System, method and computer program product for defect detection based on multiple references

1. A defect detection system for computerized detection of defects in an inspected object based on processing of an inspection image, the system comprising:

an interface for receiving inspection image data including information of an analyzed pixel and of a plurality of reference pixels;

a memory; and

a processor operatively coupled to the interface and the memory to:

identify an analyzed pixel from a first cell of the inspection image of the inspected object;

identify a plurality of reference pixels from a plurality of cells of the inspection image of the inspected object, the plurality of cells being other cells in the inspection image than the first cell of the inspection image;

obtain an inspected value representative of the analyzed pixel of the first cell of the inspected image, and a plurality of reference values of the plurality of cells of the inspection image, wherein each of the plurality of reference values is representative of a reference pixel among the plurality of reference pixels from the plurality of cells other than the first cell;

for each of the plurality of reference pixels, calculate a difference between the reference value of a respective reference pixel and the inspected value of the analyzed pixel of the first cell;

compute a representative difference value based on a plurality of the differences; and

determine a presence of a defect in the analyzed pixel based on the representative difference value.

2. The system according to claim 1 , wherein the inspected value and the plurality of reference values are gray-level values.

3. The system according to claim 1 , wherein the plurality of reference pixels are selected based on a selection rule which is based on at least one of:

in-die periodicity or design data used for the manufacturing of the inspected object.

4. The system according to claim 1 , wherein the inspected object is selected from a group consisting of an electronic circuit, a wafer, a photomask, and a reticle.

5. The system according to claim 1 , wherein the processor is to compute the representative difference value by:

selecting a proper subset of the plurality of differences based on the plurality of calculated differences between the inspected value of the analyzed pixel and the reference value of the respective reference pixel; and

defining the representative difference value based on the proper subset of differences.

6. The system according to claim 5 , wherein the processor is to select the proper subset according to a selection rule which is defined based on characteristics of an inspection of the inspected object and on requirements of the defect detection system.

7. The system according to claim 1 , wherein the processor is to compute the representative difference value by selecting one of the plurality of differences as the representative difference value.

8. The system according to claim 1 , wherein the processor is to determine the presence of the defect in the analyzed pixel based on the representative difference value and on a representative noise-indicative value that is computed based on a plurality of noise-indicative values which comprises a noise-indicative value and multiple reference noise-indicative values, the inspected noise-indicative value representative of an analyzed pixel and each of the reference noise-indicative values representative of a reference pixel.

9. A method for computerized detection of defects in an inspected object based on processing of an inspection image of the inspected object, the method comprising:

identifying an analyzed pixel from a first cell of the inspection image of the inspected object;

identifying, by a processor, a plurality of reference pixels from a plurality of cells of the inspection image of the inspected object, the plurality of cells being other cells in the inspection image than the first cell of the inspection image;

obtaining an inspected value representative of the analyzed pixel of the first cell of the inspected image, and a plurality of reference values of the plurality of cells of the inspection image, wherein each of the plurality of reference values is representative of a reference pixel among the plurality of reference pixels from the plurality of cells other than the first cell;

for each of the plurality of reference pixels, calculating a difference between the reference value of a respective reference pixel and the inspected value of the analyzed pixel of the first cell;

computing a representative difference value based on a plurality of the differences; and

determining a presence of a defect in the analyzed pixel based on the representative difference value.

10. The method according to claim 9 , wherein the inspected object is selected from a group consisting of an electronic circuit, a wafer, a photomask, and a reticle.

11. The method according to claim 9 , wherein the computing comprises:

selecting a proper subset of the plurality of differences based on the plurality of calculated differences between the inspected value of the analyzed pixel and the reference value of the respective reference pixel; and

defining the representative difference value based on the proper subset of differences.

12. The method according to claim 9 , wherein the computing comprises selecting one of the plurality of differences as the representative difference value.

13. The method according to claim 9 , wherein the determining of the presence of the defect in the analyzed pixel is based on the representative difference value and on a representative noise-indicative value that is computed based on a plurality of noise-indicative values which comprises a noise-indicative value and multiple reference noise-indicative values, the inspected noise-indicative value representative of an analyzed pixel and each of the reference noise-indicative values is representative of a reference pixel.

14. The method according to claim 9 , wherein the plurality of reference pixels are selected from a same column as the first cell.

15. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform operations for computerized detection of defects in an inspected object based on processing of an inspection image, the operations comprising:

identifying an analyzed pixel from a first cell of the inspection image of the inspected object;

identifying, by the machine, a plurality of reference pixels from a plurality of cells of the inspection image of the inspected object, the plurality of cells being other cells in the inspection image than the first cell of the inspection image;

obtaining an inspected value representative of the analyzed pixel of the first cell, and a plurality of reference values of the plurality of cells of the inspection image, wherein each of the plurality of reference values is representative of a reference pixel among the plurality of reference pixels from the plurality of cells other than the first cell;

for each of the plurality of reference pixels, calculating a difference between the reference value of a respective reference pixel and the inspected value of the analyzed pixel of the first cell;

computing a representative difference value based on a plurality of the differences; and

determining a presence of a defect in the analyzed pixel based on the representative difference value.

16. The program storage device according to claim 15 , wherein the inspected object is selected from a group consisting of an electronic circuit, a wafer, a photomask, and a reticle.

17. The program storage device according to claim 15 , wherein the computing comprises:

selecting a proper subset of the plurality of differences based on the plurality of calculated differences between the inspected value of the analyzed pixel and the reference value of the respective reference pixel; and

defining the representative difference value based on the proper subset of differences.

18. The program storage device according to claim 15 , wherein the computing comprises selecting one of the plurality of differences as the representative difference value.

19. The program storage device according to claim 15 , wherein the determining of the presence of the defect in the analyzed pixel is based on the representative difference value and on a representative noise-indicative value that is computed based on a plurality of noise-indicative values which comprises a noise-indicative value and multiple reference noise-indicative values, the inspected noise-indicative value representative of an analyzed pixel and each of the reference noise-indicative values representative of a reference pixel.

20. The program storage device according to claim 15 , wherein the plurality of reference pixels are selected from a same column as the first cell.