US 12,353,970 B2
Finding semiconductor defects using convolutional context attributes
Abdurrahman Sezginer, Monte Sereno, CA (US); Gordon Rouse, Dublin, CA (US); and Manikandan Mariyappan, San Jose, CA (US)
Assigned to KLA CORPORATION, Milpitas, CA (US)
Filed by KLA Corporation, Milpitas, CA (US)
Filed on Nov. 13, 2020, as Appl. No. 17/098,256.
Claims priority of provisional application 62/939,534, filed on Nov. 22, 2019.
Prior Publication US 2021/0158223 A1, May 27, 2021
Int. Cl. G06N 20/00 (2019.01); G01N 21/95 (2006.01); G01N 21/956 (2006.01); G06N 20/10 (2019.01)
CPC G06N 20/10 (2019.01) [G01N 21/9501 (2013.01); G01N 21/95607 (2013.01); G01N 21/95623 (2013.01); G01N 2021/95615 (2013.01)] 26 Claims
OG exemplary drawing
 
1. A method, comprising:
calculating context attributes for optical imaging of a patterned layer of a semiconductor die, comprising calculating convolutions of a pattern of the patterned layer with respective kernels of a plurality of kernels, wherein the plurality of kernels is orthogonal; and
finding defects on the semiconductor die in accordance with the context attributes, comprising:
optically imaging the semiconductor die to generate a target image;
comparing the target image of the semiconductor die to a reference image of the semiconductor die, to generate a difference image of the semiconductor die;
adjusting a defect-detection filter for different portions of the semiconductor die based at least in part on the context attributes;
detecting defects in the difference image using the defect-detection filter; and
storing a list of the detected defects.