Microscopy system and method for reducing an image vignetting

1 . A method for reducing an image vignetting, comprising:

capturing a microscope image of a sample using a microscope, wherein the microscope image exhibits a vignetting;

inputting at least the microscope image and at least one vignetting-free central area of a further microscope image of the sample jointly into an image processing model which has been trained, using training data comprising input images with vignetting and associated target images without vignetting, to calculate a vignetting-free microscope image from at least the input microscope image; and

outputting a calculated vignetting-free microscope image calculated from at least both the input microscope image and the at least one vignetting-free central area of the further microscope image using the image processing model.

2 . The method according to claim 1 , further comprising:

carrying out a training of the image processing model using training data obtained from the sample used to capture the microscope image whose vignetting is to be reduced.

3 . The method according to claim 1 ,

wherein a vignetting-free central area of the microscope image whose vignetting is to be reduced is used as one of the target images of the training data, and wherein a central area with computationally added vignetting is used as an associated input image of the training data.

4 . The method according to claim 1 , further comprising:

capturing a plurality of laterally offset, overlapping microscope images showing the same sample areas once with vignetting and once vignetting-free; and

forming input images for the training data from areas of the microscope images with vignetting, and forming associated target images from vignetting-free areas of the microscope images relating to the same sample areas.

5 . The method according to claim 1 ,

wherein the further microscope image, whose central area is also input into the image processing model, is selected from a plurality of captured microscope images, according to a criterion based on which an image selected as the further microscope image shows a largest number of objects of interest among the plurality of captured microscope images.

6 . The method according to claim 1 , further comprising:

capturing the further microscope image of the same sample area, wherein the further microscope image is captured with a different contrast method than the microscope image;

calculating the vignetting-free microscope image using the image processing model, which is trained to use the further microscope image to calculate a virtually stained microscope image, which is used in the calculation of the vignetting-free microscope image.

7 . The method according to claim 6 ,

wherein, in order to calculate the vignetting-free microscope image, a central area of the microscope image with vignetting is retained and a shaded edge area is replaced by a corresponding image area of the virtually stained microscope image.

8 . The method according to claim 1 ,

wherein the image processing model is trained to receive both the microscope image with vignetting and the further microscope image as inputs and to calculate the vignetting-free microscope image directly therefrom as an output.

9 . A microscopy system with

a microscope configured to carry out the method according to claim 1 .

10 . A non-transitory computer-readable medium including a computer program comprising commands which, when the program is executed by a computer, cause said computer to execute the method according to claim 1 .

11 . The method of claim 1 , further comprising:

capturing at least one vignetting-free contrast image of a sample area captured with the microscope image, wherein the contrast image is captured with a different contrast method than the microscope image;

calculating a virtually stained microscope image from the at least one vignetting-free contrast image using a virtual staining model, wherein a contrast method of the virtually stained microscope image corresponds to a contrast method of the microscope image;

using the image processing model to generate the calculated vignetting-free microscope image from the virtually stained microscope image and the microscope image.

12 . The method according to claim 11 ,

wherein calculating the vignetting-free microscope image includes:

using a central area of the microscope image with vignetting for the calculated vignetting-free microscope image; and

replacing a shaded edge area of the microscope image by a corresponding image area of the virtually stained microscope image.

13 . The method according to claim 11 ,

wherein the image processing model and the virtual staining model are formed as one model.

14 . A method for reducing an image vignetting, comprising:

capturing a microscope image of a sample area using a microscope, wherein the microscope image exhibits a vignetting;

capturing at least one vignetting-free contrast image of the sample area captured with the microscope image, wherein the contrast image is captured with a different contrast method than the microscope image;

calculating a virtually stained microscope image from the at least one vignetting-free contrast image using a virtual staining model, wherein a contrast method of the virtually stained microscope image corresponds to a contrast method of the microscope image;

inputting at least the microscope image and the virtually stained microscope image into an image processing model which includes a model trained with training data comprising input images with vignetting and associated target images without vignetting to calculate a vignetting-free microscope image; and

using the image processing model to calculate a vignetting-free microscope image from the virtually stained microscope image and the microscope image.

15 . The method according to claim 14 ,

wherein calculating the vignetting-free microscope image includes:

using a central area of the microscope image with vignetting for the calculated vignetting-free microscope image; and

replacing or modifying a shaded edge area of the microscope image by a corresponding image area of the virtually stained microscope image.

16 . The method according to claim 1 ,

wherein the microscope image and the at least one vignetting-free central area of the further microscope image show different regions of the sample, wherein the at least one vignetting-free central area of the further microscope image shows a larger number of objects of interest than the microscope image.

17 . The method according to claim 1 ,

wherein the image processing model has been trained using at least a microscope image and a further microscope image jointly as two input images to calculate one vignetting-free output image which is compared with an associated target image without vignetting.

18 . A microscopy system comprising:

a microscope configured to carry out a method for reducing an image vignetting, the method comprising:

capturing a microscope image of a sample area using the microscope, wherein the microscope image exhibits a vignetting;

capturing at least one vignetting-free contrast image of the sample area captured with the microscope image, wherein the contrast image is captured with a different contrast method than the microscope image;

calculating a virtually stained microscope image from the at least one vignetting-free contrast image using a virtual staining model, wherein a contrast method of the virtually stained microscope image corresponds to a contrast method of the microscope image;

inputting at least the microscope image and the virtually stained microscope image into an image processing model which is configured to calculate a vignetting-free microscope image from the virtually stained microscope image and the microscope image.

19 . The system according to claim 18 ,

wherein calculating the vignetting-free microscope image includes:

using a central area of the microscope image with vignetting for the calculated vignetting-free microscope image; and

replacing or modifying a shaded edge area of the microscope image by a corresponding image area of the virtually stained microscope image.

20 . The system according to claim 18 ,

wherein the image processing model includes a model trained with training data comprising input images with vignetting and associated target images without vignetting to calculate the vignetting-free microscope image.