System and method for verifying authenticity of an anti-counterfeiting element, and method for building a machine learning model used to verify authenticity of an anti-counterfeiting element
AssigneeCheck it out Co., Ltd.
InventorYing-Jen Chen
Life of this patent
Estimating the term…
Abstract
A system for verifying authenticity of an anti-counterfeiting element includes a model building module and a verifying module. The model building module has a pre-processing sub-module for receiving original images of a same reference element and for making adjustments to the original images to form a plurality of true-element images. A training sub-module trains a deep neural network based on the true-element images to build a machine learning model. The verifying module obtains images of the anti-counterfeiting element, and inputs the images into the machine learning model to determine whether the anti-counterfeiting element is authentic or unauthentic.
Claims14
A method for building a machine learning model that is related to a reference anti-counterfeiting element for verifying authenticity of a to-be-examined anti-counterfeiting element, the method comprising steps of:
- receiving a plurality of original images of the reference anti-counterfeiting element;
- with respect to each of the original images, making different adjustments to the original image to form a plurality of true-element images, respectively, wherein the plurality of true-element images formed with respect to a same one of the original images constitute a set;
- inputting plural sets of the true-element images to a deep neural network, the plural sets of the true-element images corresponding respectively to the original images;
- and
- training the deep neural network based on the plural sets of the true-element images to build the machine learning model that is related to the reference anti-counterfeiting element;
- wherein the step of making different adjustments to the original image is to add different lighting effects to the original image so as to form the plurality of true-element images, respectively.
A system for verifying authenticity of a to-be-examined (TBE) anti-counterfeiting element, said system comprising:
- a model building module, including
- a pre-processing sub-module configured to obtain a plurality of original images of a same reference anti-counterfeiting element, and configured to, with respect to each of the original images, make different adjustments to the original image to form a plurality of true-element-images, respectively, wherein the plurality of true-element images formed with respect to a same one of the original images constitute a set, and
- a training sub-module connected to said pre-processing sub-module, and configured to train a deep neural network based on plural sets of the true-element images to build a machine learning model that is related to the reference anti-counterfeiting element, the plural sets of true-element images corresponding respectively to the original images;
- and
- a verifying module configured to obtain a TBE image of the TBE anti-counterfeiting element and to input the TBE image into said machine learning model to determine whether the TBE anti-counterfeiting element in the TBE image is authentic or unauthentic;
- wherein said pre-processing sub-module of said model building module makes different adjustments to the original image by adding different lighting effects to the original image so as to form the plurality of true-element images, respectively.
Dependencies are inferred from the claim text.