MACHINE LEARNING ALGORITHMS FOR SPECTRAL IMAGING IN FOOD ANALYSIS: A REVIEW

Abstract

Over the past ten years, spectral imaging technology has rapidly progressed. To evaluate the quality of food, spectral imaging techniques are used. While carrying out of monitoring food quality, non-destructive testing techniques have grown in importance over time. Among the most important non-destructive methods for assessing quality that gives both spectrum and spatial information is hyperspectral imaging. A collection of machine-learning approaches can be applied to the hyperspectral images to analyze food. Various machine learning methods, including PCA, Random Forest (RF), Convolutional Neural Network (CNN), and Support Vector Machines (SVM), are discussed in this paper for food analysis in the context of hyperspectral images.