Method and apparatus for detecting pattern defects

A visual inspection method comprising:

• a classification condition setting step of setting classification conditions beforehand;
• a defect detection step of detecting defects by using inspection images acquired by imaging a target substrate, and calculating features and position information of the detected defects;
• and
• a defect classification step of classifying the defects in accordance with the classification conditions set beforehand in the classification condition setting step by using the features and position information of the defects calculated in the defect detection step;
• wherein said classification condition setting step further includes:
• a collection step of detecting a large number of defects by using inspection images acquired by imaging a target substrate, and calculating features and position information of the large number of detected defects, and collecting the calculated feature and position information of each defect over the large number of defects to store the collected feature and position information of the each defect;
• a defect sampling step of extracting clusters based on the position information over the large number of defects collected by the collection step, and for defects except the extracted clusters, creating defect feature distribution indicating defect occurrence distribution based on the feature of each defect over the large number of defects collected in the collection step, and performing sampling review defects based on the created defect feature distribution;
• and
• a reviewing step of giving at least defect classes to a plurality of review defects by reviewing for the review defects sampled in the defect sampling step;
• wherein for the defects except the extracted clusters, the defect classes of the review defects given in the reviewing step for the features of the review defects collected in the collection step, are set as training data of the classification conditions.

A visual inspection method comprising:

• a classification condition setting step of setting classification conditions beforehand;
• a defect detection step of detecting defects by using inspection images acquired by imaging a target substrate, and calculating features and position information of the detected defects;
• and
• a defect classification step of classifying the defects in accordance with the classification conditions set beforehand in the classification condition setting step by using the features and position information of the defects calculated in the defect detection step;
• wherein said classification condition setting step further includes:
• a collection step of detecting a large number of defects by using inspection images acquired by imaging a target substrate, and calculating features and position information of the large number of detected defects, and collecting the calculated feature and position information of each defect over the large number of defects to store the collected feature and position information of the each defect;
• a defect sampling step of extracting clusters based on the position information of each defect over the large number of defects collected in the collection step, and for the extracted clusters, performing sampling to check a few defects, whereas for defects except the extracted clusters, creating defect feature distribution indicating defect occurrence distribution based on the feature of each defect over the large number of defects collected in the collection step and performing sampling review defects based on the created defect feature distribution;
• and
• a reviewing step of, for the cluster defects sampled in the defect sampling step, giving a defect class that is checked by reviewing the cluster defects, whereas for the plurality of review defects sampled in the defect sampling step, giving at least a defect class determined by reviewing the review defects;
• wherein for the cluster defects, setting the defect class checked in the reviewing step as training data of the classification conditions, whereas for defects except the cluster defects, setting, as training data of the classification conditions, the defect class of the review defects given in the reviewing step for the features of the review defects collected in the collection step.

A visual inspection apparatus comprising:

• a classification condition setting means which sets classification conditions beforehand;
• a defect detection means which detects defects by using inspection images acquired by imaging a target substrate, and calculates features of the detected defects;
• and
• a defect classification unit which classifies the defects in accordance with the classification conditions set beforehand by the classification condition setting means by using the defect features calculated by the defect detection means;
• wherein said classification condition setting means further includes:
• a collection unit which detects a large number of defects by using inspection images acquired by imaging a target substrate by the defect detection means, and which calculates feature of each defect over the large number of detected defects, and which collects the calculated feature of each defect over the large number of detected defects to store the collected feature of the each defect;
• a defect sampling unit which creates defect feature distribution indicating defect occurrence distribution based on the feature of each defect over the large number of defects collected in the collection unit, and which performs sampling review defects based on the created defect feature distribution;
• an input means which inputs at least defect classes given as a result of reviewing by a review tool for a plurality of review defects sampled by the defect sampling unit;
• and
• a classification condition setting unit which sets the defect classes of the review defects inputted by the input means for the features of the review defects collected by the collection unit as training data of the classification conditions.

A visual inspection apparatus comprising:

• a classification condition setting means which sets classification conditions beforehand;
• a defect detection means which detects defects by using inspection images acquired by imaging a target substrate, and which calculates features and position information of the detected defects;
• and
• a defect classification unit which classifies the defects in accordance with the classification conditions set beforehand by the classification condition setting means by using the features and position information of the defect calculated by the defect detection means;
• wherein said classification condition setting means further includes:
• a collection unit which detects a large number of defects by using inspection images acquired by imaging a target substrate, and calculates features and position information of the large number of detected defects, and which collects the calculated features and position information of each defect over the large number of defects to store the collected feature and position information of the each defect;
• a defect sampling unit which extracts clusters based on the position information over the large number of defects collected by the collection unit, and which, for defects except the extracted clusters, creates defect feature distribution indicating defect occurrence distribution based on the feature of each defect over the large number of defects collected by the collection unit, and which performs sampling review defects based on the created defect feature distribution;
• a input means which inputs at least defect classes given as a result of reviewing by a review tool for a plurality of review defects sampled by the defect sampling unit;
• and
• a classification condition setting unit which, for the defects except the clusters, sets the defect classes of the review defects inputted by the input means for the features of the review defects collected by the collection unit as training data of the classification conditions.

A classification condition setting apparatus comprising:

• a collection unit which collects feature of each defect over a large number of defects acquired from a visual inspection apparatus, and which stores the feature of the each defect;
• a defect sampling unit which creates defect feature distribution indicating defect occurrence distribution based on the feature of each defect over the large number of defects collected by the collection unit, and which performs sampling review defects based on the created defect feature distribution;
• an input means which inputs at least defect classes given as a result of reviewing by a review tool for a plurality of review defects sampled by the defect sampling unit;
• and
• a classification condition setting unit which sets the defect classes of the review defects inputted by the input means for the features of the review defects collected by the collection unit as training data of the classification conditions.

A classification condition setting apparatus comprising:

• a collection unit which collects feature and inspection image of each defect over a large number of defects acquired from a visual inspection apparatus to store the feature and inspection image of the each defect;
• a defect sampling unit which creating defect feature distribution indicating defect occurrence distribution based on the defect feature of the each defect over the large number of defects collected by the collection unit, and which performs sampling review defects based on the created defect feature distribution;
• an input means which inputs review images acquired by reviewing by a review tool for a plurality of review defects sampled by the defect sampling unit;
• a defect class training means which displays a list of images on a monitor, said list including inspection images of a plurality of review defects collected by the collection unit and review images of the plurality of review defects inputted by the input means, and which teaches a defect classes for the displayed list of the images;
• and
• a classification condition setting unit which sets the defect classes of the review defects taught by the defect class training means for the features of the review defects collected by the collection unit.