Thermal imaging for self-driving cars

1. A computer-implemented method, comprising:

receiving, at a computing device, one or more infrared images of an environment;

receiving a current spatial point cloud of the environment;

applying, using the computing device, a trained machine learning system on the one or more infrared images to determine an identified object type in the environment by at least:

identifying objects in the environment based on the current spatial point cloud;

calculating emissivity values for the identified objects; and

determining a current thermal map associated with the environment based on the one or more infrared images; and

determining the identified object type based on the current thermal map and the calculated emissivity values.

2. The computer-implemented method of claim 1 , wherein the identified object type is a living object.

3. The computer-implemented method of claim 1 , wherein determining the current thermal map comprises:

receiving, from a plurality of sensors communicatively coupled to the computing device, a current thermal state of the environment; and

based on the current thermal state, determining one or more prior thermal maps from a plurality of prior thermal maps stored on the computing device, wherein determining the current thermal map is further based on the determined prior thermal maps.

4. The computer implemented method of claim 3 , wherein the plurality of sensors comprises of at least one of: an ambient temperature sensor, a global positioning system (GPS) sensor, and a solar sensor operable to determine a current position of the sun.

5. The computer-implemented method of claim 3 , wherein the computing device is incorporated into a scanning laser system, wherein each of the one or more prior thermal maps is associated with an angle of the scanning laser system, and wherein determining the one or more prior thermal maps is further based on a current angle of the scanning laser system.

6. The computer implemented method of claim 3 , wherein the one or more prior thermal maps comprise one or more regions that have been altered, redacted, or deleted based on a removed object type.

7. The computer-implemented method of claim 3 , wherein the one or more prior thermal maps comprise thermal maps determined by a second computing device positioned in a different location in the environment from the computing device.

8. The computer-implemented method of claim 7 , wherein the thermal maps determined by the second computing device are determined within a threshold time limit from when the computing device receives the one or more infrared images.

9. The computer-implemented method of claim 1 , wherein determining the current thermal map associated with the environment comprises calibrating a thermal contrast of the one or more infrared images.

10. The computer-implemented method of claim 1 , wherein the current thermal map comprises the current spatial point cloud projected onto the one or more infrared images.

11. The computer-implemented method of claim 1 , further comprising:

determining current visible light images based on data from a camera system;

identifying, using the current visible light images, objects in the environment; and

calculating emissivity values for the identified objects, wherein determining the identified object type is further based on the calculated emissivity values.

12. The computer-implemented method of claim 1 , wherein the one or more infrared images comprises a set of temporally successive infrared images of the environment, and wherein applying the trained machine learning system on the one or more infrared images comprises a spatio-temporal association of the identified object type between the one or more infrared images.

13. The computer-implemented method of claim 12 , wherein the spatio-temporal association comprises assigning unique identifiers to each object in the environment that is of the identified object type.

14. The computer-implemented method of claim 1 , wherein the trained machine learning system comprises a convolutional neural network.

15. The computer-implemented method of claim 14 , wherein the convolutional neural network is a segmentation network having an encoding path and a decoding path.

16. The computer-implemented method of claim 1 , wherein the computing device is part of an autonomous vehicle system.

17. A computer-implemented method, comprising:

receiving, at a computing device, one or more infrared images of an environment, wherein the one or more infrared images comprise infrared images with spectral intensity information corresponding to a plurality of respective spectral wavebands;

training, using the computing device, a machine learning system on the one or more infrared images to determine an identified object type in the environment by at least:

training the machine learning system to determine one or more prior thermal maps associated with the environment,

using the one or more prior thermal maps and the one or more infrared images, training the machine learning system to determine a current thermal map associated with the environment, wherein training the machine learning system to determine the current thermal map associated with the environment comprises calculating the emissivity of objects in the environment; and

training the machine learning system to determine the identified object type using the current thermal map.

18. A computing device, comprising:

one or more processors; and

data storage, wherein the data storage has stored thereon computer-executable instructions that, when executed by the one or more processors, cause the computing device to carry out functions comprising:

receiving one or more infrared images taken of an environment;

receiving a current spatial point cloud of the environment;

applying a trained machine learning system to the one or more infrared images to determine an identified object type in the environment by at least:

identifying objects in the environment using the current spatial point cloud; and

calculating emissivity values for the identified objects,

determining one or more prior thermal maps associated with the environment,

determining a current thermal map associated with the environment based on the one or more infrared images; and

determining the identified object type using the current thermal map and the calculated emissivity values.

19. The computing device of claim 18 , wherein the trained machine learning system comprises a convolutional neural network.

20. The computing device of claim 19 , wherein the convolutional neural network is a segmentation network having an encoding path and a decoding path.