Hologram detection service providing server and hologram detection method
AssigneeKakaoBank Corp.
InventorDong Yul Lee
Life of this patent
Estimating the term…
Abstract
A hologram detection method according to an aspect of the disclosure, includes: inputting a first image, obtained by capturing a detection object on the basis of first flash intensity, to a neural network model to obtain a first detection result value representing the detection or not of a hologram for each of predetermined at least one detection unit regions; and comparing a threshold value with the first detection result value obtained for each detection unit region to determine the detection or not of a hologram in the first image and a first detection unit region where a hologram is detected.
Claims6
A hologram detection method, the hologram detection method comprising:
- setting a plurality of detection unit regions located at different positions in a detection object;
- obtaining a first image by capturing the detection object in a flash off state;
- inputting the first image to a neural network model to obtain a first detection result value representing whether a hologram is detected and a detection region for each of the plurality of detection unit regions;
- changing flash intensity in a flash on state;
- obtaining a second image by capturing the detection object on the basis of the changed flash intensity;
- inputting the second image to the neural network model to obtain a second detection result value representing whether a hologram is detected and a detection region for each of the plurality of detection unit regions;
- comparing the detection region of the first detection result value and the second detection result value to determine the authenticity or not of the detection object, and when the detection region where the hologram is detected is different from each other, determining that the detection object is normal,
- wherein the neural network model is learned by using a plurality of learning images,
- wherein each of the plurality of learning images includes a plurality of regions, where the presence of a hologram has been checked, and each of the plurality of regions includes labeling value representing whether a hologram is present.
A hologram detection service providing server, the hologram detection service providing server comprising:
- a detection unit region setting unit configured to set a plurality of detection unit regions located at different positions in a detection object;
- an image receiver configured to receive a first image obtained by capturing the detection object in a flash off state and a second image obtained by capturing the detection object on the basis of a flash intensity in a flash on state;
- a hologram detector configured to input the first image to a neural network model to obtain a first detection result value representing whether a hologram is detected and a detection region for each of the plurality of detection unit regions, and input the second image to the neural network model to obtain a second detection result value representing whether a hologram is detected and a detection region for each of the plurality of detection unit regions;
- an authenticity determiner configured to compare the detection region of the first detection result value and the second detection result value to determine the authenticity or not of the detection object, and determine that the detection object is normal when the detection region where the hologram is detected is different from each other;
- wherein the neural network model is learned by using a plurality of learning images,
- wherein each of the plurality of learning images includes a plurality of regions, where the presence of a hologram has been checked, and each of the plurality of regions includes labeling value representing whether a hologram is present.
Dependencies are inferred from the claim text.