Acoustic scattering AI for non-invasive object classifications: A case study on hair assessment

2025-06-17

AI Summary

This study addresses non-invasive, fine-grained classification of biological tissues—specifically hair type and moisture content—using contactless acoustic scattering techniques. Method: We propose a physics-informed acoustic scattering signal acquisition framework, integrated with self-supervised pretraining and full-parameter fine-tuning, ensuring privacy-preserving operation. Our end-to-end acoustic AI pipeline encompasses signal acquisition, time-frequency feature extraction, embedded representation learning, and lightweight discriminative classification. Contribution/Results: To our knowledge, this is the first work to model acoustic scattering properties for biological tissue characterization. Experimental evaluation on real human scalp samples achieves 89.7% classification accuracy—significantly outperforming conventional vision-based approaches. The results demonstrate the feasibility, privacy compliance, and industrial deployability of acoustic scattering AI in unobtrusive, non-contact, and privacy-sensitive scenarios.

Technology Category

• Machine Learning: Privacy
• Humans and AI: Brain-Sensing and Analysis
• Computer Vision: Medical and Biological Imaging

Application Category

• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization

Abstract

This paper presents a novel non-invasive object classification approach using acoustic scattering, demonstrated through a case study on hair assessment. When an incident wave interacts with an object, it generates a scattered acoustic field encoding structural and material properties. By emitting acoustic stimuli and capturing the scattered signals from head-with-hair-sample objects, we classify hair type and moisture using AI-driven, deep-learning-based sound classification. We benchmark comprehensive methods, including (i) fully supervised deep learning, (ii) embedding-based classification, (iii) supervised foundation model fine-tuning, and (iv) self-supervised model fine-tuning. Our best strategy achieves nearly 90% classification accuracy by fine-tuning all parameters of a self-supervised model. These results highlight acoustic scattering as a privacy-preserving, non-contact alternative to visual classification, opening huge potential for applications in various industries.

Problem

• Non-invasive hair classification using acoustic scattering
• AI-driven deep learning for sound-based property analysis
• Privacy-preserving alternative to visual object assessment

Innovation

• Acoustic scattering for non-invasive classification
• AI-driven deep learning sound analysis
• Self-supervised model fine-tuning achieves 90% accuracy