2026-10-14T20:20:38Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34007799 2026-09-30T00:25:09Z category_28840 portal_12 item_type_5 month_year_09_2026

Isurie Akarawita: Machine Vision for Classifying Bolus Formation in Robotic Chewing Isurie Akarawita (17062131) Bangxiang Chen (4072387) Jaspreet Singh Dhupia (25119067) Martin Stommel (25119073) Weiliang Xu (5982089) Intelligent robotics Chewing phase classification Food bolus transformation Image texture analysis Machine learning Mouth process model Robotic chewing <p dir="ltr">This study presents a non-contact machine-vision framework for classifying progressive bolus formation during robotic mastication. A total of 1,024 bolus images were collected from two chewing trajectories: vertical compression (T1) and combined vertical–shear motion (T13). Images were segmented using SAM2, and deep VGG16 features were fused with classical texture descriptors, including GLCM features, mean lightness, and intensity variance. Four chew-count-defined stages: Unchewed, Initial, Mid, and Final were classified using LDA, SVM, and Random Forest models with trial-grouped cross-validation. LDA achieved the highest accuracy, reaching 65% for T1 and 68% for T13. The shear-assisted T13 trajectory provided improved stage separability, while most classification errors occurred between neighbouring chewing stages. These findings demonstrate the potential of machine vision for non-invasive monitoring of bolus transformation without interrupting robotic chewing or food bolus extraction.</p> 2026-09-30T00:25:09Z Image Poster 10.17608/k6.auckland.34007799.v3 https://figshare.com/articles/poster/Untitled_Item/34007799 CC BY 4.0