Reconstruction of a distorted image of an array of structural elements of a specimen
There is provided a method and a system configured to compensate for image distortions. An example method includes first receiving a warped image of an array of cells of a specimen. Each cell of the array comprises one or more structural elements of a substrate. A reference image of a region associated with a cell of the array of cells is generated and a first cell of the array of cells is identified using the reference image and at least part of the warped image. One or more locations comprising cells that differ from the first cell are identified and, based at least on the one or more locations, a warped compensation transform is determined. The warped compensation transform is applied on the warped image to generate an undistorted image.
1. A system, comprising:
a memory device; and
a processing device, operatively coupled to the memory device, the processing device to:
receive a warped image of an array of cells of a specimen, wherein cell of the array of cells comprises one or more structural elements of a substrate;
segment each cell of the array of cells into a respective plurality of sub-cells, wherein a size of one or more of the respective plurality of sub-cells is determined using a recipe and wherein the one or more of the respective plurality of sub-cells comprises distinguishable information;
generate a set of reference images of a set of regions associated with a set of sub-cells from one or more of the respective plurality of sub-cells;
identify a first cell of the array of cells using a reference image of the set of reference images and at least part of the warped image;
select a first sub-cell from a plurality of sub-cells of the first cell;
identify one or more locations comprising sub-cells that differ from the first sub-cell;
determine, based at least on the one or more locations, a warped compensation transform; and
apply the warped compensation transform on the warped image to generate an undistorted image.
2. The system of claim 1 , wherein the processing device is further to:
determine whether array entities of the array of cells are cells or sub-cells based on at least one of (a) a size of a cell, (b) a predefined location accuracy, and (c) a content of a cell.
3. The system of claim 1 , wherein a location of each array entity is a location of a center of the array entity.
4. The system of claim 1 , wherein the processing device is further to:
search for suspected-location array entities that are associated with unreliable determined locations; and
apply the warped compensation transform by determining locations of the suspected-location array entities based on locations of other array entities.
5. The system of claim 1 , wherein the processing device is further to:
apply the warped compensation transform by smoothing locations of different array entities based on differences between location deviations between adjacent array entities.
6. The system of claim 1 , wherein the processing device is further to:
find a location of an array entity by (a) performing a correlation between a reference image of the array entity and a part of the warped image that comprises the array entity and a vicinity of the array entity, and (b) performing a two-dimensional sub-pixel matching.
7. The system of claim 1 , wherein the processing device is further to:
determine the warped compensation transform by defining at least one spline kernel to be applied on at least one part of the warped image.
8. A method, comprising:
receiving a warped image of an array of cells of a specimen, wherein each cell of the array of cells comprises one or more structural elements of a substrate;
segmenting each cell of the array of cells into a respective plurality of sub-cells, wherein a size of one or more of the respective plurality of sub-cells is determined using a recipe and wherein the one or more of the respective plurality of sub-cells comprises distinguishable information;
generating a set of reference images of a set of regions associated with a set of sub-cells from one or more of the respective plurality of sub-cells;
identifying a first cell of the array of cells using a reference image of the set of reference images and at least part of the warped image;
selecting a first sub-cell from a plurality of sub-cells of the first cell;
identifying one or more locations comprising sub-cells that differ from the first sub-cell;
determining, based at least on the one or more locations, a warped compensation transform; and
applying the warped compensation transform on the warped image to generate an undistorted image.
9. The method of claim 8 , further comprising:
determining whether array entities of the array of cells are cells or sub-cells based on at least one of (a) a size of a cell, (b) a predefined location accuracy, and (c) a content of a cell.
10. The method of claim 8 , wherein a location of each array entity is a location of a center of the array entity.
11. The method of claim 8 , further comprising:
searching for suspected-location array entities that are associated with unreliable determined locations; and
applying the warped compensation transform by determining locations of the suspected-location array entities based on locations of other array entities.
12. The method of claim 8 , further comprising:
applying the warped compensation transform by smoothing locations of different array entities based on differences between location deviations between adjacent array entities.
13. The method of claim 8 , further comprising:
finding a location of an array entity by (a) performing a correlation between a reference image of the array entity and a part of the warped image that comprises the array entity and a vicinity of the array entity, and (b) performing a two-dimensional sub-pixel matching.
14. The method of claim 8 , further comprising:
determining the warped compensation transform by defining at least one spline kernel to be applied on at least one part of the warped image.
15. A non-transitory computer readable storage medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising:
receiving a warped image of an array of cells of a specimen, wherein each cell of the array of cells comprises one or more structural elements of a substrate;
segmenting each cell of the array of cells into a respective plurality of sub-cells, wherein a size of one or more of the respective plurality of sub-cells is determined using a recipe and wherein the one or more of the respective plurality of sub-cells comprises distinguishable information;
generating a set of reference images of a set of regions associated with a set of sub-cells from one or more of the respective plurality of sub-cells;
identifying a first cell of the array of cells using a reference image of the set of reference images and at least part of the warped image;
selecting a first sub-cell from a plurality of sub-cells of the first cell;
identifying one or more locations comprising sub-cells that differ from the first sub-cell;
determining, based at least on the one or more locations, a warped compensation transform; and
applying the warped compensation transform on the warped image to generate an undistorted image.
16. The non-transitory computer readable storage medium of claim 15 , wherein the operations further comprise:
determining whether array entities of the array of cells are cells or sub-cells based on at least one of (a) a size of a cell, (b) a predefined location accuracy, and (c) a content of a cell.
17. The non-transitory computer readable storage medium of claim 15 , wherein a location of each array entity is a location of a center of the array entity.
18. The non-transitory computer readable storage medium of claim 15 , wherein the operations further comprise:
searching for suspected-location array entities that are associated with unreliable determined locations; and
applying the warped compensation transform by determining locations of the suspected-location array entities based on locations of other array entities.
19. The non-transitory computer readable storage medium of claim 15 , wherein the operations further comprise:
applying the warped compensation transform by smoothing locations of different array entities based on differences between location deviations between adjacent array entities.
20. The non-transitory computer readable storage medium of claim 15 , wherein the operations further comprise:
finding a location of an array entity by (a) performing a correlation between a reference image of the array entity and a part of the warped image that comprises the array entity and a vicinity of the array entity, and (b) performing a two-dimensional sub-pixel matching.