System and a method for automatic recipe validation and selection

Abstract

A system, a non-transitory computer program product and a method for selecting an inspection recipe, the method includes: (i) obtaining an image of a structural element of the semiconductor device; (ii) calculating multiple types of distances between the image of the structural element and each of a plurality of reference images obtained by applying a plurality of inspection recipes; and (iii) automatically selecting at least one selected inspection recipe out of the plurality of inspection recipes based on values of the multiple types of distances.

We claim: 
     
         1 . A method for selecting an inspection recipe, the method comprises:
 obtaining an image of a structural element of the semiconductor device;   calculating, by a processor, multiple types of distances between an image of the structural element and each of a plurality of reference images obtained by applying a plurality of inspection recipes; wherein the multiple types of distances are calculated by using different algorithms;   wherein different inspection recipes of the plurality of inspection recipes are optimized for different conditions of the semiconductor device; wherein each reference image is associated with an inspection recipe before the obtaining of the image of the structural element; and   automatically selecting, by the processor, at least one selected inspection recipe out of the plurality of inspection recipes based on values of the multiple types of distances.   
     
     
         2 . The method according to  claim 1 , wherein the calculating, by the processor, of the multiple types of distances comprises calculating, by the processor, each one out of (i) a difference between gray level histograms, (ii) a rectilinear (L 1 ) distance, (iii) Euclidian (L 2 ) length, (iv) Chi-square distance, (v) Bhattacharyya distance, (vi) Wasserstein metric, (vii) Swain metric, and (viii) normalized correlation. 
     
     
         3 . The method according to  claim 1 , wherein the calculating, by the processor of the multiple types of distances comprises calculating, by the processor, a majority of distances out of a group consisting of: (i) a difference between gray level histograms, (ii) a rectilinear (L 1 ) distance, (iii) Euclidian (L 2 ) length, (iv) Chi-square distance, (v) Bhattacharyya distance, (vi) Wasserstein metric, (vii) Swain metric, and (viii) normalized correlation. 
     
     
         4 . The method according to  claim 1 , comprising generating, by the processor, a match value for each reference image based on values of the multiple types of distances between the image of the structural element and the reference image; and selecting each inspection recipe associated with a reference image that has a match value within an allowable match value range. 
     
     
         5 . The method according to  claim 1 , comprising selecting, by the processor, multiple selected inspection recipes. 
     
     
         6 . The method according to  claim 1 , comprising changing, by the processor, the type of distances to be calculated based on evaluation results obtained by applying the at least one selected inspection recipe. 
     
     
         7 . The method according to  claim 1 , wherein the structural elements are coated by a coating material that changes an optical property over time, wherein the plurality of inspection recipes differ from each other by a value of optical property of the coating material they are tuned to. 
     
     
         8 . An inspection system, comprising:
 an image obtaining module that is adapted to obtain an image of a structural element of the semiconductor device;   a processor that comprises:   a distance calculator that is adapted to calculate multiple types of distances between the image of the structural element and each of a plurality of reference images; wherein the multiple types of distances are calculated by using different algorithms;   wherein the plurality of reference images are obtained by applying a plurality of inspection recipes; wherein different inspection recipes of the plurality of inspection recipes are optimized for different conditions of the semiconductor device; wherein each reference image is associated with an inspection recipe before the obtaining of the image of the structural element; and   a selecting module that is adapted to automatically select at least one selected inspection recipe out of the plurality of inspection recipes based on values of the multiple types of distances.   a controller that is adapted to control the image obtaining module to apply each of the at least one selected inspection recipe.   
     
     
         9 . The system according to  claim 8 , wherein a calculation of the multiple types of distances by the distance calculator comprises calculating (i) a difference between gray level histograms, (ii) a rectilinear (L 1 ) distance, (iii) Euclidian (L 2 ) length, (iv) Chi-square distance, (v) Bhattacharyya distance, (vi) Wasserstein metric, (vii) Swain metric, and (viii) normalized correlation. 
     
     
         10 . The system according to  claim 8 , wherein a calculation of the multiple types of distances by the distance calculator comprises calculating a majority of distances out of a group consisting of: (i) a difference between gray level histograms, (ii) a rectilinear (L 1 ) distance, (iii) Euclidian (L 2 ) length, (iv) Chi-square distance, (v) Bhattacharyya distance, (vi) Wasserstein metric, (vii) Swain metric, and (viii) normalized correlation. 
     
     
         11 . The system according to  claim 8 , wherein the distance calculator is adapted to generate a match value for each reference image based on values of the multiple types of distances between the image of the structural element and the reference image; and selecting each inspection recipe associated with a reference image that has a match value within an allowable match value range. 
     
     
         12 . The system according to  claim 8 , wherein the selecting module is adapted to select multiple selected inspection recipes. 
     
     
         13 . The system according to  claim 8 , wherein the distance calculator is adapted to change the type of distances to be calculated based on evaluation results obtained by applying the at least one selected inspection recipe. 
     
     
         14 . The system according to  claim 8 , wherein the structural elements are coated by a coating material that changes an optical property over time, wherein the plurality of inspection recipes differ from each other by a value of optical property of the coating material they are tuned to. 
     
     
         15 . A computer program product comprising a non-transient computer readable medium that stores instructions that cause a processor to execute a method that comprises: obtaining an image of a structural element of the semiconductor device; calculating multiple types of distances between an image of the structural element and each of a plurality of reference images obtained by applying a plurality of inspection recipes; wherein the multiple types of distances are calculated by using different algorithms; wherein different inspection recipes of the plurality of inspection recipes are optimized for different conditions of the semiconductor device; wherein each reference image is associated with an inspection recipe before the obtaining of the image of the structural element; and automatically selecting at least one selected inspection recipe out of the plurality of inspection recipes based on values of the multiple types of distances. 
     
     
         16 . The computer program product according to  claim 15 , wherein the calculating of the multiple types of distances comprises calculating each type of distance out of: (i) a difference between gray level histograms, (ii) a rectilinear (L 1 ) distance, (iii) Euclidian (L 2 ) length, (iv) Chi-square distance, (v) Bhattacharyya distance, (vi) Wasserstein metric, (vii) Swain metric, and (viii) normalized correlation. 
     
     
         17 . The computer program product according to  claim 15 , further storing instructions that cause the processor to execute a method that comprises generating a match value for each reference image based on values of the multiple types of distances between the image of the structural element and the reference image; and selecting each inspection recipe associated with a reference image that has a match value within an allowable match value range. 
     
     
         18 . The computer program product according to  claim 15 , further storing instructions that cause the processor to execute a method that comprises selecting multiple selected inspection recipes. 
     
     
         19 . The computer program product according to  claim 15 , further storing instructions that cause the processor to execute a method that comprises changing the type of distances to be calculated based on evaluation results obtained by applying the at least one selected inspection recipe. 
     
     
         20 . The computer program product according to  claim 15 , wherein the structural elements are coated by a coating material that changes an optical property over time, wherein the plurality of inspection recipes differ from each other by a value of optical property of the coating material they are tuned to. 
     
     
         21 . The method according to  claim 1 , wherein the different conditions comprise a global reflection of the semiconductor device.