Unsupervised Domain Adaptation with Histogram-gated Image Translation for Delayered IC Image Analysis

TL;DR

This paper introduces HGIT, an unsupervised domain adaptation method that uses histogram-guided GAN-based image translation to improve circuit image segmentation across different datasets, reducing the need for manual annotations.

Contribution

The paper presents a novel histogram-gated image translation framework for unsupervised domain adaptation in circuit image segmentation, outperforming existing techniques.

Findings

HGIT achieves superior segmentation accuracy compared to other domain adaptation methods.

The method performs close to fully supervised benchmarks without requiring labeled target data.

Experiments validate HGIT's effectiveness across multiple target datasets.

Abstract

Deep learning has achieved great success in the challenging circuit annotation task by employing Convolutional Neural Networks (CNN) for the segmentation of circuit structures. The deep learning approaches require a large amount of manually annotated training data to achieve a good performance, which could cause a degradation in performance if a deep learning model trained on a given dataset is applied to a different dataset. This is commonly known as the domain shift problem for circuit annotation, which stems from the possibly large variations in distribution across different image datasets. The different image datasets could be obtained from different devices or different layers within a single device. To address the domain shift problem, we propose Histogram-gated Image Translation (HGIT), an unsupervised domain adaptation framework which transforms images from a given source dataset to the domain of a target dataset, and utilize the transformed images for training a segmentation network. Specifically, our HGIT performs generative adversarial network (GAN)-based image translation and utilizes histogram statistics for data curation. Experiments were conducted on a single labeled source dataset adapted to three different target datasets (without labels for training) and the segmentation performance was evaluated for each target dataset. We have demonstrated that our method achieves the best performance compared to the reported domain adaptation techniques, and is also reasonably close to the fully supervised benchmark.