Paper Contents
Abstract
In the process of IC design, lithography can be defined as the process of reissuing the pattern of the mask on a Silicon wafer. Lithography is an essential step in this process as it enables point size to drop which further helps in dwindling device size. This nonstop drop in point size may lead to printability issues and hotspots. Presence of hotspots can beget the circuit to fail, so it's veritably important to descry these hotspots with high delicacy. preliminarily colorful simulation, machine leaning and deep literacy grounded ways have been enforced to break this issue. In this paper, a system to identify hotspots using Vision Mills is proposed. Other deep literacy ways, similar as CNNs and ANNs have also been used for comparison purposes. All three ways are enforced on five datasets. ViT gives an overall average delicacy of98.05 which is1.39 advanced than delicacy of CNNs and2.04 advanced than delicacy given by ANNs. Although the ViTs prove the stylish in terms of overall delicacy, but at dataset position its performance can be bettered. Three out of five datasets have delicacy advanced than 99 and for rest two it's slightly above 95. In future, we wish to ameliorate delicacy for these two datasets by perfecting the model and reducing imbalance in the datasets. Keywords: Deep literacy, Lithography, Hotspot Detection, Vision Transformer, complication Neural Network, Artificial Neural Network.
Copyright
Copyright © 2024 Ali Akbar P. This is an open access article distributed under the Creative Commons Attribution License.
