Substrate inspecting unit and substrate treating apparatus including the same

A unit for inspecting a substrate comprising:

• a feature extracting module for extracting a feature from training data included in each class in response to a plurality of training data related to image data of a substrate being classified according to a predefined class;
• a validity evaluating module for evaluating validity of the feature;
• a class verifying module for verifying the predefined class;
• and
• a data reconstructing module for reconstructing the plurality of training data based on a feature determined as valid and a verified class,
• wherein reconstructed training data is utilized when inspecting the substrate,
• wherein the feature extracting module extracts the features using transfer learning of a pre-trained model that is a result obtained by training using data related to the image data of the substrate equal to or greater than a reference amount,
• wherein a validity evaluating module evaluates validity of the feature by applying a dimension reduction method to the feature,
• wherein a class verifying module verifies the predefined class using DBSCAN based on unsupervised learning,
• wherein the plurality of training data is image data of the substrate, onto which a substrate treating solution is discharged, and is image data of at least one type of image data in a good state and image data in a defective state.

An apparatus for treating a substrate comprising:

• a process treating unit for supporting the substrate while the substrate is treated;
• an inkjet head unit for discharging a substrate treating solution onto the substrate;
• a gantry unit, in which the inkjet head unit is installed, and for moving the inkjet head unit on the substrate;
• and
• a substrate inspecting unit for inspecting the substrate;
• wherein the substrate inspecting unit comprises,
• a feature extracting module for extracting a feature from training data included in each class in response to a plurality of training data related to image data of a substrate being classified according to a predefined class;
• a validity evaluating module for evaluating validity of the feature;
• a class verifying module for verifying the predefined class;
• and
• a data reconstructing module for reconstructing the plurality of training data based on a feature determined as valid and a verified class,
• wherein reconstructed training data is utilized when inspecting the substrate.