Disclosed are a method of forming an optical proximity correction (OPC) model and/or a method of fabricating a semiconductor device using the same. The method of forming the OPC model may include obtaining a scanning electron microscope (SEM) image, which is an average image of a plurality of images taken using one or more scanning electron microscopes, and a graphic data system (GDS) image, which is obtained by imaging a designed layout, aligning the SEM image and the GDS image, performing an image filtering process on the SEM image, extracting a contour from the SEM image, and verifying the contour. The verifying of the contour may be performed using a genetic algorithm. Variables in the genetic algorithm may include first parameters related to the image alignment process, second parameters related to the image filtering process, and third parameters related to a critical dimension (CD) measurement process.

A fluorescence microscopy inspection system includes light sources able to emit light that causes a specimen to fluoresce and light that does not cause a specimen to fluoresce. The emitted light is directed through one or more filters and objective channels towards a specimen. A ring of lights projects light at the specimen at an oblique angle through a darkfield channel. One of the filters may modify the light to match a predetermined bandgap energy associated with the specimen and another filter may filter wavelengths of light reflected from the specimen and to a camera. The camera may produce an image from the received light and specimen classification and feature analysis may be performed on the image.

A damage evaluation device, a method, and a program that can automatically evaluate a damage of an outer layer of a structure occurring with respect to construction of the structure are provided. In a damage evaluation device of a structure including a processor, the processor is configured to perform image acquisition processing of acquiring a captured image of the structure, perform damage detection processing of detecting damages (cracks) of the structure based on the acquired image, perform feature region detection processing of detecting a structure feature region (a region of a P cone mark) related to construction of the structure based on the acquired image, perform selection processing of selecting a specific damage (settlement crack) related to the detected structure feature region among the detected damages, and perform information output processing of outputting information about the selected specific damage. By outputting the information about the specific damage, the damage of the outer layer of the structure occurring with respect to the construction of the structure can be automatically evaluated, and application to validity verification of a construction method and improvement of the construction method can be made.

A device for analyzing a grain sample including a light source, an image sensor, and a controller. The light source is configured for illuminating the grain sample. The image sensor is configured for capturing images of the grain sample. The controller is coupled to the image sensor and is configured for receiving the images of the grain sample therefrom and for analyzing the images to detect at least one material other than grain in the grain sample. The light source is configured for illuminating the grain sample with a local light spot having a size that is smaller than a width of an average wheat kernel. The image analysis and the detection of material other than grain may, at least partly, be performed using trained neural networks and other artificial intelligence algorithms.

The coating quality prediction device includes: a learned model that has learned a relationship between characteristics of a paint, conditions at a time of applying the paint, and a smoothness of a surface of a coating film obtained by applying the paint under the conditions; and a calculation unit that uses the learned model to calculate the smoothness of the surface of the coating film from the characteristics of the paint and the conditions at the time of applying the paint.

An artificial neural network-based method for detecting a surface type of an object includes: receiving a plurality of object images, wherein a plurality of spectra of the plurality of object images are different from one another and each of the object images has one of the spectra; transforming each object image into a matrix, wherein the matrix has a channel value that represents the spectrum of the corresponding object image; and executing a deep learning program by using the matrices to build a predictive model for identifying a target surface type of the object. Accordingly, the speed of identifying the target surface type of the object is increased, further improving the product yield of the object.

An inspection apparatus (100) detects an inspection object (90) from first image data (10) in which the inspection object (90) is included. The inspection apparatus (100) generates second image data (20) by performing a geometric transform on the first image data (10) in such a way that a view of the detected inspection object (90) becomes a view satisfying a predetermined reference. In an inference phase, the inspection apparatus (100) detects, by using an identification model for detecting an abnormality of the inspection object (90), an abnormality of the inspection object (90) included in the second image data (20). Further, in a learning phase, the inspection apparatus (100) learns, by using the second image data (20), an identification model for detecting an abnormality of the inspection object (90).

An automated method of inspecting a pipe includes: positioning the pipe with respect to a laser scanner using a positioning apparatus; scanning a size of the positioned pipe by the laser scanner; identifying a specification and historical data of the pipe's type by inputting the scanned size to an artificially intelligent module trained through machine learning to match input size data to standardized pipe types and output corresponding specifications and historical data of the pipe types; scanning dimensions of the positioned pipe by the laser scanner using a dimension portion of the identified historical data; comparing the scanned dimensions with standard dimensions from the identified specification; detecting a dimension nonconformity when the scanned dimensions are not within acceptable tolerances of the standard dimensions; and in response to detecting the dimension nonconformity, generating an alert and updating the dimension portion of the identified historical data to reflect the detected dimension nonconformity.

An artificial neural network-based method for selecting a surface type of an object includes receiving at least one object image, performing surface type identification on each of the at least one object image by using a first predictive model to categorize the object image to one of a first normal group and a first abnormal group, and performing surface type identification on each output image in the first normal group by using a second predictive model to categorize the output image to one of a second normal group and a second abnormal group.

A defect inspection system provides an image of a surface of a hard drive media to a machine learning model that is trained to identify predefined classifications of abnormal surface patterns on the hard drive media, each of the predefined classifications being associated in system memory with a severity indicator. The defect inspection model analyzes the image and generates and output indicating that the image includes a pattern consistent with a select classification of the predefined classifications of abnormal surface patterns. When the severity indicator for the select classification satisfies a failure condition, the defect inspection system automatically implements a corrective action.