Inspection system and a method for inspecting a diced wafer

1. A method for inspecting a diced wafer that comprises multiple dice, the method comprises;

acquiring, by one or more detectors of an image acquisition unit of an inspection system, multiple images of multiple portions of the diced wafer according to a predefined image acquisition scheme; wherein the multiple portions comprise multiple dice; wherein the acquiring of the multiple images comprises illuminating, by an illuminator, the multiple portions of the diced wafer; wherein the diced wafer is generated by (a) dicing a pre-diced wafer to separate between different dies of the wafer, and (b) preforming an expanding process that involves pulling away the different dice from each other;

locating multiple unique features within the multiple images;

determining, by a processor of the inspection system, positions of the multiple dice in response to locations of the multiple unique features;

and

assigning a die index to each die of the multiple dice in response to uncertainties associated with the dicing and expanding processes;

associating between the multiple dice and multiple reference dice in response to locations of the multiple unique features and to at least one expected die dimension; and

performing at least one out of (i) measuring, by the processor, at least one feature of at least one die, and (ii) detecting, by the processor, defects based upon a comparison between pixels of a die image and pixels of an associated reference die image.

2. The method according to claim 1 wherein the associating comprises determining positions of multiple dice in response to locations of the multiple unique features and in response to and at least one die uncertainty parameter that reflects rotations, shifts and movement of a die during the dicing and the expanding process.

3. The method according to claim 1 wherein the associating comprises generating a diced wafer map that comprises images of the multiple dice and associated die indexes.

4. The method according to claim 1 wherein each die is characterized by a die coordinate system and wherein the method comprises calculating a coordinate transformation from each die coordinate system to a global coordinate system.

5. The method according to claim 4 wherein the calculating comprises determining a relationship between each die coordinate system to a frame coordinate system; wherein the multiple images are aligned with the frame coordinate system; wherein the coordinate transformation is responsive to the determined relationship and to a relationship between the global coordinate system and the frame coordinate system.

6. The method according to claim 1 wherein the acquiring comprises introducing a continuous mechanical translation between the diced wafer and an image acquisition unit.

7. The method according to claim 1 wherein the locating comprises locating a first unique feature and the associating comprises assigning a first die index to a first dice that comprises the unique feature.

8. The method according to claim 7 wherein the locating comprising locating another unique feature that belongs to another dice and assigning another die index to the other dice; wherein a relationship between the first and the other die indexes is responsive to a spatial relationship between the first and other unique features.

9. The method according to claim 1 further comprising generating a reference wafer map.

10. The method according to claim 1 wherein the associating comprises performing a registration of the diced wafer map in relation of a reference wafer map.

11. The method according to claim 1 further comprising detecting defects based upon the comparison between pixels of the die image and pixels of the associated reference die image.

12. The method according to claim 1 wherein the stage of locating multiple unique features at least partially overlaps the stage of acquiring multiple images.

13. The method according to claim 1 wherein the stage of associating between the multiple dice and the multiple reference dice at least partially overlaps the stage of acquiring multiple images.

14. The method according to claim 1 wherein the stage of locating comprises applying normalized correlation or geometric hashing.

15. The method according to claim 1 wherein the stage of locating comprises applying multiple image recognition techniques.

16. The method according to claim 1 wherein the stage of associating is followed by measuring the at least one feature of the at least one die.

17. The method according to claim 1 wherein the acquiring comprises acquiring partially overlapping images.

18. An inspection system, the system comprises:

an illuminator that is adapted to illuminate multiple portions of a diced wafer;

an image acquisition unit that comprises one or more detectors that are adapted to acquire multiple images, according to a predefined image acquisition scheme, of the multiple portions of the diced wafer that comprises multiple dice;

wherein the multiple portions comprise multiple dice; wherein the diced wafer is generated by (a) dicing a pre-diced wafer to separate between different dies of the wafer, and (b) preforming an expanding process that involves pulling away the different dice from each other; and

a processor adapted to (a) locate multiple unique features within the multiple images, (b) determine positions of the multiple dice in response to locations of the multiple unique features; (c) assign a die index to each dice of the multiple dice in response to uncertainties associated with the dicing and expanding processes; associate between the multiple dice and multiple reference dice in response to locations of the multiple unique features, to at least one expected die; and perform at least one out of (i) measure at least one feature of at least one die, and (ii) detect defects based upon a comparison between pixels of a die image and pixels of an associated reference die image.

19. The system according to claim 18 wherein the processor is adapted to determine positions of multiple dice in response to locations of the multiple unique features and in response to at least one die uncertainty parameter that reflects rotations, shifts and movement of a die during the dicing and the expanding process.

20. The system according to claim 18 wherein the processor is adapted to generate a diced wafer map that comprises images of the multiple dice and associated die indexes.

21. The system according to claim 18 wherein each die is characterized by a die coordinate system and wherein the processor is adapted to calculate a coordinate transformation from each die coordinate system to a global coordinate system.

22. The system according to claim 21 wherein the processor is adapted to determine a relationship between each die coordinate system to a frame coordinate system; wherein the multiple images are aligned with the frame coordinate system; wherein the coordinate transformation is responsive to the determined relationship and to a relationship between the global coordinate system and the frame coordinate system.

23. The system according to claim 18 further comprising a mechanical stage adapted to introduce a continuous mechanical translation between the diced wafer and the image acquisition unit.

24. The system according to claim 18 wherein the processor is adapted to locate a first unique feature and assign a first die index to a first dice that comprises the unique feature.

25. The system according to claim 24 wherein the processor is adapted to locate another unique feature that belongs to another dice and assign another die index to the other dice; wherein a relationship between the first and the other die indexes is responsive to a spatial relationship between the first and the other unique features.

26. The system according to claim 18 wherein the system comprises a storage unit adapted to store a reference wafer map.

27. The system according to claim 18 wherein the processor is adapted to perform a registration of the diced wafer map in relation of a reference wafer map.

28. The system according to claim 18 the processor is adapted to detect defects based upon a the comparison between pixels of a the die image and pixels of an the associated reference die image.

29. The system according to claim 18 wherein the processor is adapted to locate at least one unique feature while the image acquisition unit acquires at least one image.

30. The system according to claim 18 wherein the processor is adapted to associate between at least one die and at least one reference die while the image acquisition unit acquires at least one image.

31. The system according to claim 18 wherein the processor is adapted to locate at least one unique feature by applying normalized correlation or geometric hashing.

32. The system according to claim 18 wherein the processor is adapted to locate at least one unique feature by applying multiple image recognition techniques.

33. The system according to claim 18 wherein the processor is adapted to measure the at least one feature of the at least one die.

34. The system according to claim 18 wherein the image acquisition unit is adapted to acquire partially overlapping images.

35. The system according to claim 18 wherein the processor is adapted to replace the die indexes by dice indexes that are related to dice indexes in the pre-diced wafer.

36. The method according to claim 1 comprising replacing the die indexes by dice indexes that are related to dice indexes in the pre-diced wafer.