Multispectrum, multi-polarization (MSMP) filtering for improved perception of difficult to perceive colors

1. A method comprising, by a computing system:

accessing first image data generated by a first image sensor having a first filter array that has a first filter pattern, wherein the first filter pattern comprises a first plurality of filter types, one of which is a first filter type corresponding to a spectrum of interest;

accessing second image data generated by a second image sensor having a second filter array that has a second filter pattern different from the first filter pattern, wherein the second filter pattern comprises a second plurality of filter types, wherein the second plurality of filter types and the first plurality of filter types have at least one filter type in common;

determining a correspondence between one or more first pixels of the first image data and one or more second pixels of the second image data based on a portion of the first image data associated with the at least one filter type in common and a portion of the second image data associated with the at least one filter type in common;

generating a tensor comprising composite data and spatial information using the first image data, the second image data, and the correspondence between the first image data and the second image data, wherein the composite data comprises data associated with the first filter type corresponding to the spectrum of interest; and

detecting, based at least on the composite data of the tensor, one or more objects corresponding to the spectrum of interest.

2. The method of claim 1 , further comprising calculating depth information of the first image data and the second image data based on data of a plurality of polarization filters associated with the first image data or the second image data.

3. The method of claim 2 , wherein calculating the depth information comprises determining a spatial offset between a pixel of the first image data and a corresponding pixel of the second image data based on the portion of the first image data associated with the at least one filter type in common and the portion of the second image data associated with the at least one filter type in common.

4. The method of claim 1 , wherein detecting the one or more objects comprises comparing the data associated with the first filter type corresponding to the spectrum of interest with stored data of objects having the spectrum of interest.

5. The method of claim 1 , wherein relative locations of the at least one filter type in common in the first filter pattern and the second filter pattern are the same.

6. The method of claim 1 , wherein the first filter type corresponding to the spectrum of interest is absent from the second filter pattern.

7. The method of claim 1 , further comprising:

accessing third image data, wherein generating the composite data further comprises using one or more image data generated by one or more additional image sensors, wherein at least one of the one or more additional sensors has a third filter array that has a third filter pattern different from the first filter pattern or the second filter pattern.

8. The method of claim 1 , wherein the detecting comprises comparing, by a machine-learning algorithm, the composite data with stored composite data of one or more objects of interest.

9. The method of claim 1 , wherein the spatial information comprises spatial coordinates associated with the composite data.

10. One or more computer-readable non-transitory storage media embodying software that is operable when executed to cause one or more processors to perform operations comprising:

accessing first image data generated by a first image sensor having a first filter array that has a first filter pattern, wherein the first filter pattern comprises a first plurality of filter types, one of which is a first filter type corresponding to a spectrum of interest;

accessing second image data generated by a second image sensor having a second filter array that has a second filter pattern different from the first filter pattern, wherein the second filter pattern comprises a second plurality of filter types, wherein the second plurality of filter types and the first plurality of filter types have at least one filter type in common;

determining a correspondence between one or more first pixels of the first image data and one or more second pixels of the second image data based on a portion of the first image data associated with the at least one filter type in common and a portion of the second image data associated with the at least one filter type in common;

generating a tensor comprising composite data and spatial information using the first image data, the second image data, and the correspondence between the first image data and the second image data, wherein the composite data comprises data associated with the first filter type corresponding to the spectrum of interest; and

detecting, based at least on the composite data of the tensor, one or more objects corresponding to the spectrum of interest.

11. The media of claim 10 , wherein the software is further operable to calculate depth information of the first image data and the second image data based on data of a plurality of polarization filters associated with the first image data or the second image data.

12. The media of claim 11 , wherein the software is further operable to determine a spatial offset between a pixel of the first image data and a corresponding pixel of the second image data based on the portion of the first image data associated with the at least one filter type in common and the portion of the second image data associated with the at least one filter type in common.

13. The media of claim 10 , wherein the software is further operable to compare the data associated with the first filter type corresponding to the spectrum of interest with stored data of objects having the spectrum of interest.

14. The media of claim 10 , wherein relative locations of the filter type in common in the first filter pattern and the second filter pattern are the same.

15. The media of claim 10 , wherein the first filter type corresponding to the spectrum of interest is absent from the second filter pattern.

16. A computing system comprising:

one or more processors; and

one or more computer-readable non-transitory storage media coupled to one or more of the processors, the one or more computer-readable non-transitory storage media comprising instructions operable when executed by one or more of the processors to cause the computing system to perform operations comprising:

accessing first image data generated by a first image sensor having a first filter array that has a first filter pattern, wherein the first filter pattern comprises a first plurality of filter types, one of which is a first filter type corresponding to a spectrum of interest;

accessing second image data generated by a second image sensor having a second filter array that has a second filter pattern different from the first filter pattern, wherein the second filter pattern comprises a second plurality of filter types, wherein the second plurality of filter types and the first plurality of filter types have at least one filter type in common;

determining a correspondence between one or more first pixels of the first image data and one or more second pixels of the second image data generated by the one or more additional image sensors based on a portion of the first image data associated with the at least one filter type in common and a portion of the second image data associated with the at least one filter type in common;

generating a tensor comprising composite data and spatial information using the first image data the second image data, and the correspondence between the first image data and the second image data, wherein the composite data comprises data associated with the first filter type corresponding to the spectrum of interest; and

detecting, based at least on the composite data of the tensor, one or more objects corresponding to the spectrum of interest.

17. The system of claim 16 , wherein the processors are further operable to calculate depth information of the first image data and the second image data generated by the one or more additional sensors based on data of a plurality of polarization filters associated with the first image data or the second image data.

18. The system of claim 17 , wherein the processors are further operable to determine a spatial offset between a pixel of the first image data and a corresponding pixel of the second image data generated by the one or more additional sensors based on the portion of the first image data associated with the at least one filter type in common and the portion of the second image data associated with the at least one filter type in common.

19. The system of claim 16 , wherein the processors are further operable to compare the data associated with the first filter type corresponding to the spectrum of interest with stored data of objects having the spectrum of interest.

20. The system of claim 16 , wherein relative locations of the filter type in common in the first filter pattern and the second filter pattern are the same.