Defect classification apparatus and defect classification method
Provided is a defect classification apparatus classifying images of defects of a sample included in images obtained by capturing the sample, the apparatus including an image storage unit for storing the images of the sample acquired by an external image acquisition unit, a defect class storage unit for storing types of defects included in the images of the sample, an image processing unit for extracting images of defects from the images from the sample, processing the extracted images of defects and generating a plurality of defect images, a classifier learning unit for learning a defect classifier using the images of defects of the sample extracted by the image processing unit and data of the plurality of generated defect images, and a defect classification unit for processing the images of the sample by using the classifier learned by the classifier learning unit, to classify the images of defects of the sample.
1. A defect classification apparatus comprising:
an image storage unit for storing images of a sample;
a defect class storage unit for storing types of defects included in the images of the sample;
an image processing unit for processing the images of the sample and generating a plurality of images; and
a classifier learning unit for learning a defect classifier using the images of the sample and the plurality of images, wherein
the image processing unit performs any of a rotation process, a horizontal inversion process or a class unchangeable deformation process, which is performed while the type of a defect image is unchanged or performs a combination thereof to the images of the sample, and generates the plurality of images, and
the channel information accompanying the plurality of generated images is renewed according to the rotation process or the inversion process.
2. The defect classification apparatus according to claim 1 , wherein
the image processing unit performs any of a rotation process, an inversion process or a class unchangeable deformation process, which is performed while the type of defect image is unchanged or performs a combination thereof to the extracted images of defects, performs any of extension, contraction or an aspect change or performs a combination thereof to the extracted images of defects, and generates new images of defects from the extracted images of defects.
3. The defect classification apparatus according to claim 1 , wherein
the classifier learning unit further includes a selection unit for selecting whether the data of the plurality of generated defect images is used to learn the defect classifier.
4. A defect classification method comprising steps of:
storing images of a sample in an image storage unit;
storing types of defects included in the images of the sample in a defect class storage unit;
processing the images of the sample and generating a plurality of images; and
learning a defect classifier in a classifier learning unit by using the images of the sample and the plurality of images, wherein
an image processing unit performs any of a rotation process, a horizontal inversion process or a class unchangeable deformation process, which is performed while the type of a defect image is unchanged or performs a combination thereof to the images of the sample and generates the plurality of images, and
the channel information accompanying the plurality of generated images is renewed according to the rotation process or the inversion process.
5. The defect classification method according to claim 4 , wherein
the processing of the extracted images of defects to generate the plurality of defect images by the image processing unit is conducted by performing any of a rotation process, an inversion process or a class unchangeable deformation process, which is performed while the type of defect image is unchanged or performing a combination thereof to the extracted images of defects, and performing any of extension, contraction or an aspect change or performing a combination thereof to the extracted images of defects to generate new images of defects from the extracted images of defects.
6. The defect classification method according to claim 4 , wherein
whether the data of the plurality of generated defect images is used to learn the defect classifier by the classifier learning unit is selected by a selection unit.