Deep photometric learning (DPL) systems, apparatus and methods

1. An imaging system, comprising:

an imaging apparatus comprising a plurality of light sources positioned at a plurality of positions and a plurality of angles relative to a stage configured to support a specimen, wherein the imaging apparatus is configured to capture a plurality of images of a surface of the specimen; and

a computing system in communication with the imaging apparatus, the computing system configured to generate a 3D-reconstruction of the surface of the specimen by:

training a deep learning model to generate 3D-reconstructions of surfaces of specimens by:

generating a plurality of training data sets, wherein the plurality of training data sets comprises a plurality of synthetic height maps, the plurality of training data sets comprising a first training data set without noise and without tilt, a second training data set comprising a fixed amount of noise and no tilt, and a third training data set comprising tilt and no noise, and

training the deep learning model using the plurality of training data sets;

receiving, from the imaging apparatus, the plurality of images of the surface of the specimen;

generating, by the deep learning model, a height map of the surface of the specimen based on the plurality of images; and

outputting the 3D-reconstruction of the surface of the specimen based on the height map generated by the deep learning model.

2. The imaging system of claim 1 , wherein the deep learning model comprises a convolutional neural network.

3. The imaging system of claim 2 , wherein the convolutional neural network is configured to receive, as input, the plurality of images and output a single image.

4. The imaging system of claim 1 , wherein the computing system is further configured to test the deep learning model by:

generating a plurality of microscopic images of an example specimen; and

testing an accuracy of the deep learning model using the plurality of microscopic images.

5. The imaging system of claim 4 , wherein the plurality of microscopic images comprise images taken with a fixed camera angle and varying illumination directions.

6. A method of generating a 3D-reconstruction of a surface of a specimen, comprising:

generating, by a computing system, a plurality of training data sets, wherein the plurality of training data sets comprises a plurality of synthetic height maps, the plurality of training data sets comprising a first training data set without noise and without tilt, a second training data set comprising a fixed amount of noise and no tilt, and a third training data set comprising tilt and no noise;

training, by the computing system, a deep learning model using the plurality of training data sets;

receiving, by the computing system from an imaging apparatus, a plurality of images of a surface of a specimen positioned on a stage of the imaging apparatus;

generating, by the computing system via the deep learning model, a height map of the surface of the specimen based on the plurality of images; and

outputting, by the computing system, a 3D-reconstruction of the surface of the specimen based on the height map generated by the deep learning model.

7. The method of claim 6 , wherein the deep learning model comprises a convolutional neural network.

8. The method of claim 7 , wherein the convolutional neural network is configured to receive, as input, the plurality of images and output a single image.

9. The method of claim 6 , further comprising:

generating, by the computing system, a plurality of microscopic images of an example specimen; and

testing, by the computing system, an accuracy of the deep learning model using the plurality of microscopic images.

10. The method of claim 9 , wherein the plurality of microscopic images comprises images taken with a fixed camera angle and varying illumination directions.

11. A non-transitory computer readable medium comprising one or more sequences of instructions, which, when executed by one or more processors, cause a computing system to perform operations, comprising:

generating, by the computing system, a plurality of training data sets, wherein the plurality of training data sets comprises a plurality of synthetic height maps, the plurality of training data sets comprising a first training data set without noise and without tilt, a second training data set comprising a fixed amount of noise and no tilt, and a third training data set comprising tilt and no noise;

training, by the computing system, a deep learning model using the plurality of training data sets;

receiving, by the computing system from an imaging apparatus, a plurality of images of a surface of a specimen positioned on a stage of the imaging apparatus;

generating, by the computing system via the deep learning model, a height map of the surface of the specimen based on the plurality of images; and

outputting, by the computing system, a 3D-reconstruction of the surface of the specimen based on the height map generated by the deep learning model.

12. The non-transitory computer readable medium of claim 11 , wherein the deep learning model comprises a convolutional neural network.

13. The non-transitory computer readable medium of claim 11 , further comprising:

generating, by the computing system, a plurality of microscopic images of an example specimen; and

testing, by the computing system, an accuracy of the deep learning model using the plurality of microscopic images.

14. The non-transitory computer readable medium of claim 13 , wherein the plurality of microscopic images comprises images taken with a fixed camera angle and varying illumination directions.