Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study evaluates the effectiveness of vision-based tactile sensors for anomaly detection in small industrial components, addressing the limitations of standard optical cameras in capturing fine surface geometries. By systematically comparing four unsupervised anomaly detection methods—SPADE, PaDiM, FAPM, and InReaCh—on a tactile dataset obtained from a GelSight Mini sensor, the research identifies key performance metrics influenced by contact conditions and sensor resolution. The findings reveal that vision-based tactile sensing can significantly enhance industrial quality control by providing reliable anomaly detection capabilities, even in challenging scenarios with limited data.
Vision-based tactile sensing can outperform traditional optical methods in detecting anomalies in small industrial components, revealing critical insights into surface geometry that standard cameras miss.
Automated inspection of small industrial components, including sub-centimetre-scale parts where defects are geometry-driven and poorly resolved by standard optical cameras, calls for sensing modalities that can directly capture fine surface geometry. Vision-based tactile sensors address this need by converting contact imprints into high-resolution image-like data compatible with existing deep-learning pipelines, yet their effective use for industrial anomaly detection (AD) remains largely unexplored. This work systematically evaluates unsupervised AD methods on a real tactile dataset covering five genuine industrial components acquired with a GelSight Mini sensor mounted on a collaborative robot. Four feature-embedding methods, SPADE, PaDiM, FAPM, and InReaCh, are compared under three validations explicitly motivated by the deployment constraints of contact-based sensing: a Good Fraction analysis establishing the minimum number of nominal contacts for stable performance, directly bounded by gel wear since every acquisition degrades the soft interface; a cross-position evaluation assessing generalization across different contact locations observing the same recurring surface pattern; and a low- versus high-resolution comparison evaluating the cost-benefit of higher-resolution tactile acquisition. Overall, this systematic benchmarking study provides practical guidance for researchers and practitioners adopting vision-based tactile sensing for industrial AD and shows how this modality can serve as a viable alternative for industrial quality-control tasks.