Applications of machine learning coupled with computer vision, electronic nose and untargeted analysis for food quality control

Application of artificial intelligence in food quality control has been a new trend, bringing a complete change to the traditional way of food quality control, helping to shorten analysis and detection time and carrying out real time analysis due to nondestructive sample preparation, simpler operations because of using of sensors, collecting a large amount of information thanks to taking all measurement data... This article provides a preliminary view of identification, discrimination and classification of food samples based on using machine learning and deep learning models coupled with analytical data obtained from spectral measurements, using sensors instead of the nose (electronic- nose), camera and image analysis (computer vision) for food quality control purposes such as determining food freshness, authenticating origin as well as detecting adulteration of foods. The published studies show that the application of machine learning models especially in rapid analysis and sample-free analysis has great potential as an alternative to targeted analysis methods with specific analytes in samples.