Abstract: A machine learning system and a machine learning method capable of selecting a pretrained model to be used in transfer learning in a short time without actually executing the transfer learning includes a pretrained model acquisition unit which acquires a pretrained model from a pretrained model storage unit storing a plurality of pretrained models obtained by learning a transfer source task under respective conditions; a transfer learning dataset storage unit configured to store dataset related to a transfer target task; a pretrained model adaptability evaluation unit configured to evaluate adaptability of each pretrained model acquired by the pretrained model acquisition unit to the dataset related to the transfer target task; and a transfer learning unit configured to execute, based on an evaluation result of the pretrained model adaptability evaluation unit, transfer learning using a selected pretrained model and the dataset, and outputs a learning result as a trained model.

Abstract: The purpose of the present invention is to provide a pattern matching device and computer program that carry out highly accurate positioning even if edge positions and numbers change. The present invention proposes a computer program and a pattern matching device wherein a plurality of edges included in first pattern data to be matched and a plurality of edges included in second pattern data to be matched with the first pattern data are associated, a plurality of different association combinations are prepared, the plurality of association combinations are evaluated using index values for the plurality of edges, and matching processing is carried out using the association combinations selected through the evaluation.

Abstract: The preset invention aims to provide a method of presenting a learning condition which enables an improvement in the accuracy of image analysis. An information processing apparatus for machine learning is provided which includes a true/false information generating unit which generates true/false information of an image analysis result, a reliability determining unit which determines reliability related to analysis in image analysis processing, and a learning condition output unit which presents a learning condition, based on the true/false information and the reliability.

Abstract: A machine learning system performs transfer learning to output a trained model by performing training using a parameter of a pre-trained model by using a given dataset and a given pre-trained model. The machine learning system includes a dataset storage unit that stores one or more datasets, and a first training unit that performs training using a dataset stored in the dataset storage unit to generate the pre-trained model, and stores the generated pre-trained model in a pre-trained model database. The dataset storage unit stores tag information including any one or more of domain information indicating a target object of data included in a dataset to be stored, class information indicating a class included in data, and data acquisition condition information related to an acquisition condition of data and a dataset in a manner that the tag information and the dataset are associated with each other.

Abstract: It is an object of the present invention to provide a semiconductor inspection apparatus capable of well carrying out position alignment and correctly determining whether the position alignment has been carried out successfully or has ended in a failure without operator interventions even if an inspected image is an image having few characteristics as is the case with a repetitive pattern or the inspected image is an image having a complicated shape.

Abstract: The purpose of the present invention is to provide a pattern matching device and computer program that carry out highly accurate positioning even if edge positions and numbers change. The present invention proposes a computer program and a pattern matching device wherein a plurality of edges included in first pattern data to be matched and a plurality of edges included in second pattern data to be matched with the first pattern data are associated, a plurality of different association combinations are prepared, the plurality of association combinations are evaluated using index values for the plurality of edges, and matching processing is carried out using the association combinations selected through the evaluation.

Abstract: Image processing apparatus includes: interpolation process image acquisition means for acquiring an interpolation process image of prescribed size which includes an interpolation point of an inputted image; Fourier transform means for subjecting the interpolation process image which is acquired with the interpolation process image acquisition means to Fourier transform; phase change means for changing, a phase of each value of the transformed interpolation process image which has been subjected to Fourier transform by the Fourier transform means, such that the interpolation point migrates to a desired nearby integer coordinate position; inverse Fourier transform means for subjecting the interpolation process image whose phase has been changed by the phase change means, to inverse Fourier transform; interpolation value determination means for adopting an interpolation point, a value of a pixel at the integer coordinate position, from the transformed interpolation process image subjected to inverse Fourier tran

Abstract: An evaluation value indicative of the extent of lines in each direction is calculated for a pre-processed image in which 0s are filled in and extended in the lateral direction of the inputted image and which has been reduced ?th in the longitudinal direction. To obtain the angle of rotation of an image from the change in the evaluation value obtained while the angle relative to the lateral direction of the pre-processed image is modified in small steps, a parallel line is drawn for each direction, a projection is taken, and the sum of squares serves as the evaluation value of the direction. The direction having the highest evaluation value serves as the obtained direction of rotation from the normal position. The projection of each direction references the point of intersection between the parallel line drawn for each direction and the coordinate line of the horizontal axis.

Abstract: An inspection apparatus and method are provided, wherein even when an image that cannot be processed by a current image processing algorithm is input to an image processing unit while a working line is in operation, the inspection can be continued by newly generating an image processing algorithm optimized in keeping with a particular image. The apparatus includes an erroneous recognition detector, a teacher data generator and a switching unit for switching the current image processing algorithm to a new image processing algorithm generated based on an updated teacher data group. As a result, the inspection can be continued without extremely decreasing the accuracy even when an unexpected image is input to the working line.

Abstract: In order to provide a computer program, an image processing device, and a pattern matching method that perform pattern matching at a high level of accuracy without relying on edge deformation, contrast fluctuations, etc., in one embodiment, the disclosed pattern matching method and device perform pattern matching over an image using a template produced on the basis of the below mentioned design data. The pattern matching method and device determine the characteristic quantities of the image for an inner region and/or an outer region that are divided by a line that defines the contour of a pattern, and determine positions at which said characteristic quantities satisfy predetermined conditions to be matching positions, matching position candidates, or erroneous matching positions.

Abstract: The present invention is a template matching processing device capable of evaluating a similarity degree which supports even a case of intensive morphological change between a design image and a photographic image. In the template matching processing device, matching processing between the design image and the photographic image is performed, a partial design image is obtained by clipping a portion having the highest correlation (step 101), and processing for deforming the photographic image in accordance with the clipped design image (steps 102 to 105) is performed, so that correlation between the deformed image obtained and the design image is taken to be set as the similarity degree.

Abstract: A pattern matching method for a scanning electron microscope comprises a step of performing pattern matching of only an upper layer pattern between an image (101) in which a pattern consisting of plural layers is represented and a template (104) in which the upper layer pattern of the plural layer pattern is selectively represented, thereby identifying the position of the pattern consisting of the plural layers. Then, information about the upper layer pattern is subtracted from the image (101), thus extracting shape information (108) about the lower layer pattern. Consequently, stable positioning or selective information extraction on a certain layer is enabled regardless of the state of the depths of a pattern formed in three dimensions or of the charge state of a sample.

Abstract: Image processing apparatus includes: interpolation process image acquisition means for acquiring an interpolation process image of prescribed size which includes an interpolation point of an inputted image; Fourier transform means for subjecting the interpolation process image which is acquired with the interpolation process image acquisition means to Fourier transform; phase change means for changing, a phase of each value of the transformed interpolation process image which has been subjected to Fourier transform by the Fourier transform means, such that the interpolation point migrates to a desired nearby integer coordinate position; inverse Fourier transform means for subjecting the interpolation process image whose phase has been changed by the phase change means, to inverse Fourier transform; interpolation value determination means for adopting an interpolation point, a value of a pixel at the integer coordinate position, from the transformed interpolation process image subjected to inverse Fourier tran

Abstract: Provided is a template matching method and a template matching apparatus, where the degree of matching between a template and the actual image upon template matching is maintained at a high level, without depending on a partial appearance of a lower layer. Proposed as one embodiment, is a method and an apparatus for template matching, where either an area is set in which comparison of the template and the image is not conducted, or a second area is set inside the template where comparison different from comparison conducted in a first comparison area is to be conducted, and the template matching is conducted on the basis either of comparison excluding the non-comparison area, or of comparison using the first and second areas.

Abstract: An object of the present invention is to provide a pattern measuring apparatus which performs high-accuracy concavity/convexity determination (e.g., distinguishing between a line segment and space) while simultaneously reducing the dose of a beam falling onto a pattern to be measured. To attain the object, this invention proposes a pattern measuring apparatus which specifies a pattern in a measurement object area by scanning a tilted bean with respect to another area different from the measurement object area and then performs measurement based on the pattern-specifying result. With such arrangement, it becomes possible to perform measurement without the risk of wrong pattern designation while lowering the dose of a beam hitting the measurement object area.

Abstract: An object of the present invention is to provide a pattern measuring apparatus which performs high-accuracy concavity/convexity determination (e.g., distinguishing between a line segment and space) while simultaneously reducing the dose of a beam falling onto a pattern to be measured. To attain the object, this invention proposes a pattern measuring apparatus which specifies a pattern in a measurement object area by scanning a tilted bean with respect to another area different from the measurement object area and then performs measurement based on the pattern-specifying result. With such arrangement, it becomes possible to perform measurement without the risk of wrong pattern designation while lowering the dose of a beam hitting the measurement object area.

Abstract: An object of the present invention is to provide a pattern measuring apparatus which performs high-accuracy concavity/convexity determination (e.g., distinguishing between a line segment and space) while simultaneously reducing the dose of a beam falling onto a pattern to be measured. To attain the object, this invention proposes a pattern measuring apparatus which specifies a pattern in a measurement object area by scanning a tilted bean with respect to another area different from the measurement object area and then performs measurement based on the pattern-specifying result. With such arrangement, it becomes possible to perform measurement without the risk of wrong pattern designation while lowering the dose of a beam hitting the measurement object area.

Abstract: It is an object of the present invention to provide a semiconductor inspection apparatus capable of well carrying out position alignment and correctly determining whether the position alignment has been carried out successfully or has ended in a failure without operator interventions even if an inspected image is an image having few characteristics as is the case with a repetitive pattern or the inspected image is an image having a complicated shape.

Abstract: An evaluation value indicative of the extent of lines in each direction is calculated for a pre-processed image in which 0s are filled in and extended in the lateral direction of the inputted image and which has been reduced ?th in the longitudinal direction. To obtain the angle of rotation of an image from the change in the evaluation value obtained while the angle relative to the lateral direction of the pre-processed image is modified in small steps, a parallel line is drawn for each direction, a projection is taken, and the sum of squares serves as the evaluation value of the direction. The direction having the highest evaluation value serves as the obtained direction of rotation from the normal position. The projection of each direction references the point of intersection between the parallel line drawn for each direction and the coordinate line of the horizontal axis.

Abstract: An object of the present invention is to provide a pattern measuring apparatus which performs high-accuracy concavity/convexity determination (e.g., distinguishing between a line segment and space) while simultaneously reducing the dose of a beam falling onto a pattern to be measured. To attain the object, this invention proposes a pattern measuring apparatus which specifies a pattern in a measurement object area by scanning a tilted bean with respect to another area different from the measurement object area and then performs measurement based on the pattern-specifying result. With such arrangement, it becomes possible to perform measurement without the risk of wrong pattern designation while lowering the dose of a beam hitting the measurement object area.