Methods and apparatus to autonomously detect thermal anomalies

1. An apparatus to detect engine anomalies comprising:

at least one memory to store instructions; and

processor circuitry to execute the instructions to:

control a plurality of infrared cameras to capture a baseline image set, the baseline image set including at least two thermal images;

generate emissivity data based on the baseline image set, wherein the emissivity data is generated based on (1) associating a recognized object with an emissivity value stored in a database or (2) applying multi-color pyrometry to electromagnetic radiation of two different wavelength bands;

provide the baseline image set and the emissivity data to an artificial intelligence model, the artificial intelligence model to generate a reconstructed image set;

determine a difference between the baseline image set and the reconstructed image set based on:

a ratio of the baseline image set to the reconstructed image set; and

a neighboring area estimation; and

in response to the difference exceeding a threshold, generate an alert indicating detection of an engine anomaly.

2. The apparatus of claim 1 , wherein the artificial intelligence model includes at least one of a variational autoencoder and a generative adversarial neural network.

3. The apparatus of claim 1 , wherein the difference is determined based on a ratio of an image of the baseline image set to an image of the reconstructed image set.

4. The apparatus of claim 1 , wherein in response to a first camera of the plurality of infrared cameras detecting the engine anomaly, the processor circuitry is to execute the instructions to direct a second camera of the plurality of infrared cameras to image the engine anomaly.

5. The apparatus of claim 1 , wherein the processor circuitry is to execute the instructions to generate a two-dimensional temperature map.

6. The apparatus of claim 1 , wherein the engine anomaly is an engine leak in an under cowl area of a gas turbine engine.

7. A non-transitory computer readable medium comprising instructions, which, when executed, cause processor circuitry to at least:

control a plurality of infrared cameras to capture a baseline image set, the baseline image set including at least two thermal images;

generate emissivity data based on the baseline image set, wherein the emissivity data is generated based on (1) associating a recognized object with an emissivity value stored in a database or (2) applying multi-color pyrometry to electromagnetic radiation of two different wavelength bands;

provide the baseline image set and the emissivity data to an artificial intelligence model, the artificial intelligence model to generate a reconstructed image set;

determine a difference between the baseline image set and the reconstructed image set based on:

a ratio of the baseline image set to the reconstructed image set; and

a neighboring area estimation; and

in response to the difference exceeding a threshold, generate an alert indicating detection of an engine anomaly.

8. The non-transitory computer readable medium of claim 7 , wherein the artificial intelligence model includes at least one of a variational autoencoder and a generative adversarial neural network.

9. The non-transitory computer readable medium of claim 7 , wherein the difference is determined based on a ratio of an image of the baseline image set to an image of the reconstructed image set.

10. The non-transitory computer readable medium of claim 7 , wherein in response to a first camera of the plurality of infrared cameras detecting the engine anomaly, the processor circuitry is to execute the instructions to direct a second camera of the plurality of infrared cameras to image the engine anomaly.

11. The non-transitory computer readable medium of claim 7 , wherein the processor circuitry is to execute the instructions to generate a two-dimensional temperature map.

12. The non-transitory computer readable medium of claim 7 , wherein the engine anomaly is an engine leak in an under cowl area of a gas turbine engine.

13. A method comprising:

controlling a plurality of infrared cameras to capture a baseline image set, the baseline image set including at least two thermal images;

generating emissivity data based on the baseline image set, wherein the emissivity data is generated based on (1) associating a recognized object with an emissivity value stored in a database or (2) applying multi-color pyrometry to electromagnetic radiation of two different wavelength bands;

providing the baseline image set and the emissivity data to an artificial intelligence model, the artificial intelligence model to generate a reconstructed image set;

determining a difference between the baseline image set and the reconstructed image set based on:

a ratio of the baseline image set to the reconstructed image set; and

a neighboring area estimation; and

in response to the difference exceeding a threshold, generating an alert indicating detection of an engine anomaly.

14. The method of claim 13 , wherein the artificial intelligence model includes at least one of a variational autoencoder and a generative adversarial neural network.

15. The method of claim 13 , wherein determining the difference is based on a ratio of an image of the baseline image set to an image of the reconstructed image set.

16. The method of claim 13 , further including generating a two-dimensional temperature map.

17. The method of claim 13 , wherein the engine anomaly is an engine leak in an under cowl area of a gas turbine engine.