AI-Enabled Shrinkage Analysis and Morphology Control in Food Processing: Mechanisms, Multimodal Perception, Modeling,...

Food shrinkage and morphological distortion during processing and storage are major causes of quality deterioration, affecting appearance, texture, internal structure, and consumer acceptance. These deformations arise from complex couplings among moisture migration, heat and mass transfer, and matrix mechanical evolution, making accurate prediction and regulation highly challenging using conventional empirical approaches. Recent advances in artificial intelligence (AI) provide new opportunities for intelligent morphology analysis and control in food systems. This review systematically summarizes the current progress of AI-enabled shrinkage characterization, predictive modeling, and intelligent regulation in food processing. First, the multiphysics mechanisms governing shrinkage formation and morphological evolution are critically discussed. Subsequently, a multimodal perception framework integrating visual imaging, acoustic sensing, spectral analysis, and microstructural imaging is presented for comprehensive characterization of appearance, texture, and internal structural changes. The review further evaluates the evolution of shrinkage modeling from traditional machine learning to deep learning and physics-informed neural networks (PINNs), highlighting their capabilities and current limitations in mechanism interpretability and cross-domain generalization. In addition, representative intelligent regulation pathways, including visual feedback closed-loop control, multiobjective optimization, smart pretreatment, active packaging, and programmable physical field intervention, are systematically analyzed. Finally, future trends involving multimodal data fusion, digital twins, edge computing, and AI-driven active morphology design are discussed. This review provides a comprehensive theoretical and technological framework for the intelligent transformation of food morphology engineering and precision food manufacturing.