A method for determining a measurement recipe for measuring a parameter of interest from a compound structure on a substrate. The method includes obtaining first training data relating to measurements of reference targets, the targets including: parameter of interest targets, each parameter of interest target having an induced set value which is varied over the parameter of interest targets; and one or more isolated feature targets, each including repetitions of one or more features. Second training data is obtained, the second training data including compound structure measurement signals obtained from measurement of one or more instances of the compound structure. One or more machine learning models are trained using the first training data and second training data to infer a value for the parameter of interest from a measurement signal related to the compound structure corrected for a feature asymmetry contribution.

A substrate including a target. The target including a plurality of sub-targets, the plurality of sub-targets including at least a first sub-target and second sub-target, each of the plurality of sub-targets including at least one subsegmented periodic structure having repetitions of a first region and a second region, wherein at least one of the first regions or second regions comprise subsegmented regions formed of periodic sub-features. The first sub-target includes subsegmentation characteristics for its subsegmented regions and the second sub-target comprises second subsegmentation characteristics for its subsegmented regions, the first subsegmentation characteristics and second subsegmentation characteristics being different in terms of at least one subsegmentation parameter.

In some implementations, a metrology optimization system may obtain three-dimensional (3D) model information associated with an object. The metrology optimization system may obtain optical sensor information associated with a metrology system that is to measure the object. The metrology optimization system may determine, based on the 3D model information and the optical sensor information, an initial configuration for the metrology system. The metrology optimization system may determine, based on the 3D model information, the optical sensor information, and the initial configuration, an optimal configuration for the metrology system. The metrology optimization system may provide the optimal configuration for the metrology system.

Methods and systems for training a machine learning based measurement model conditioned by at least one regularization control parameter are described herein. A ML based measurement model conditioned by at least one regularization control parameter is trained for different control parameter values. A regularization control value provided as input to the trained ML based measurement model defines the regularization condition at inference. In a further aspect, an optimal value of a regularization control parameter is selected based on measurement performance on a set of measurement data. As measurement conditions change, the optimal value is reevaluated based on measurement performance on an updated set of measurement data that reflects the changing measurement conditions. In another further aspect, changes in measurement conditions and reevaluation of a regularization control value are performed automatically as measurement data is collected by a measurement system without interruption of the measurement process.

A method for multi-merit adaptive sampling may include receiving an initial sampling map and a first set of metrology data. The method may further include calculating figure of merit metrics. The method may include ranking each initial sampling point in the initial sampling map based on the figure of merit metrics and levels of process variation. The method may include generating an adjusted sampling plan for at least one of a second sample in the first lot or a future process layer of the first sample based on the rank of each initial sampling point. An adjusted set of sampling points may be generated by increasing a number of sampling points in a first sample region associated with a first level of process variation and decreasing a number of sampling points in a second sample region associated with a second level of process variation.

A measuring method for measuring distortion of a substrate includes capturing images of a plurality of partial regions of a pattern formed in a device region of the substrate, measuring a position of the pattern based on the images acquired by the image capturing, and processing of determining a period of the pattern based on the images acquired by the image capturing, and determining distortion in the region of the substrate where the pattern is formed, based on the determined period and a measurement result acquired in the measuring.

A method for analyzing a wafer includes acquiring first measurement data for a first wafer and second measurement data for a second wafer, generating normalization data including first normalization data and second normalization data obtained by scaling the first measurement data and the second measurement data, respectively, separating each of the first normalization data and the second normalization data into at least one component to generate component data including first component data and second component data, and outputting a similarity of the first wafer and the second wafer calculated based on the component data.

A measuring device includes a light source that irradiates a measurement spot on a wafer formed with memory holes and slits with a multi-wavelength light, a first imaging unit that acquires a first pupil plane intensity distribution image of reflected light from the measurement spot, a second imaging unit that acquires a second pupil plane intensity distribution image of the reflected light, and a detection unit that analyzes the second pupil plane intensity distribution image to measure overlay. The measuring device includes an overlay analysis unit that acquires the first and second pupil plane intensity distribution images while moving a position of the measurement spot and selects a measurement spot not including the slit based on the first pupil plane intensity distribution image, and uses the overlay obtained by analyzing the second pupil plane intensity distribution image of the selected measurement spot as the overlay of the memory hole and a slit.

An image reading unit reads image data obtained by imaging a sample in which a plurality of regions is disposed including a pattern formed based on identical design information. An evaluation value acquisition unit acquires, from among sampling images at a plurality of sampling points set for each of the plurality of regions, an evaluation value based on luminance of pixels for a plurality of sampling images at the same sampling point of the plurality of regions. A reference evaluation value acquisition unit acquires a reference evaluation value relative to the evaluation value for a plurality of sampling images at the same sampling point in the plurality of regions. An equalized evaluation value acquisition unit acquires an equalized evaluation value indicating a degree of deviation of the evaluation value from the reference evaluation value, for each of the sampling images of the plurality of regions.

A swath imaging system may include an illumination source configured to illuminate a sample with an illumination beam, a stage to scan the sample with a scan pattern including swaths extending along a scan direction when implementing the inspection recipe, one or more TDI sensors configured to capture swath images of the sample, and a controller. The plurality of swaths may be distributed along a step direction orthogonal to the scan direction, and at least some of the plurality of swath images overlap along the step direction. The controller may implement the inspection recipe by receiving the plurality of swath images, combining the plurality of swath images into a uniformized image where overlapping portions of the plurality of swath images are combined within the uniformized image, and generating one or more measurements of the sample based on the uniformized image.