Detecting targeted locations in a semiconductor specimen

An examination system to detect defects in a specimen, the system comprising a processing and memory circuitry (PMC) for:

• receiving data specifying at least one targeted location of interest within at least a part of the specimen, wherein the at least one targeted location of interest differs from locations of potential defects previously revealed in accordance with a detection recipe, and wherein the detection recipe comprises a set of instructions for determining a score for each pixel of the plurality of pixels, and detecting potential defects based on the score satisfying a threshold condition;
• obtaining an image of the at least a part of the specimen, the image acquired by an examination tool;
• processing the image in run-time, in accordance with the detection recipe, to derive information related to the at least one targeted location of interest;
• and
• outputting the information related to the at least one targeted location usable for examining the at least one targeted location of interest.

A method of specimen examination, comprising:

• obtaining, by a processing and memory circuitry (PMC), an image of at least a part of a specimen, the image comprising a plurality of pixels, the image acquired by an examination tool;
• processing, by the PMC in accordance with a detection recipe, the image to derive information related to potential defects in the specimen, wherein:
• the detection recipe comprises a set of instructions for determining a score for each pixel of the plurality of pixels, and detecting potential defects based on the score satisfying a threshold condition;
• and
• said information comprises a first data set informative of first locations identified, in accordance with the detection recipe, as locations of potential defects, and a second data set informative of second locations not identified, in accordance with the detection recipe, as locations of potential defects;
• receiving, by the PMC, data specifying at least one targeted location of interest within the at least a part of the specimen;
• determining, by the PMC, that the first data set is not informative of the at least one targeted location of interest;
• in response to determining that the first data set is not informative of the at least one targeted location of interest, generating, by the PMC, a third data set by adding, to the first data set from the second data set, added information related to the at least one targeted location of interest, the added information bearing an indication that it corresponds to the at least one targeted location of interest;
• and
• outputting the third data set usable for examining the at least one targeted location of interest.