Background. Deep learning (DL) stands as a highly effective big data analytics method, boasting a remarkable track record across diverse domains such as image processing, voice recognition, object detection, illness diagnosis, prediction and clinical decision support systems. Its applicability extends even further into emerging areas like food science and engineering, where it has witnessed a surge in its application. Over the past decade, DL has demonstrated its utility in various facets of the food industry, including food quality assessment, the detection, differentiation, identification of diseases, phenotyping of plant stressors, monitoring and smart farming practices. The integration of DL technologies has been pivotal in revolutionizing the food business and its associated supply chains, facilitating advancements in food quality evaluation, food recognition and spectroscopic analysis. Notably, hyperspectral imaging and acoustic data have emerged as key modalities leveraged by DL techniques in these applications.
Results and conclusions. This study aims to delve into the recent strides made in Deep Learning frameworks within the food industry, exploring their diverse applications and functionalities. Of particular interest is the exploration of DL’s role in food sensory analysis and consumer research, where it presents a promising avenue for sophisticated data mining techniques. Through a comprehensive examination of comparative performances, architectural nuances and potential future applications, this paper aims to shed light on the evolving landscape of DL in the realm of food science and engineering.