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 method for image classification, comprising:
scoring image data at an endpoint system to generate a probability;
determining whether the probability is above a first threshold and below a second threshold;
transmitting the image data to a central system for scoring if the probability is above the first threshold and below the second threshold; and
denying access to a resource if the probability is above the second threshold.
2. The method of claim 1 further comprising generating a second probability with a TensorFlow Inception V3 neural network at the central system if the probability is above the first threshold and below the second threshold.
3. The method of claim 1 further comprising a plurality of endpoint systems, wherein each of the endpoint systems is further configured to use a Mobilenet neural network to generate the probability.
4. The method of claim 1 further comprising providing access to a resource if the probability is below the first threshold.
5. A method for data processing, comprising:
scoring image data at an access system to generate a probability;
determining whether the probability is above a first threshold and below a second threshold;
transmitting the image data to a central system for scoring if the probability is above the first threshold and below the second threshold; and
providing access to a resource if the probability is below the first threshold.
6. The method of claim 5 further comprising generating a second probability with a TensorFlow Inception V3 neural network at the central system if the probability is above the first threshold and below the second threshold.
7. The method of claim 5 wherein the access system is further configured to use a Mobilenet neural network to generate the probability.
8. The method of claim 5 further comprising denying access to a resource if the probability is above the second threshold.
9. The method of claim 5 further comprising a plurality of access systems, wherein each access system is further configured to use a Mobilenet neural network to generate the probability.
10. The method of claim 5 further comprising generating a second probability with an image data recognition neural network at the central system if the probability is above the first threshold and below the second threshold.
11. The method of claim 5 wherein an accuracy of scoring the image data at the access system is lower than an accuracy of scoring the image data at the central system.
12. The method of claim 5 further comprising transmitting an approval from the central system to the access system if the probability generated at the central system is below a predetermined level.
13. The method of claim 5 further comprising transmitting a denial from the central system to the access system if the probability generated at the central system is above a predetermined level.
14. The method of claim 5 further comprising transmitting an approval from the central system to the access system if a score generated at the central system meets one or more predetermined criteria.
15. The method of claim 5 further comprising generating an incident report at the central system if a score generated at the central system meets one or more predetermined criteria.
16. The method of claim 5 further comprising transmitting a denial from the central system to the access system if a score generated at the central system meets one or more predetermined criteria.
17. A method for data processing, comprising:
scoring image data at an access system to generate a probability;
determining whether the probability is above a first threshold and below a second threshold;
transmitting the image data to a central system for scoring if the probability is above the first threshold and below the second threshold; and
wherein an accuracy of scoring the image data at the access system is lower than an accuracy of scoring the image data at the central system.
18. The method of claim 17 further comprising generating a second probability with a neural network at the central system if the probability is above the first threshold and below the second threshold.
19. The method of claim 17 wherein the access system is further configured to use a neural network to generate the probability.
20. The method of claim 17 further comprising denying access to a resource if the probability is above the second threshold.