Method and device for identifying an object in a search image

1. A method for identifying an object in a search image, the method comprising:

recording the object and surroundings around the object with an optics unit of a camera to generate a recorded image;

swivelling the optics unit in the surroundings during an integration of a recorded image; and

back calculating, with an image processing unit, the search image from the recorded image using a point spread function and a swivelling movement of the optics unit; and

searching the search image for the object to identify the object in the search image.

2. The method according to claim 1 , wherein the back calculating step comprises using a convolution kernel of the point spread function and of a point blurring by the swivelling movement, and calculating the search image by inverse convolution with the convolution kernel.

3. The method according to claim 1 , which comprises completely covering a search space by a sequence of recorded images that are recorded during continuous swivelling of the optics unit.

4. The method according to claim 1 , which comprises defining a swivelling movement during the integration of the recorded image over at least 10% and at most 50% of an image diameter in the swivelling direction.

5. The method according to claim 1 , which comprises using a Wiener filter in the back calculating step.

6. The method according to claim 1 , which comprises using noise in a frequency space caused by the optics unit as a parameter in the back calculating step.

7. The method according to claim 1 , which comprises using readout noise caused by readout electronics of the camera as a parameter in the back calculating step.

8. The method according to claim 1 , which comprises using detector noise in the camera as a parameter in the back calculating step.

9. The method according to claim 1 , which comprises using background noise of the surroundings ascertained from an earlier image recording as a parameter in the back calculating step.

10. The method according to claim 9 , which comprises ascertaining the background noise with spatial dependency by way of a sequence of recorded images.

11. An apparatus for identifying an object in a search image, the apparatus comprising:

a camera with an optics unit, a detector, and a drive for swivelling said optics unit in the surroundings;

a swivel capture unit configured for capturing a swivelling of said optics unit;

an image processing unit configured to control image recording during driven swivelling of said optics unit and to capture the swivelling by way of the swivel capture unit during the image recording, said image processing unit being configured to back calculate a search image from the recorded image using a point spread function and the captured swivelling movement of said optics unit and to search the search image for the object.