Beyond Lux thresholds: a systematic pipeline for classifying biologically relevant light contexts from wearable data

2025-12-05

AI Summary

This study addresses the challenge of classifying natural versus artificial light in wearable spectral data, where performance is highly susceptible to illumination intensity variations. To overcome reliance on heuristic illuminance thresholds, we propose a robust, illumination-invariant method: (1) L2-normalization to eliminate total spectral intensity effects; (2) hour-level medoid clustering to construct a joint spectral-temporal representation; and (3) sinusoidal/cosine time encoding coupled with logarithmic spectral transformation to enhance modeling of diurnal periodicity. A multilayer perceptron (MLP) serves as the classifier, validated via strict participant-out cross-validation. Evaluated on real-world wearable spectral recordings, our approach achieves an AUC of 0.938 (accuracy: 88%), significantly outperforming conventional illuminance-threshold-based methods. This work establishes, for the first time, a reproducible analytical framework for biologically relevant lighting environment classification that is illumination-robust, generalizable across individuals, and grounded in physiological plausibility.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Multimodal Learning

Application Category

• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Background: Wearable spectrometers enable field quantification of biologically relevant light, yet reproducible pipelines for contextual classification remain under-specified. Objective: To establish and validate a subject-wise evaluated, reproducible pipeline and actionable design rules for classifying natural vs. artificial light from wearable spectral data. Methods: We analysed ActLumus recordings from 26 participants, each monitored for at least 7 days at 10-second sampling, paired with daily exposure diaries. The pipeline fixes the sequence: domain selection, log-base-10 transform, L2 normalisation excluding total intensity (to avoid brightness shortcuts), hour-level medoid aggregation, sine/cosine hour encoding, and MLP classifier, evaluated under participant-wise cross-validation. Results: The proposed sequence consistently achieved high performance on the primary task, with representative configurations reaching AUC = 0.938 (accuracy 88%) for natural vs. artificial classification on the held-out subject split. In contrast, indoor vs. outdoor classification remained at feasibility level due to spectral overlap and class imbalance (best AUC approximately 0.75; majority-class collapse without contextual sensors). Threshold baselines were insufficient on our data, supporting the need for spectral-temporal modelling beyond illuminance cut-offs. Conclusions: We provide a reproducible, auditable baseline pipeline and design rules for contextual light classification under subject-wise generalisation. All code, configuration files, and derived artefacts will be openly archived (GitHub + Zenodo DOI) to support reuse and benchmarking.

Problem

• Develops a pipeline to classify natural vs. artificial light from wearable spectral data
• Addresses the need for reproducible methods beyond simple illuminance thresholds
• Validates the approach using participant-wise evaluation to ensure generalizability

Innovation

• Pipeline with L2 normalization excluding total intensity
• Hour-level medoid aggregation with sine/cosine encoding
• MLP classifier evaluated under participant-wise cross-validation