Image enhancement for defect inspection based on machine learning

• Main Authors: RESHEF, BOSER, LEVANT BORIS, JU RMINEN, TOMER, SHERMAN BORIS, YACOBI, RAN
• Format: Patent
• Language: Chinese; English
• Published: 19.03.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

A system and method for defect inspection of a semiconductor sample is provided. The method includes: obtaining an original image of a semiconductor sample, the original image having a first region annotated as surrounding a defect feature; specifying a second region in an original image including a first region, thereby generating a background region between the first region and the second region; in the target image of the semiconductor sample, identifying a candidate region set matched with the background region according to the matching measurement; selecting one or more candidate regions from the candidate region set; and pasting the first region or a portion of the first region with respect to the one or more candidate regions, thereby producing an enhanced target image usable for defect inspection on the semiconductor sample. 提供了一种用于对半导体样品进行缺陷检查的系统和方法。该方法包括：获得半导体样品的原始图像，该原始图像具有被注释为包围缺陷特征的第一区域；在包含第一区域的原始图像中指定第二区域，从而在所述第一区域与所述第二区域之间产生背景区域；在半导体样品的目标图像中，根据匹配测量来识别与背景区域匹配的候选区域集合；从该候选区域集合中选择一个或多个候选区域；以及相对于一个或