EN FR

Logging Core Defects Using Computer Vision: Enhancing Speed and Accuracy in Geotechnical Investigations

Declan Vanderhor

In the proceedings of: GeoManitoba 2025: 78th Canadian Geotechnical Conference & 9th Canadian Permafrost Conference

Session: Rock Mechanics & Engineering Geology

ABSTRACT: Manual logging of borehole core defects remains a critical but time-intensive part of subsurface investigation. These traditional methods, while familiar and trusted, often introduce delays, inconsistencies, and integration challenges with digital platforms. At TabLogs, we are developing an AI-powered solution that interprets both core images and legacy borehole logs to accelerate and standardize data capture. This paper explores the system architecture, early trial findings, and its potential for transforming how geotechnical data is digitized and reviewed. Initial feedback indicates that the approach has the potential to reduce log processing time by up to 60% and improve defect detection accuracy while retaining essential human oversight.


RÉSUMÉ: L'enregistrement manuel des défauts des carottes de forage reste un élément essentiel, mais chronophage, de l'investigation du sous-sol. Ces méthodes traditionnelles, bien que familières et fiables, engendrent souvent des retards, des incohérences et des difficultés d'intégration avec les plateformes numériques. Chez TabLogs, nous développons une solution basée sur l'IA qui interprète à la fois les images de carottes et les diagraphies de forage existantes afin d'accélérer et de standardiser la capture des données. Cet article explore l'architecture du système, les premiers résultats des essais et son potentiel pour transformer la numérisation et l'analyse des données géotechniques. Les premiers retours indiquent que cette approche pourrait réduire le temps de traitement des diagraphies jusqu'à 60 % et améliorer la précision de la détection des défauts, tout en conservant la surveillance humaine essentielle.


Submit a Data Update Form for this paper
Please include this code when submitting a data update via other methods: GEO2025_51

Access this article:
Canadian Geotechnical Society members can access to this article, along with all other Canadian Geotechnical Conference proceedings, in the Member Area. Conference proceedings are also available in many libraries.

Cite this article:
Vanderhor, Declan (2025) Logging Core Defects Using Computer Vision: Enhancing Speed and Accuracy in Geotechnical Investigations in GEO2025. Ottawa, Ontario: Canadian Geotechnical Society.

@inproceedings{Vanderhor_GEO2025_51, author = {{Vanderhor, Declan}}
title = {Logging Core Defects Using Computer Vision: Enhancing Speed and Accuracy in Geotechnical Investigations }
booktitle = {Proceedings of the 78th Canadian Geotechnical Conference & 9th Canadian Permafrost Conference}
year = {2025}
organization = {The Canadian Geotechnical Society},
address = {Ottawa, Canada} }
Abstracts are Copyright © the Authors and used with permission. Online database Copyright © 2026 The Canadian Geotechnical Society. All rights reserved.