Compressive sensing capturing device and method
A compressive sensing capturing device has circuitry, which is configured to obtain compressive sensing image data; and to set a device attribute based on image attribute data, wherein the image attribute data are based on a machine learning algorithm performed in the compressing domain on the obtained compressive sensing image data.
1. A compressive sensing capturing device, comprising circuitry configured to:
obtain compressive sensing image data; and
set a device attribute based on image attribute data, wherein the image attribute data are based on a machine learning algorithm performed in the compressing domain on the obtained compressive sensing image data.
2. The compressive sensing capturing device of claim 1 , wherein the device attribute includes at least one of: autofocus, luminance, shutter speed, aperture size, shoot mode, flash mode, object detected.
3. The compressive sensing capturing device of claim 1 , wherein the machine learning algorithm is based on an artificial neural network using the obtained compressive sensing image data as input.
4. The compressive sensing capturing device of claim 3 , wherein the compressive sensing image data are block-based.
5. The compressive sensing capturing device of claim 4 , wherein one block of compressive sensing image data is used as input for the artificial neural network.
6. The compressive sensing capturing device of claim 5 , wherein for each block of compressive sensing image data at least two different measurements are provided to the artificial neural network.
7. The compressive sensing capturing device of claim 1 , wherein the circuitry further comprises a compressive imaging sensor.
8. The compressive sensing capturing device of claim 7 , wherein the device is configured as digital camera.
9. The compressive sensing capturing device of claim 7 , wherein the compressive sensing image data are obtained from the compressive imaging sensor.
10. The compressive sensing capturing device of claim 7 , wherein the circuitry is further configured to control the image sensor based on the set device attribute.
11. A compressive sensing capturing method, comprising:
obtaining compressive sensing image data; and
setting a device attribute based on image attribute data, wherein the image attribute data are based on a machine learning algorithm performed in the compressing domain on the obtained compressive sensing image data.
12. The compressive sensing capturing method of claim 11 , wherein the device attribute includes at least one of: autofocus, luminance, shutter speed, aperture size, shoot mode, flash mode, object detected.
13. The compressive sensing capturing method of claim 11 , wherein the machine learning algorithm is based on an artificial neural network using the obtained compressive sensing image data as input.
14. The compressive sensing capturing method of claim 13 , wherein the compressive sensing image data are block-based.
15. The compressive sensing capturing method of claim 14 , wherein one block of compressive sensing image data is used as input for the artificial neural network.
16. The compressive sensing capturing method of claim 15 , wherein for each block of compressive sensing image data at least two different measurements are provided to the artificial neural network.
17. The compressive sensing capturing method of claim 11 , further comprising generating the image attribute data.
18. The compressive sensing capturing method of claim 17 , wherein the image attribute data are generated by performing the machine learning algorithm on the obtained compressive sensing image data.
19. The compressive sensing capturing device of claim 11 , wherein the compressive sensing image data are obtained from a compressive imaging sensor.
20. The compressive sensing capturing device of claim 19 , further comprising controlling the image sensor based on the set device attribute.