A defective part recognition device includes a microscope for obtaining a magnified image of a unit area for recognizing a defective part on the surface of a multi-layer film substrate; a spectral camera having an imaging surface where the magnified image is formed; and an information processing part for processing the spectrum information from the spectral camera. The information processing part includes a machine learning part for a clustering process on the spectrum information for each pixel, and a defect recognition part for recognizing a defective part from the result of the machine learning part. The machine learning part sets a cluster in the unit area and generates a histogram with a frequency, the number of pixels clustered into the cluster. The defect recognition part compares the frequency distribution of the generated histogram with that of a histogram free of defects and recognizes a defective part.

The image inspection apparatus includes a camera that images a workpiece placed on a stage and generates a workpiece image, and a second camera having an imaging field-of-view wider than that of the camera, which images the workpiece and generates a bird's eye view image. The apparatus detects a position of the workpiece based on the bird's eye view image, and positions the workpiece based on the position of the workpiece so that the workpiece is located in or near an imaging field-of-view of the camera, thereby imaging the workpiece with the camera to generate the workpiece image. The apparatus specifies a detailed position and an orientation of the workpiece in the workpiece image generated by the camera, and determines an inspection point of the workpiece in the workpiece image based on the detailed position and the orientation of the workpiece specified, thereby executing a predetermined inspection process.

An information processing apparatus includes a reception unit configured to receive an input specifying a position of a detection target included in an image, an acquisition unit configured to acquire a storage amount of training data including a pair of information indicating the image and information indicating the position specified by the input, a training unit configured to train a training model to detect the detection target from the image based on the stored training data, and a display control unit configured to control a display unit to display the storage amount, and a reference amount of the training data set as an amount of the training data necessary for the training unit to train the training model, in a comparable way.

A defect characterization method includes: a first scanning image and target defect coordinates in the first scanning image are obtained; a first defect image is obtained according to the target defect coordinates in the first scanning image, the first defect image containing a defect area where a target defect is located and a noise area not containing the target defect; the noise area is marked, Automatic Defect Review (ADR) calculation is performed on the defect area, and a pixel level value of a defect in the defect area is obtained; coordinates of the defect with a maximum pixel level value are obtained, and a second defect image is obtained according to the coordinates of the defect with the maximum pixel level value; and the defect with the maximum pixel level value is classified according to the second defect image.

A detection device, a detection apparatus, and a detection method are provided. The detection device includes a detection assembly and a driving assembly. At least two rows of detection units are provided, and the detection units are arranged in a matrix along the first direction and the second direction. During detection, the driving assembly may drive one or both of the to-be-detected object and the detection assembly to move along at least one of the first direction and the second direction, so that the detection units scan different areas of the surface of the to-be-detected object, thereby realizing a scanning of the entire surface. That is, the scanning of the surface of the to-be-detected object is completed by using multiple detection units, which improves detection efficiency for the to-be-detected surface, and realizes a relatively high imaging quality.

There is provided a surface abnormality detection device, and a system, capable of detecting an abnormal portion having a displacement below the distance measurement accuracy when detecting the abnormal portion on the surface of a structure. A surface abnormality detection device includes a classification means for classifying an object under measurement into one or more clusters having the same structure, based on position information at a plurality of points on a surface of the object under measurement; a determination means for determining a reflection brightness normal value of the cluster based on a distribution of reflection brightness values at a plurality of points on a surface of the cluster; and an identification means for identifying an abnormal portion on the surface of the cluster based on a difference between the reflection brightness normal value and the reflection brightness value at each of the plurality of points.

The objective of the present invention is provide a defect inspection apparatus that increases defect position precision and can easily align a coordinate origin offset between a reviewing apparatus and the defect inspection apparatus, even when design data cannot be obtained or it is difficult to sufficiently use the design data. The defect inspection apparatus according to the present invention acquires a wafer swath image necessary for inspection, and uses the swath image to detect defects and calculate a positional deviation amount. During the calculation of the positional deviation amount, a template pattern is acquired from one arbitrary swath image via an image processing unit, and the template pattern and a plurality of swath images of the entire wafer are compared, whereby the positional deviation amount for a position corresponding to the template pattern on the wafer is calculated. For positions at which the template pattern is not present, an interpolated positional deviation amount is calculated by executing an interpolation operation by using the calculated positional deviation amount. A defect position is corrected on the basis of the positional deviation amount and the interpolated positional deviation amount, or by using a positional deviation map in which these positional deviation amounts have been mapped on the entire wafer.

The present disclosure discloses a method for detecting a wafer backside defect, comprising: Step 1, providing a signal database comprising signal data corresponding to various different defects, the defects comprising convex defects and concave defects, the signal data reflecting 3D information of the corresponding defect; Step 2, performing backside scanning on a tested wafer by using oblique incident light, and collecting corresponding emitted and scattered light data; and Step 3, comparing the collected emitted and scattered light data with the signal data, and fitting a defect 3D distribution map of the backside of the tested wafer. The present disclosure can test the height or depth of a wafer backside defect and form a 3D distribution map of the wafer backside defect, which is beneficial for analyzing the source of the wafer backside defect and processing it in time, reducing the troubleshooting time and improving the product yield.

The invention relates to a method for providing a coating composition-related prediction program, the method comprising: providing a database (204, 904) comprising associations of qualitative and/or quantitative characterizations of coating surfaces and one or more parameters; training a machine learning model for providing a predictive model (M2, M3) having learned to correlate qualitative and/or quantitative characterizations of one or more coating surfaces with one or more of the parameters; and providing a composition-quality-prediction program configured for using the predictive model (M2) for predicting the properties of a coating surface to be produced from one or more input parameters; and/or providing a composition-specification-prediction program configured for using the predictive model (M3) for predicting, based on an input specifying at least a desired coating surface characterization, one or more output parameters related to a coating composition predicted to generate a coating surface having the input surface characterizations.

A method for qualitative and/or quantitative characterization of a coating surface is provided, comprising: providing a program recognizing coating surface defect types; determining, by the program, whether a camera(s) coupled to the program is within a predefined distance range and/or within a predefined image acquisition angle range relative to a currently presented coating surface; depending on the determination: generating a feedback signal indicative of whether adjustment of the position of the camera(s) is within predefined distance range and/or within the predefined image acquisition angle range, and/or automatically adjusting the relative distance of the camera and and/or automatically adjusting the angle of the camera; enabling the camera to acquire an image of the coating surface only when the camera(s) is/are within the predefined distance range and/or image acquisition angle range; processing the digital image for recognizing coating surface defects; and outputting a characterization of the coating surface.