Double-layered image classification endpoint solution
A system for image classification is disclosed that includes a central system configured to provide high reliability image data processing and recognition and a plurality of endpoint systems, each configured to provide image data processing and recognition with a lower reliability than the central system and to generate probability data. A decision switch disposed at each of the plurality of endpoint systems is configured to receive the probability data and to determine whether to deny access, grant access or generate a referral message to the central system, wherein the referral message includes at least a set of image data generated at the endpoint system.
1. A system for image classification, comprising:
a central system configured to provide high reliability image data processing and recognition;
a plurality of endpoint systems, each configured to provide image data processing and recognition with a lower reliability than the central system and to generate probability data; and
a decision switch disposed at each of the plurality of endpoint systems, the decision switch configured to receive the probability data and to determine whether to deny access, grant access or generate a referral message to the central system, wherein the referral message includes at least a set of image data generated at the endpoint system.
2. The system of claim 1 wherein the central system further comprises a TensorFlow Inception V3 neural network.
3. The system of claim 1 wherein each of the endpoint systems further comprises a Mobilenet neural network.
4. The system of claim 1 wherein the decision switch further comprises a policy engine configured to receive an output from a neural network and to determine whether the output is greater than a first threshold or less than a second threshold.
5. The system of claim 4 further comprising an incident system configured to receive a control signal from the policy engine if the output is greater than the first threshold and to deny access to a resource.
6. The system of claim 4 further comprising an approval system configured to receive a control signal from the policy engine if the output is less than the second threshold and to allow access to a resource in response.
7. The system of claim 4 further comprising an uncertainty analysis system configured to receive a control signal from the policy engine if the output is less than the first threshold and greater than a second threshold and to generate a control signal for the central system in response.
8. A system for image classification, comprising:
a central system configured to provide high reliability image data processing and recognition;
a plurality of endpoint systems, each configured to provide image data processing and recognition with a lower reliability than the central system and to generate probability data; and
means for receiving the probability data and determining whether to deny access, grant access or generate a referral message to the central system, wherein the referral message includes at least a set of image data generated at the endpoint system.
9. The system of claim 8 wherein the central system further comprises a TensorFlow Inception V3 neural network.
10. The system of claim 8 wherein each of the endpoint systems further comprises a Mobilenet neural network.
11. The system of claim 8 further comprising means for receiving an output from a neural network and determining whether the output is greater than a first threshold or less than a second threshold.
12. The system of claim 11 further comprising an incident system configured to receive a control signal from the policy engine if the output is greater than the first threshold and to deny access to a resource.
13. The system of claim 11 further comprising an approval system configured to receive a control signal from the policy engine if the output is less than the second threshold and to allow access to a resource in response.
14. The system of claim 11 further comprising an uncertainty analysis system configured to receive a control signal from the policy engine if the output is less than the first threshold and greater than a second threshold and to generate a control signal for the central system in response.