In accordance with some embodiments of the present disclosure, an inspection method of a photomask includes performing a first inspection process, unloading the photomask from the inspection system, and performing a second inspection process. In the first inspection process, a common Z calibration map of an objective lens of an optical module with respect to the photomask is generated and stored, and a first image of the photomask is captured by using an image sensor while focusing the objective lens of the optical module based on the common Z calibration map. The photomask is unloaded from the inspection system. In the second inspection process, the photomask is loaded on the inspection system and a second image of the photomask is captured by using an image sensor while focusing an objective lens of an optical module based on the common Z calibration map generated in the first inspection process.

Methods and apparatuses are disclosed herein for parameter estimation for metrology. An example method at least includes optimizing, using a parameter estimation network, a parameter set to fit a feature in an image based on one or more models of the feature, the parameter set defining the one or more models, and providing metrology data of the feature in the image based on the optimized parameter set.

A semiconductor structure and a method for managing semiconductor wafer stress are disclosed. The semiconductor structure includes a semiconductor wafer, a first stress layer disposed on and in contact with a backside of the semiconductor wafer, and a second stress layer on and in contact with the first stress layer. The first stress layer exerts a first stress on the semiconductor wafer and the second layer exerts a second stress on the semiconductor wafer that is opposite the first backside stress. The method includes forming a first stress layer on and in contact with a backside of a semiconductor wafer, and further forming a second stress layer on and in contact with the first stress layer. The first stress layer exerts a first stress on the semiconductor wafer and the second stress layer exerts a second stress on the semiconductor wafer that is opposite to the first stress.

A defect observation method includes, as steps executed by a computer system, a first step of acquiring, as a bevel image, an image captured using defect candidate coordinates in a bevel portion as an imaging position by using a microscope or an imaging apparatus; and a second step of detecting a defect in the bevel image. The second step includes a step of determining whether there is at least one portion among a wafer edge, a wafer notch, and an orientation flat in the bevel image, a step of switching and selectively applying a defect detection scheme of detecting the defect from the bevel image from a plurality of schemes which are candidates based on a determination result, and a step of executing a process of detecting the defect from the bevel image in conformity with the switched scheme.

An image-based acceptance learning device (2) learns a result of determination as to whether a planar object (1) is acceptable or defective based on at least one of a three-dimensional shape or a color on a surface of the planar object (1). The device (2) includes a surface image receiver (3) to receive an inputted two-dimensional data being image data of the surface of the planar object (1), a determination information receiver (4) to receive an inputted determination information indicating the result of determination as to whether the planar object (1) corresponding to the two-dimensional data is acceptable or defective, and a learner (5) to learn, based on the two-dimensional data and the determination information, a relevant area (1R) including the three-dimensional shape or the color on the surface in the two-dimensional data. The relevant area is a basis for the determination information.

An image-based acceptance learning device (2) learns a result of determination as to whether a planar object (1) is acceptable or defective based on at least one of a three-dimensional shape or a color on a surface of the planar object (1). The device (2) includes a surface image receiver (3) to receive an inputted two-dimensional data being image data of the surface of the planar object (1), a determination information receiver (4) to receive an inputted determination information indicating the result of determination as to whether the planar object (1) corresponding to the two-dimensional data is acceptable or defective, and a learner (5) to learn, based on the two-dimensional data and the determination information, a relevant area (1R) including the three-dimensional shape or the color on the surface in the two-dimensional data. The relevant area is a basis for the determination information.

A method for inspecting a surface of a wafer, includes steps of: irradiating a surface of the wafer with a laser beam having three or more distinct wavelengths; detecting a reflected light from the surface of the wafer when the surface of the wafer is irradiated with the laser beam; and determining whether a foreign matter exists on the surface of the wafer based on reflectances of the surface of the wafer with respect to the laser beam having the three or more distinct wavelengths, wherein the step of determining whether the foreign matter exists includes a step of determining whether the foreign matter is a metal or a non-metal.

An optical system configured to measure a raised or receded surface feature on a surface of a sample may comprise a broadband light source; a tunable filter configured to filter broadband light emitted from the broadband light source and to generate a first light beam at a selected wavelength; a linewidth control element configured to receive the first light beam and to generate a second light beam having a predefined linewidth and a predetermined coherence length; collimating optics optically coupled to the second light beam and configured to collimate the second light beam; collinearizing optics optically coupled to the collimating optics and configured to align the collimated second light beam onto the raised or receded surface feature of the sample, and a processor system and at least one digital imager configured to measure a height of the raised surface or depth of the receded surface from light reflected at least from those surfaces.

An optical system configured to measure a raised or receded surface feature on a surface of a sample may comprise a broadband light source; a tunable filter configured to filter broadband light emitted from the broadband light source and to generate a first light beam at a selected wavelength; a linewidth control element configured to receive the first light beam and to generate a second light beam having a predefined linewidth and a predetermined coherence length; collimating optics optically coupled to the second light beam and configured to collimate the second light beam; collinearizing optics optically coupled to the collimating optics and configured to align the collimated second light beam onto the raised or receded surface feature of the sample, and a processor system and at least one digital imager configured to measure a height of the raised surface or depth of the receded surface from light reflected at least from those surfaces.

A system and methods for OCD metrology are provided including receiving multiple first sets of scatterometric data, dividing each set into k sub-vectors, and training, in a self-supervised manner, k2 auto-encoder neural networks that map each of the k sub-vectors to each other. Subsequently multiple respective sets of reference parameters and multiple corresponding second sets of scatterometric data are received and a transfer neural network (NN) is trained. Initial layers include a parallel arrangement of the k2 encoder neural networks. Target output of the transfer NN training is set to the multiple sets of reference parameters and feature input is set to the multiple corresponding second sets of scatterometric data, such that the transfer NN is trained to estimate new wafer pattern parameters from subsequently measured sets of scatterometric data.