Machine learning and image processing to monitor strain and tensile forces with mechanochromic sensors

Author(s)

de Castro, Lucas D. C. ; Scabini, Leonardo ; Ribas, Lucas C. ; Bruno, Odemir M. ; Oliveira Jr, Osvaldo N.

• Document type: Journal article
• Source: EXPERT SYSTEMS WITH APPLICATIONS; v. 212, p. 7-pg., 2023-02-01.

Abstract

A computer vision (CV) system is proposed for real-time prediction of strain by monitoring the color-changing feature of mechanochromic sensors. Pictures of the sensors subjected to calibration tensile tests were treated with standard image processing methods and analyzed using supervised machine learning (ML) algorithms. Visual strain sensing was demonstrated by linear regression models capable of learning a relation between the applied strain and the reflected structural color. The ElasticNet regression model provided the highest accuracy in the strain prediction task, with a remarkable performance in monitoring real-time strain variation of sensors during a tensile-relaxion cycle. Using calibration curves, the predicted strain can also be employed for estimating the tensile force applied on the mechanochromic sensors. Taken together these results point to potential intelligent systems for noninvasive in-situ visual monitoring of deformations and tensions. (AU)

• FAPESP's process: 20/02938-0 - Coloured patterns based on bioinspired photonic crystals for mechanocromic applications
• Grantee: Lucas Daniel Chiba de Castro
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• FAPESP's process: 16/18809-9 - Deep learning and complex networks applied to computer vision
• Grantee: Odemir Martinez Bruno
• Support Opportunities: Research Grants - Research Partnership for Technological Innovation - PITE
• FAPESP's process: 18/22214-6 - Towards a convergence of technologies: from sensing and biosensing to information visualization and machine learning for data analysis in clinical diagnosis
• Grantee: Osvaldo Novais de Oliveira Junior
• Support Opportunities: Research Projects - Thematic Grants
• FAPESP's process: 14/08026-1 - Artificial vision and pattern recognition applied to vegetal plasticity
• Grantee: Odemir Martinez Bruno
• Support Opportunities: Regular Research Grants
• FAPESP's process: 21/07289-2 - Learning Representations using artificial neural networks and complex networks with applications in sensors and biosensors
• Grantee: Lucas Correia Ribas
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• FAPESP's process: 19/07811-0 - Artificial neural networks and complex networks: an integrative study of topological properties and pattern recognition
• Grantee: Leonardo Felipe dos Santos Scabini
• Support Opportunities: Scholarships in Brazil - Doctorate