Harnessing Group-Oriented Consistency Constraints for Semi-Supervised Semantic Segmentation in CdZnTe Semiconductors

2025-08-18

AI Summary

To address error accumulation and confirmation bias in conventional semi-supervised segmentation methods applied to CdZnTe semiconductor images—where multiple unlabeled views share a single annotation (“many-to-one”) and exhibit low-contrast defect boundaries—this paper proposes ICAF, a multi-view group-aware semi-supervised semantic segmentation framework. Its core innovation is the first introduction of a group-oriented consistency enhancement mechanism: intra-group view sampling establishes group-level reference representations, while a pseudo-label correction network—comprising view augmentation and correction modules—simultaneously enhances boundary delineation and suppresses label noise. Evaluated on the CdZnTe dataset with DeepLabV3+, ICAF achieves 70.6% mIoU using only two annotated view groups (0.5% labeling ratio), demonstrating substantial improvement in few-shot segmentation performance.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

Labeling Cadmium Zinc Telluride (CdZnTe) semiconductor images is challenging due to the low-contrast defect boundaries, necessitating annotators to cross-reference multiple views. These views share a single ground truth (GT), forming a unique ``many-to-one'' relationship. This characteristic renders advanced semi-supervised semantic segmentation (SSS) methods suboptimal, as they are generally limited by a ``one-to-one'' relationship, where each image is independently associated with its GT. Such limitation may lead to error accumulation in low-contrast regions, further exacerbating confirmation bias. To address this issue, we revisit the SSS pipeline from a group-oriented perspective and propose a human-inspired solution: the Intra-group Consistency Augmentation Framework (ICAF). First, we experimentally validate the inherent consistency constraints within CdZnTe groups, establishing a group-oriented baseline using the Intra-group View Sampling (IVS). Building on this insight, we introduce the Pseudo-label Correction Network (PCN) to enhance consistency representation, which consists of two key modules. The View Augmentation Module (VAM) improves boundary details by dynamically synthesizing a boundary-aware view through the aggregation of multiple views. In the View Correction Module (VCM), this synthesized view is paired with other views for information interaction, effectively emphasizing salient regions while minimizing noise. Extensive experiments demonstrate the effectiveness of our solution for CdZnTe materials. Leveraging DeepLabV3+ with a ResNet-101 backbone as our segmentation model, we achieve a 70.6% mIoU on the CdZnTe dataset using only 2 group-annotated data (5 extperthousand). The code is available at href{https://github.com/pipixiapipi/ICAF}{https://github.com/pipixiapipi/ICAF}.

Problem

• Overcoming low-contrast defect boundaries in CdZnTe semiconductor image labeling
• Addressing many-to-one relationship limitations in semi-supervised segmentation
• Reducing error accumulation and confirmation bias in low-contrast regions

Innovation

• Intra-group Consistency Augmentation Framework (ICAF)
• Pseudo-label Correction Network (PCN)
• View Augmentation and Correction Modules