Abstract: A best optical inspection mode to detect defects can be determined when no defect examples or only a limited number of defect examples are available. A signal for a defect of interest at the plurality of sites and for the plurality of modes can be determined using electromagnetic simulation. A ratio of the signal for the defect of interest to the noise at each combination of the plurality of sites and the plurality of modes can be determined. A mode with optimized signal-to-noise characteristics can be determined based on the ratios.

Abstract: Methods and systems for determining a layer on which a defect detected on a wafer is located are provided. One method includes detecting defects on a wafer by directing light to the wafer at first and second angles of incidence and determining locations of the defects on the wafer based on the output corresponding to the defects. For one of the defects detected in the output generated for one spot illuminated on the wafer with the light directed to the wafer at the first and second angles, the method includes comparing the locations of the one of the defects determined based on the output generated with the light directed to the one spot on the wafer at the first and second angles. The method further includes determining a layer of the wafer on which the one of the defects is located based on results of the comparing.

Abstract: Systems and methods for providing an augmented input data to a convolutional neural network (CNN) are disclosed. Wafer images are received at a processor. The wafer image is divided into a plurality of references images each associated with a die in the wafer image. Test images are received. A plurality of difference images are created by differences the test images with the reference images. The reference images and difference images are assembled into the augmented input data for the CNN and provided to the CNN.

Abstract: Systems and methods for detecting defects on a reticle are provided. One system includes computer subsystem(s) that include one or more image processing components that acquire images generated by an inspection subsystem for a wafer, a main user interface component that provides information generated for the wafer and the reticle to a user and receives instructions from the user, and an interface component that provides an interface between the one or more image processing components and the main user interface. Unlike currently used systems, the one or more image processing components are configured for performing repeater defect detection by applying a repeater defect detection algorithm to the images acquired by the one or more image processing components, and the repeater defect detection algorithm is configured to detect defects on the wafer using a hot threshold and to identify the defects that are repeater defects.

Abstract: Methods and systems for performing one or more functions for a specimen using output simulated for the specimen are provided. One system includes one or more computer subsystems configured for acquiring output generated for a specimen by one or more detectors included in a tool configured to perform a process on the specimen. The system also includes one or more components executed by the one or more computer subsystems. The one or more components include a learning based model configured for performing one or more first functions using the acquired output as input to thereby generate simulated output for the specimen. The one or more computer subsystems are also configured for performing one or more second functions for the specimen using the simulated output.

Abstract: Systems and methods for tool health monitoring and matching through integrated real-time data collection, event prioritization, and automated determination of matched states through image analysis are disclosed. Data from the semiconductor production tools can be received in real-time. A control limit impact (CLI) of the parametric data and the defect attributes data can be determined and causation factors can be prioritized. Image analysis techniques can compare images and can be used to judge tool matching, such as by identifying one of the states at which the two or more of the semiconductor manufacturing tools match.

Abstract: Methods and systems for training a learning based defect classifier are provided. One method includes training a learning based defect classifier with a training set of defects that includes identified defects of interest (DOIs) and identified nuisances. The DOIs and nuisances in the training set include DOIs and nuisances identified on at least one training wafer and at least one inspection wafer. The at least one training wafer is known to have an abnormally high defectivity and the at least one inspection wafer is expected to have normal defectivity.

Abstract: Methods and systems are disclosed that provide nuisance reduction in images, such as semiconductor images that include one or more metal lines. A potential defect is correlated against pixel grey level intensity charts for two perpendicular axes. A position of the potential defect relative to a pattern, such as a metal line, is determined along the two axes. The potential defect can be classified as a defect of interest or nuisance event.

Abstract: Defect detection is performed by comparing a test image and a reference image with a rendered design image, which may be generated from a design file. This may occur because a comparison of the test image and another reference image was inconclusive due to noise. The results of the two comparisons with the rendered design image can indicate whether a defect is present in the test image.

Abstract: Methods and systems for identifying a source of nuisance defects on a wafer are provided. One method includes detecting defects on a wafer by applying a hot threshold to output generated for the wafer by a detector of an inspection subsystem such that at least a majority of the detected defects include nuisance defects and determining locations of the detected defects with respect to design information for the wafer. In addition, the method includes stacking information for the detected defects based on the determined locations relative to a structure on the wafer such that the detected defects having the same locations relative to the structure are coincident with each other in results of the stacking. The method further includes identifying a source of the nuisance defects based on the results of the stacking.

Abstract: Gray level histograms for a test image and a reference image are adjusted by histogram scaling. Parameters from the histogram scaling are applied to the test image and the reference image. After the parameters are applied, the reference image and the test image are compared to produce a difference image, such as by subtracting the reference image from the test image. Noise in the difference image can be reduced, which improves defect identification in the difference image. In addition, noisy structures in the difference image which are elongated in vertical or horizontal direction can be found. If the noise exceeds a certain threshold, the structures may not be inspected.

Abstract: Systems and methods for tool health monitoring and matching through integrated real-time data collection, event prioritization, and automated determination of matched states through image analysis are disclosed. Data from the semiconductor production tools can be received in real-time. A control limit impact (CLI) of the parametric data and the defect attributes data can be determined and causation factors can be prioritized. Image analysis techniques can compare images and can be used to judge tool matching, such as by identifying one of the states at which the two or more of the semiconductor manufacturing tools match.

Abstract: Methods and systems for detecting defects on a specimen are provided. One method includes identifying first and second portions of dies on a specimen as edge dies and center dies, respectively. The method also includes determining first and second inspection methods for the first and second portions, respectively. Parameter(s) of comparisons performed in the first and second inspection methods are different. The method further includes detecting defects in at least one of the edge dies using the first inspection method and detecting defects in at least one of the center dies using the second inspection method.

Abstract: Defects from a hot scan can be saved, such as on persistent storage, random access memory, or a split database. The persistent storage can be patch-based virtual inspector virtual analyzer (VIVA) or local storage. Repeater defect detection jobs can determined and the wafer can be inspected based on the repeater defect detection jobs. Repeater defects can be analyzed and corresponding defect records to the repeater defects can be read from the persistent storage. These results may be returned to the high level defect detection controller.

Abstract: Criticality of a detected defect can be determined based on context codes. The context codes can be generated for a region, each of which may be part of a die. Noise levels can be used to group context codes. The context codes can be used to automatically classify a range of design contexts present on a die without needing certain information a priori.

Abstract: Systems and methods for classifying defects using hot scans and convolutional neural networks (CNNs) are disclosed. Primary scanning modes are identified by a processor and a hot scan of a wafer is performed. Defects of interest and nuisance data are selected and images of those areas are captured usa7ing one or more secondary scanning modes. Image sets are collected and divided into subsets. CNNs are trained using the image subsets. An ideal secondary scanning mode is determined and a final hot scan is performed. Defects are filtered and classified according to the final hot scan and the ideal secondary scanning mode. Disclosed systems for classifying defects utilize image data acquisition subsystems such as a scanning electron microscope as well as processors and electronic databases.

Abstract: Systems and methods of a two-pass inspection methodology that dynamically creates micro care areas for inspection of repeater defects. Micro care areas can be formed around each location of a repeater defect. After inspection, additional repeater defects in the micro care areas can be identified. Attributes of the repeater defects can be compared and any repeater defects with attributes that deviate from an expected group attribute distribution can be classified as nuisance.

Abstract: Systems and methods for providing an augmented input data to a convolutional neural network (CNN) are disclosed. Wafer images are received at a processor. The wafer image is divided into a plurality of references images each associated with a die in the wafer image. Test images are received. A plurality of difference images are created by differences the test images with the reference images. The reference images and difference images are assembled into the augmented input data for the CNN and provided to the CNN.

Abstract: Noise reduction in a difference image of an optical inspection tool is provided by calculating a difference image across layers of a multi-layered wafer. A first wafer image of a first wafer layer and a second wafer image of a second wafer layer are used. The first wafer image and the second wafer image are at a same planar location on the multi-layered wafer, but of different layers and/or after different process steps. A first difference image is calculated between the first wafer image and the second wafer image to reduce wafer noise. Defects can be identified using the first difference image. A system with an image data acquisition subsystem can be used to perform this technique.

Abstract: Systems and methods for detecting defects on a reticle are provided. One system includes computer subsystem(s) that include one or more image processing components that acquire images generated by an inspection subsystem for a wafer, a main user interface component that provides information generated for the wafer and the reticle to a user and receives instructions from the user, and an interface component that provides an interface between the one or more image processing components and the main user interface. Unlike currently used systems, the one or more image processing components are configured for performing repeater defect detection by applying a repeater defect detection algorithm to the images acquired by the one or more image processing components, and the repeater defect detection algorithm is configured to detect defects on the wafer using a hot threshold and to identify the defects that are repeater defects.