Prediction of the Key Rheological Characteristics of Wheat Dough Based on the Spatial Deformation Difference Before and After the Bursting of the Dough

Dough toughness and extensibility are key rheological properties directly linked to the quality of dough‐based products. Thus, this research employed a custom‐built device to inflate dough sheets into bubbles, with a 3D imaging system recording real‐time point cloud data of these formations. The investigation specifically targeted point cloud imagery capturing both the frame displaying maximum spatial deformation height in dough bubbles and its subsequent frame. Four novel computational methods were developed to quantify four distinct deformation characteristics from the selected two‐frame bubble point cloud images. These parameters were subsequently fed into three machine learning architectures for autonomous prediction of dough toughness and extensibility. Experimental findings revealed optimal detection accuracy when utilizing deformation feature n's immediate recovery index ( E n ) with the ELM algorithm, attaining R p coefficients of 0.882 and0.940 for extensibility and toughness assessments, respectively.