Layout hotspot detection method based on geometric feature analysis

By generating a layout sample set through undersampling and oversampling, and combining geometric feature extraction of corner point count, short-circuit sensitivity, and open-circuit sensitivity, an ensemble learning model is constructed. This solves the problems of sample imbalance and low recognition rate of unseen patterns in layout hotspot detection, and achieves high-precision and low-false-alarm hotspot detection.

CN118134889BActive Publication Date: 2026-06-26XIDIAN UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
XIDIAN UNIV
Filing Date
2024-03-21
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

In existing technologies, there is a severe imbalance between hot spot category samples and non-hot spot category samples on the map, which leads to reduced model detection accuracy and low hot spot recognition rate and high false alarm rate for map design patterns that have never been seen before.

Method used

A layout sample set is generated using undersampling and oversampling methods. Geometric features of the layout, such as the number of corner points, short-circuit sensitivity, and open-circuit sensitivity, are extracted. An ensemble learning model is then constructed to detect hotspots, avoiding reliance on the model to learn the geometry of hotspot patterns.

Benefits of technology

It improves the accuracy and recall rate of hotspot detection on the map, reduces the false alarm rate, and increases detection efficiency, making it suitable for real-time hotspot detection on the map.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118134889B_ABST
    Figure CN118134889B_ABST
Patent Text Reader

Abstract

The application discloses a layout hotspot detection method based on geometric feature analysis, and the implementation steps are as follows: an under-sampling and over-sampling method is used to generate a layout sample set; the number of corner points, short-circuit sensitivity and open-circuit sensitivity of the layout sample are extracted to form a feature vector of the layout sample; an integrated learning model is trained by using a feature vector training set; and the feature vector test set is input into the trained integrated learning model to output a detection result. The application overcomes the problems in the prior art that the imbalance of layout category samples leads to reduced model detection precision, and that the recognition rate of a layout mode that has never been seen before is low and the false positive rate is high. Geometric features are extracted for photolithography hotspots in the layout, so that the application can maintain high detection precision and low false positive rate when detecting a layout that has never been seen before and a complex layout mode.
Need to check novelty before this filing date? Find Prior Art