Method and system for detecting drift in image streams
AssigneeCapital One Services, LLC
InventorsKeegan Hines, Christopher Bayan Bruss
Life of this patent
Estimating the term…
Abstract
Methods and systems disclosed herein may quantify a representation of a type of input an image analysis system should expect. The image analysis system may be trained on the type of input the image analysis system should expect using a first image stream. A first model of the type of input that the image analysis system should expect may be built from the first image stream. After the first model is built, a second image, or a second image stream, may be compared to the first model to determine a difference between the second image, or second image stream, and the first image stream. When the difference is greater than or equal to a threshold, a drift may be detected and steps may be taken to determine the cause of the drift.
Claims20
A method comprising:
- receiving, by a first device, a first image stream;
- building a first model based on the first image stream, wherein the first model represents a first probability distribution of a first plurality of features appearing in the first image stream;
- receiving a second image stream;
- building a second model based on the second image stream, wherein the second model represents a second probability distribution of a second plurality of features appearing in the second image stream;
- comparing the second model to the first model to determine a difference between the second image stream and the first image stream;
- indicating that the second image stream is significantly different from the first image stream when the difference is greater than or equal to a threshold, wherein significantly different indicates that the second image stream is outside an accepted range of tolerance.
A computing device comprising:
- an interface configured to receive a plurality of image streams;
- one or more processors;
- and
- memory storing instructions that, when executed by the one or more processors, cause the computing device to:
- receive a first image stream;
- build a first model based on the first image stream, wherein the first model represents a first probability distribution of a first plurality of features appearing in the first image stream;
- receive a second image stream;
- build a second model based on the second image stream, wherein the second model represents a second probability distribution of a second plurality of features appearing in the second image stream;
- compare the second model to the first model to determine a difference between the second image stream and the first image stream;
- indicate that the second image stream is significantly different from the first image stream when the difference is greater than or equal to a threshold, wherein significantly different indicates that the second image stream is outside an accepted range of tolerance.
One or more non-transitory media storing instructions that, when executed by one or more processors, cause the one or more processors to perform steps comprising:
- receiving a first image stream;
- building a first model based on the first image stream, wherein the first model represents a first probability distribution of a first plurality of features appearing in the first image stream;
- receiving a second image stream;
- building a second model based on the second image stream, wherein the second model represents a second probability distribution of a second plurality of features appearing in the second image stream;
- comparing the second model to the first model to determine a difference between the second image stream and the first image stream;
- and
- indicating that the second image stream is significantly different from the first image stream when the difference is greater than or equal to a threshold, wherein significantly different indicates that the second image stream is outside an accepted range of tolerance.
Dependencies are inferred from the claim text.