attributes for classification. These methods can be powerful to identify or categorize images because the correlations can be highly accurate. Moreover. CNNs may be updated or retrained frequently to accurately capture image parameters.
NETL
NSTL