Deep supervised learning to estimate true rough line images from SEM images

Chaudhary, Narendra; Savari, Serap A.; Yeddulapalli, Sai Swaroop

doi:10.1117/12.2324341

Cited by 6 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, suitably optimised DnCNN method can be successfully used for denoising of scanning electron microscopy (SEM) images. 16,17 This also improves the accuracy nanometre-scale SEM measurements. Some scholars applied the previously proposed network to mapping between images, giving the network structure a new application prospect.…”

Section: Related Workmentioning

confidence: 92%

See 1 more Smart Citation

Imaging error compensation method for through‐focus scanning optical microscopy images based on deep learning

Jia-jun

Peng

et al. 2021

Journal of Microscopy

View full text Add to dashboard Cite

Through-focus scanning optical microscopy (TSOM) is a model-based nanoscale metrology technique which combines conventional bright-field microscopy and the relevant numerical simulations. A TSOM image is generated after throughfocus scanning and data processing. However, the mechanical vibration and optical noise introduced into the TSOM image during image generation can affect the measurement accuracy. To reduce this effect, this paper proposes a imaging error compensation method for the TSOM image based on deep learning with U-Net.Here, the simulated TSOM image is regarded as the ground truth, and the U-Net is trained using the experimental TSOM images by means of a supervised learning strategy. The experimental TSOM image is first encoded and then decoded with the U-shaped structure of the U-Net. The difference between the experimental and simulated TSOM images is minimised by iteratively updating the weights and bias factors of the network, to obtain the compensated TSOM image. The proposed method is applied for optimising the TSOM images for nanoscale linewidth estimation. The results demonstrate that the proposed method performs as expected and provides a significant enhancement in accuracy.

show abstract

Section: Related Workmentioning

confidence: 92%

“…The DnCNN can be modified into a new network model to meet different image requirements. For example, suitably optimised DnCNN method can be successfully used for denoising of scanning electron microscopy (SEM) images 16,17 . This also improves the accuracy nanometre‐scale SEM measurements.…”

Section: Related Workmentioning

confidence: 99%

Imaging error compensation method for through‐focus scanning optical microscopy images based on deep learning

Jia-jun

Peng

et al. 2021

Journal of Microscopy

View full text Add to dashboard Cite

show abstract

“…Since deep learning 36 has altered the practices of signal, image, and video processing, its rapid advances may offer the potential to better address these issues, particularly since line and contour measurements currently require an average of multiple lines or contours in order to sufficiently improve precision (see, e.g., [ 29,30 directly outputs a matrix of dimension 2 × 1024 with the estimated left and right edge positions of the line; we will discuss EDGENet and the simulated dataset used to study it, and we point the audience to References 4 and 24 for a sample of the extensive literature on LER and its estimation. Since CD-SEM metrology artifacts affect the accuracy of LER measurements, 38 we propose denoising as a first step in constructing prediction intervals and apply our group's Poisson denoising CNN SEMNet, 30,42 which was designed for the same dataset as EDGENet. We use various computer vision and image processing techniques in combination with the conformal prediction and conformalized quantile regression frameworks to examine how the EDGENet LER prediction errors are related to the "noise image" defined as the absolute difference between a noisy input image and its associated denoised output image from SEMNet.…”

Section: Background On Lermentioning

confidence: 99%

“…D is the output of the denoiser SEMNet 30,42 for input x i , and the corresponding residual or absolute prediction error; this approach appears to be related to meta-learning 25 for this application since there is a correlation between noise and other artifacts and the difficulty of edge detection. 38 The normalized nonconformity score associated with γ is…”

Section: On Conformal Prediction and Conformalized Quantile Regressionmentioning

confidence: 99%

On the Construction of Distribution-Free Prediction Intervals for an Image Regression Problem in Semiconductor Manufacturing

Akpabio¹,

Savari²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

The high-volume manufacturing of the next generation of semiconductor devices requires advances in measurement signal analysis. Many in the semiconductor manufacturing community have reservations about the adoption of deep learning; they instead prefer other model-based approaches for some image regression problems, and according to the 2021 IEEE International Roadmap for Devices and Systems (IRDS) report on Metrology a SEMI standardization committee may endorse this philosophy. The semiconductor manufacturing community does, however, communicate a need for state-of-the-art statistical analyses to reduce measurement uncertainty. Prediction intervals which characterize the reliability of the predictive performance of regression models can impact decisions, build trust in machine learning, and be applied to other regression models. However, we are not aware of effective and sufficiently simple distribution-free approaches that offer valid coverage for important classes of image data, so we consider the distribution-free conformal prediction and conformalized quantile regression framework. The image regression problem that is the focus of this paper pertains to line edge roughness (LER) estimation from noisy scanning electron microscopy images. LER affects semiconductor device performance and reliability as well as the yield of the manufacturing process; the 2021 IRDS emphasizes the crucial importance of LER by devoting a white paper to it in addition to mentioning or discussing it in the reports of multiple international focus teams. It is not immediately apparent how to effectively use normalized conformal prediction and quantile regression for LER estimation. The modeling techniques we apply appear to be novel for finding distribution-free prediction intervals for image data and will be presented at the 2022 SEMI Advanced Semiconductor Manufacturing Conference.

show abstract

Simultaneous Denoising and Edge Estimation from SEM Images using Deep Convolutional Neural Networks

Chaudhary

Savari

2019

2019 30th Annual SEMI Advanced Semiconductor Manufacturing Conference (ASMC)

View full text Add to dashboard Cite

Deep supervised learning to estimate true rough line images from SEM images

Cited by 6 publications

References 18 publications

Imaging error compensation method for through‐focus scanning optical microscopy images based on deep learning

Imaging error compensation method for through‐focus scanning optical microscopy images based on deep learning

On the Construction of Distribution-Free Prediction Intervals for an Image Regression Problem in Semiconductor Manufacturing

Simultaneous Denoising and Edge Estimation from SEM Images using Deep Convolutional Neural Networks

Contact Info

Product

Resources

About